Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
listlengths 1
91
| keywords
listlengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
The development of modern and affordable sequencing technologies has allowed the study of viral populations to an unprecedented depth . This is of particular interest for the study of within-host RNA viral populations , where variation due to error-prone polymerases can lead to immune escape , antiviral resistance and adaptation to new host species . Methods to sequence RNA virus genomes include reverse transcription ( RT ) and polymerase chain reaction ( PCR ) . RT-PCR is a molecular biology technique widely used to amplify DNA from an RNA template . The method itself relies on the in vitro synthesis of copy DNA from RNA followed by multiple cycles of DNA amplification . However , this method introduces artefactual errors that can act as confounding factors when the sequence data are analysed . Although there are a growing number of published studies exploring the intra- and inter-host evolutionary dynamics of RNA viruses , the complexity of the methods used to generate sequences makes it difficult to produce probabilistic statements about the likely sources of observed sequence variants . This complexity is further compounded as both the depth of sequencing and the length of the genome segment of interest increase . Here we develop a Bayesian method to characterise and differentiate between likely structures for the background viral population . This approach can then be used to identify nucleotide sites that show evidence of change in the within-host viral population structure , either over time or relative to a reference sequence ( e . g . an inoculum or another source of infection ) , or both , without having to build complex evolutionary models . Identification of these sites can help to inform the design of more focussed experiments using molecular biology tools , such as site-directed mutagenesis , to assess the function of specific amino acids . We illustrate the method by applying to datasets from experimental transmission of equine influenza , and a pre-clinical vaccine trial for HIV-1 .
Reverse transcription-polymerase chain reaction ( RT-PCR ) is a common tool to generate copy DNA ( cDNA ) from RNA . All publicly available sequences of RNA viruses have been generated using this technique . The method consists of two steps: the first is an in vitro synthesis of cDNA from an RNA template in a reverse-transcription reaction ( RT ) ; and the second ( PCR ) consists of multiple cycles of DNA amplification using the cDNA generated in the RT step as a template . As in any other polymerisation reaction , misincorporations that result in artefactual mutations are generated during both steps , although at different rates ( reverse-transcriptases lack proofreading activity and thus the RT step is more error-prone , while DNA polymerases exhibit various degrees of proofreading activity ) . The current genomics revolution has generated thousands of sequences of complete RNA viral genomes . Sequences derived from the influenza viruses resource ( http://www . ncbi . nlm . nih . gov/genomes/FLU/FLU . html ) alone account for more than 175 , 000 as of October 2010 . Indeed , the advent of novel and more affordable sequencing technologies allows the study of viral populations in an unprecedented depth , up to the level of characterising within-host viral populations in a qualitative and quantitative fashion . In particular , such studies are critical to understand the mechanisms that govern the evolution of virulence or antiviral resistance , as well as the underpinning mechanisms of cross-species jumps and immune evasion . In addition , in-depth studies of genetic variation are increasingly used to elucidate the viral population dynamics and evolution ( phylodynamics ) both within and between hosts [1] . Different laboratories have explored the within-host variation and evolution of a variety of RNA viruses , ranging from those that cause acute infections such as influenza and dengue [2]–[6] , to those that persistently infect their host , like human and simian immunodeficiency viruses [7]–[10] . Despite differences in experimental design due to inherent biological features of the virus under study ( i . e . specific host , inoculation route , replication strategy ) and the biological questions being addressed ( i . e . size of transmission bottlenecks , time of appearance of antiviral resistance or immune escape variants ) , most of these experiments rely on the analysis of sequences derived from viral samples taken at different times post-infection . The underlying assumption is that if multiple samples are taken from a single host over time , it is possible to map the frequency of a particular observed sequence and its variants in a temporal fashion . However , since there are various sources of error , both in the viral replication cycle and in the experimental process , it is difficult to elicit ( probabilistically ) whether observed variants are consistent with the possibility of viral evolution , or simply a result of random misincorporations occurring either within the host or during the RT-PCR/sequencing process . We propose a Bayesian method to try to make such distinctions , and to illustrate these techniques we use data from an experimental transmission study of equine influenza virus ( EIV ) in its natural host [6] , and data from a prime-boost pre-clinical vaccine trial in a non-human primate model for HIV-1 ( M . Varela and J . L . Heeney , in preparation ) . An important biological distinction between these two pathogens is the duration of the infection; while influenza infections are typically acute , lasting for only a few days , HIV infections can last for a lifetime . In addition , the experimental procedures established for the study of within-host evolution for those two infections are different ( Figure 1 ) . For HIV , single genome amplification ( SGA ) followed by direct sequencing is currently the technique of choice [7] , [8] , [11]–[15] . In SGA , viral RNA is extracted from a clinical sample ( typically a blood sample ) and copied into cDNA , which in turn is subjected to a limiting dilution step such that only one molecule is then used as a template for a PCR reaction . Thus the obtained PCR products are the result of the amplification of one single molecule of cDNA . These PCR products are then sequenced directly without cloning . An alternative experimental approach is clonal sequencing , which has been used to study intra-host viral populations of influenza and dengue [2] , [3] , [5] , [6] . With this method , RT-PCR is performed from a clinical sample , followed by subcloning of the resulting PCR products into sequencing vectors , which in turn are introduced into bacteria in order to produce the necessary quantities of DNA required for sequencing . In clonal sequencing , DNA from individual colonies ( i . e . single molecules of PCR product ) are extracted and sequenced . The statistical framework we present here is quite general , and we show how it can be used for screening data from longitudinal within-host experiments , and/or between-host transmission studies . The mechanism by which we identify “sites-of-interest” is to monitor the frequency of bases present at a particular nucleotide site in the background population of viruses . It should be noted that the approach we propose here is not meant to replace methods to study selection analysis , for which there are already many excellent algorithms and software packages available ( e . g . [16] ) . Instead the method is designed to flag up single sites that exhibit changes in the structure of the distributions of bases either over time , or relative to a reference sequence ( such as that obtained from an inoculum sample ) . Furthermore it aims to provide a weight-of-evidence in favour of population structures that suggest higher frequencies of mutations than would be expected if all mutations arose randomly without further propagation ( i . e . de novo ) . There are various biological mechanisms that could cause these observed changes , for example competition or selection within the host , and we discuss various options in more detail in the Materials and Methods and Discussion sections . The method can also be used to inform subsequent experiments that aim to target the role of individual nucleotide variants in defined phenotypes . In both studies described here , viral sequences have been generated using capillary sequencing technologies ( i . e . Sanger sequencing ) . Although newer sequencing technologies that produce thousands of reads are available , they are not yet established for the kind of studies analysed here . This is due to the variable length of reads they produce ( 50 to 250 base pairs ) , which makes it difficult to link distant mutations , as well as for the intrinsic error rates they display .
The genetic units of interest here are individual nucleotide sites , and the output from the sequencing process is a distribution of bases present across a finite set of observed sequences . For consistency we define an observed ‘mutation’ to be a deviation away from the consensus base at a particular nucleotide site [14] . At a given nucleotide site the consensus is defined as the base present at the highest frequency in the set of observed sequences from the inoculum ( for the HIV study ) or the initial challenge animal ( for the EIV study ) . In the event that there is no clear consensus base at a particular site ( e . g . a 50∶50 split ) , then numerically the methods described subsequently are invariant to the choice of ‘consensus’ and ‘mutation’ , though care must be taken with the biological interpretation of the results . In the first instance we will consider an individual dataset containing S sequences of N nucleotides each , derived from a single clinical sample ( in this case a blood sample or a nasal swab ) . At any single nucleotide site there are three possible deviations away from the consensus base . The distributions of observed bases at a single nucleotide site can then be considered as a random draw from the background population , and can be described by a multinomial distribution ( described below ) . More formally , if we denote the number of bases of type B at site j as , then the probability of observing sequences with base , with base , with base and with the consensus base at position j is:where and . Here the parameters correspond to the proportion of each base present in the background population . For brevity we drop the complex subscript , such that and ; making only the concession that the consensus base is always indexed 4 . The goal of this work is to develop a screening mechanism to inform the development of future studies . The proposed method aims to identify nucleotide sites whose frequency of mutations differ from their expected values , which in turn are based on a given viral population and some simple assumptions about the mechanisms of random mutation events . We aim to approach this problem by using two main sources of information: the overall proportion of mutations present in the observed sequences ( denoted ) , and multiple viral samples obtained over time ( and/or within different animals ) . Given a starting population of viruses , consider initially the case that all observed mutations occur randomly without further replication . In this scenario the distribution of observed bases at a nucleotide site j will be expected to follow a multinomial distribution such that , regardless of the background structure of the s . On the other hand , if a site j exhibits a frequency of mutations such that , then it is much more likely that some form of amplification of one or more mutations has occurred , and these are defined as our “sites-of-interest” . Of course in reality will contain both “unamplified” and “amplified” mutations , as it averages over all positions . Hence using the constraint to characterise sites-of-interest will be conservative , in the sense that we are less likely to identify some truly amplified mutations due to the potential overestimate of . However , we are not modelling the biological mechanisms that cause the population structure to change , and therefore it is necessary to consider the interpretation of sites identified using this criterion . We note that any mutation must have occurred either by a biological mechanism ( “real” ) , or as an artefact of the RT-PCR process ( “artefactual” ) , and the aim of this work is to distinguish between these mechanisms in a viable manner . As in all practical discrimination algorithms there is the potential for classification error to happen , and in this case a false positive occurs when an artefact mutation is classified as a mutation-of-interest , and a false negative occurs when a real mutation is missed . In fact the distinction is more subtle than this , since real mutations that are either neutral or deleterious to the fitness of the virus are not usually of interest from a biological perspective , and if these occur then they are likely to be present at very low levels at any given time point and so will not be isolated via our screening criterion . Of course we also run the risk of missing real mutations that do confer a fitness advantage but have only just begun to replicate ( i . e . they are present at low levels in the population ) . Our method cannot make the distinction between these “real” low frequency mutations and low frequency mutations occurring as a result of RT-PCR error ( without a more complex mutation model ) . Instead we argue below that we if we can isolate high frequency mutations in a careful way , then these are more likely to constitute evidence of providing an increased fitness advantage to the virus , and hence are of particular biological importance . Of course , it is possible that single-site mutations that do show evidence of replication could have arisen during the RT-PCR process . Although this is theoretically possible , we expect that this happens at such a negligible level that it is highly unlikely that mutations isolated during our screening mechanism would have arisen in this way . For example , in clonal sequencing we amplify a large population of viruses , and expect that the amplified population will show a similar structure to the original population . If anything we might expect to miss variants that are present at low levels , since there is some concern that clonal sequencing might bias towards picking up those variants present at high levels in the population [14] , and hence we would be less likely to isolate mutations of this type using our screening criterion if this were true . In SGA the original populations are diluted down after reverse transcription in an attempt to amplify single viral molecules . In this case only mutations occurring in the RT step would count as artefacts . If an isolated mutation occurs in the early steps of the PCR and becomes amplified in the following cycles , such that it theoretically makes up a large enough proportion of the amplified population to be detected , then these sequences are removed from the analysis after visual inspection of the chromatograms . Thus errors at the PCR step are minimised . Furthermore , if we sequence multiple clinical samples then the RT-PCR processes that generate the data will be independent for each of these samples . Therefore if we saw the same mutation occurring in multiple clinical samples it is even more unlikely that this has occurred as an artefact of the RT-PCR . In either case we acknowledge the possibility that an isolated mutation could be a false positive , but consider the probability to be negligible . We reiterate that the methods described here aim to screen the data for sites-of-interest , and there may well be a small degree of false positive mutations that creep in; however , an important point is that this false positive rate will be further mediated if we observe the same mutation in multiple clinical samples , either from the same or different hosts . There is an additional subtlety however , and that is that the background population of viruses in the inoculum may not be homogeneous , and thus the variation in bases in a set of observed sequences may simply be a result of sampling from this heterogeneous background population . Therefore it is also of interest to compare the distributions of bases at a particular site to the distribution in the inoculum , or other earlier viral sample ( e . g . animal source of infection in the EIV study ) . To this end we highlight the necessity to model both frequencies and distributions of mutations . If we were interested purely in the former , then we could produce the corresponding marginal binomial distribution modelling the number of mutations observed in a set of sequences . However , if viral evolution is or has occurred , it is possible that two viral populations will carry the same frequency of mutations , but of different types . Therefore we argue here that using a method based on the full multinomial model allows comparison of the distributions and frequency of observed mutations , rather than simply the latter . To summarise , we have argued so far that we need to: These criteria then define a set of “sites-of-interest” that have a reasonable biological basis for exploration in future studies . The question then arises as to how to derive a sensible method to elicit these sites . In a classical statistical framework we would generate a null hypothesis in each case and then ask the question: under this null hypothesis how likely are we to see an observation at least as or more extreme than the observed value ? However , it is also only possible to build evidence against a single null hypothesis , and yet there are various random substitution models that may be appropriate [17]–[19] , that would ascribe different structures to the background population of bases . For example , under the Jukes-Cantor substitution model [17] the frequencies of the four nucleotides at equilibrium would be 25% . In reality , a given nucleotide is much more likely to be miscopied as a transition than a transversion [19] , and although this could be incorporated by setting different values for the proportions in our null model , these would have to be known beforehand or estimated from the data . Here we wish to compare between multiple competing models , and in addition we also want to compare between multiple distributions . The Bayesian method we propose presents a flexible alternative to both of these problems . Also , often we do not know the specific site of interest in advance , and in a classical framework it would also be necessary to account for the number of nucleotide sites being studied . One way to do this would be to use a multiple correction procedure , such as the Bonferroni or Holm-Bonferroni corrections ( that correct for the familywise error rate; see e . g . [20] ) , or the Benjamini-Hochberg correction ( that controls for the false discovery rate; [21] ) . The choice of correction procedure depends on the context of the problem posed; the former are more stringent in protecting against false positives , whereas the latter allows a proportion of false positives to be obtained in order to increase the probability of detecting all true positives . In all cases the degree-of-correction depends on the number of independent tests ( e . g . sites ) evaluated . The approach we propose here uses Bayesian models based on Bayes' Factors ( BFs; [22]–[24] ) . In contrast to the classical statistical framework where the parameters of the system are assumed fixed , in a Bayesian framework all parameters are considered to be random variables with each following a probability distribution . As such it is possible to analyse competing models in an analogous way to that of a classical hypothesis test , but with various advantages , namely: Other , more general advantages of BFs are described in Kass and Raftery [23] , and an excellent introduction to the use of BFs in general , but specifically in genetic association studies can be found in Stephens and Balding [24] . Formally , the BF is defined as the posterior odds in favour of one model against another , when the prior probability of either model is equally favourable , and is defined as:where and are competing models , and D is the observed data . We can view the competing models as competing hypotheses . The Bayesian framework can be used to generate the PPA for a given model , and this can be generalised to multiple competing models . Let denote the competing models , and let be the prior probability that model is correct , such that . Then by Bayes' Theorem:wherewith the ( unknown ) parameters on parameter space . This approach therefore integrates , or averages ( rather than maximises ) over the parameter space . If we are looking at multiple nucleotide sites , and is equal across all sites , then represents the prior proportion of sites that exhibit the phenomena of interest believed to exist in the population . This is similar to classical multiple testing procedures that account for the false discovery rate , but has the advantage that it does not depend on the number of tests performed , only the proportion of true associations believed to exist in the population [24] . To attempt to identify sites-of-interest , we will specify a set of competing models that cover a range of feasible background population structures . Therefore the set of observed sequences corresponds to a random draw from one of these population structures . In many cases we have to resort to numerical methods to calculate the likelihood , , but for the models discussed here it is possible to derive these analytically ( for mathematical details see Protocol S1 ) . For brevity the subsequent discussion assumes that we are dealing with a single nucleotide site , and we drop the site subscript . The observed data at a site are denoted , where S is the number of observed sequences . For a set of sequences obtained from a single dataset ( i . e . an individual clinical sample ) we can define ten competing structures for the background population of bases at a given site . The first five models cover a range of population structures subject to the overall mutation rate being equal to , where is the per-nucleotide mutation probability , i . e . the probability that a nucleotide in a randomly chosen sequence at a randomly chosen site differs from the consensus . We estimate by computing the overall proportion of mutations present in the data . Furthermore , we can also specify an analogous range of models in which the overall mutation rate p is allowed to vary between 0 and 1 . The derivation of for each of these models is discussed in Protocol S1 and mathematical forms given in Table S1 , along with R [25] functions to evaluate these probabilities . If multiple viral samples are available ( i . e . clinical samples obtained at different times post-infection ) , , then it is necessary to introduce some additional notation to capture the fact that different samples could have arisen as a result of sampling from different background populations . For example , consider that data from two viral samples from the same animal are available , denoted and . There are two possible scenarios: either D1 and D2 are random samples from the same population , or they are random samples from different populations . We make the assumption that at any time the population of bases at a given nucleotide site will be consistent with one of the models , and we denote the combination of models that could explain the data by using multiple subscripts corresponding to the viral sample i . e . , where corresponds to the population structure for the first viral sample ( D1 ) and to the structure for the second viral sample ( D2 ) . Thus it is necessary to calculate for any i and j . If , then by definition the background populations from which D1 and D2 are sampled are different , and so —see Protocol S1 . When then there are no free parameters over which to integrate , and so . If then there is an additional subtlety , in that D1 and D2 are either sampled from the same population , or from two different populations but with the same structure . To try to clarify this point , consider Figure 2 . This shows the case when . In Figure 2A we see that D1 and D2 are random samples from the same population described by the model . In Figure 2B we can see that D1 and D2 are random samples from two different populations , but with the same population structure , described by and respectively . To differentiate between these possibilities we introduce an additional character subscript such that cases similar to Figure 2A are denoted and cases similar to Figure 2B as . The main difference in the calculations of and relate to the parameter space over which it is necessary to integrate . These results follow from the fact that although the background populations are not independent , the sampling mechanism is . Mathematical detail is given in Protocol S1 . Summarising for the two-sample case , we have given by:where . Hence there are 109 possible competing models that could explain the joint distribution of D1 and D2 , since there are possible ways of producing random samples from two different background populations based on structures , and a further 9 ways corresponding to when D1 and D2 are samples from the same background population ( based on ) . It is possible to generalise these calculations to more than two viral samples as required , and hence it is possible to produce a PPA for all possible combinations of potential background structures . Given sequence data from multiple viral samples , we have described a method that produces weights-of-evidence in favour of the data being drawn from a particular configuration of background populations . In effect these population structures can be used to define various criteria-of-interest , which can then be assigned an overall PPA by summing across the relevant models . This can then be used to provide useful information about potential changes in the background population of viruses ( if any ) , and whether or not the frequency of mutations is higher than we would expect if there had been no propagation of mutations ( so all mutations are first generation ) . In this case we can combine the two types of sites we are seeking into a single question that can be tested using this framework: To calculate this we can append the inoculum to the observed data and treat it as an additional sample . We can then sum over the corresponding model structures that are consistent with the question of interest . In the two sample case , we have data , where is the data for the inoculum ( or initial challenge animal ) and we denote the PPA for this definition of site-of-interest as , which can be calculated as:whereIt is worth noting here that a range of questions could be asked of the data , for example we may be more stringent and ask for the probability that all viral samples obtained from an animal show a different background structure to the inoculum and exhibit a higher frequency of mutations than expected if no propagation has occurred . In which case , At the current time we use a brute-force computational approach to calculate the PPAs for all models , however it would be possible to develop an approximation based on a variation of the Occam's Window approach of [26] in order to make the calculations less computationally intensive for particularly large-scale problems .
The data in [6] consist of 2366 sequences of length 903 nucleotides , derived from 30 samples taken from 11 horses over a 15-day experiment . The number of sequences derived per sample ranged from between 44 and 154 sequences , and the breakdown by individual sample is shown in Table 1 . The aim of this study was to determine the levels of genetic variation along the course of infection in single horses and how transmission can impact on the process of within-host evolution . As such we can screen the sequence data to identify nucleotide sites that have specific properties of interest as defined earlier . The methods described here could help to locate sites and mutations that confer some fitness advantage , or perhaps neutral selection through drift founder effects which can also give some insight into the viral population dynamics . To calculate the required PPAs it is necessary to specify a prior probability of association for each of the competing models . We suggest using a range of priors to assess the strength of any observed associations . Here we choose values of 0 . 001 , 0 . 01 and 0 . 05 in favour of the phenomena of interest ( as defined in the Materials and Methods ) , split uniformly across all model structures consistent with being sites-of-interest . The remainder is split uniformly across all model structures inconsistent with our definition . Table 3 provides the PPAs that at least one sample exhibits differences in the background population structure compared to the initial challenge horse , that also shows a higher frequency of mutations than we would expect if no replication occurs ( ) . It can be seen that even with a very low prior probabilities ( 0 . 001; so less than one site a priori ) there are three sites that show very strong evidence of an association ( >0 . 97 ) , and three that show some evidence ( >0 . 07 ) . What this method allows us to do is to observe how the distributions are changing between samples , which provides useful information on which sites/mutations may be changing over the course of the experiment . It is therefore possible to explore specific sites in particular horses in more detail . As an illustrative example consider site 478 in horse 5447 . This had 4 samples taken on days 8–11 ( after initiation of the transmission chain ) , generating 69 , 44 , 50 and 51 sequences respectively ( Table 1 ) . It can be seen that the only time point at which mutations are observed is on day 11 ( Table 4 ) . To model this it is necessary to treat the samples from the challenge horse as an additional sample ( also shown in Table 1 ) , which results in 210 979 possible sets of models that could explain the output from the five viral samples . Table 5 shows the first five models returned after sorting the output by decreasing PPAs . Notice that the PPAs in favour of each of these models is equal , and this is because in the situation where there are no observed mutants in a set of sequences , then there is not enough information in the data set to distinguish between ( see Table S1 ) . What is driving this pattern is the fact that on day 11 the technique is selecting to be the most likely background population to have given rise to the data . Hence model returns the same PPA as model , or any other model that uses structures for the samples from the challenge horse and days 8–10 . It is the sum over all models that have structure 7 on day 11 that is potentially of more interest here , which results in a marginal probability of 0 . 54 . We can repeat this for all possible models for day 11 , with having a marginal probability of 0 . 37 and of 0 . 05 . We have already discussed the likely biological mechanisms behind these mutations in the Materials and Methods , and these results provide strong evidence that the mutations observed in position 478 are likely to indicate real variation and replication within this host between days 10–11 , when mutations of both type G478A and G478T occur . Interestingly , both G478A and G478T constitute non-synonymous mutations in a putative antigenic site . The data in the HIV-1 study ( M . Varela and J . L . Heeney unpublished ) consist of 439 envelope sequences of length 2544 nucleotides , derived from 10 individuals ( plus the inoculum ) at two time points over a four week period ( Table 2 ) . The purpose of the study was to identify specific changes in the HIV-1 envelope glycoprotein within a host under selective immune pressure elicited by a prime-boost vaccine . The same questions as for the EIV study can be asked , though in this case the background population of viruses in the inoculum is much more heterogeneous ( data not shown ) than that from the initial challenge horse in the EIV study . Table 6 shows the results from those sites with ( with a 0 . 001 prior ) . Interestingly , and importantly , some of these sites are identified in more than one animal , even though this was not a transmission study . It is also possible to split the results by vaccination status . Qualitatively at least , the results in Table 6 suggest that more sites show deviations from the inoculum in the vaccinated group than the unvaccinated group , which is suggestive perhaps of increased diversification due to selection pressure in response to the vaccine . It is possible as before to delve further into the nature of the mutations observed , and how the distributions have changed . For example , consider site 994; this site is identified in three animals ( R01093 , BB204 and R99004 ) , two vaccinated and one unvaccinated ( though the PPA is weaker for the unvaccinated animal ) . A summary of bases for each animal at each time point are shown in Table 7 . It is clear to see that there has been a change of consensus over the time course of infection in each of these animals , switching from G in the inoculum sample to A in each of the subsequent samples . This is backed up further by the PPAs for different model structures ( Table 8 ) , in which model was selected as the most likely model structure in two cases ( PPAs of 0 . 78 and 0 . 73 respectively ) and as the second most likely model structure in the third ( PPA = 0 . 12 ) . In the latter case the most likely model was ( PPA = 0 . 82 ) , which suggests that structure was the most likely for each sample , but that given the sample size it was not able to fully disregard the possibility of random sampling from the inoculum ( note that in the HIV-1 study there were less sequences produced per sample , and hence an increase in variability in the accuracy of the estimated distributions – nevertheless more sites-of-interest were identified overall ) . Of key importance is the fact that this site was picked up in multiple animals , and so for reasons discussed previously these differences are highly unlikely to have arisen as a result of independent RT-PCR error . In the EIV study we did not observe any sites that showed mutations occurring at a high frequency in more than one sample , however in the HIV-1 study there are various occurrences of this nature ( such as at site 994 ) . It is possible to screen specifically for these mutations specifically by placing more stringent criteria on the data; namely that we wish to identify mutations in which the data show evidence of deviating from the inoculum in both samples , as well as showing a high frequency of mutations from the consensus . These are shown in Table 9 . Although the absolute probabilities are different ( due to the resulting change in prior caused by the change in the number of models-of-interest ) , the sites observed in Table 9 are all a subset of those identified in Table 6 ( with the exception of site 261 in animal 8758 , which has a low PPA in any case ) . This illustrates a practical way in which these methods can be adapted to deal with specific questions .
Obtaining viral genetic information at multiple times post-infection either along the course of infection or along a chain of transmission , whether experimental or observational , can help us to understand the underpinning mechanisms that shape viral evolution . Nonetheless , it is difficult to obtain probabilistic information about whether these observed mutations are consistent or inconsistent with having occurred due to random mutation error . This information can provide insight into the potential fitness of single-site mutations , both in terms of survival within a host and transmission between hosts . To this end we have discussed various ways in which probabilistic measures can be derived in order to address specific questions regarding the pattern of observed single-site mutations in the data , and have applied these measures to two datasets derived from experimental studies on HIV and influenza . It should be noted that the approach we propose here is not meant to replace methods to study selection analysis but to complement them . Indeed , in [6] ( for the EIV study ) we estimated the mean numbers of non-synonymous substitutions per site and synonymous substitutions per site using the SLAC algorithm available in Datamonkey [16] . Interestingly , the mutations that we have picked in this manuscript as sites-of-interest were not identified by the aforementioned selection analysis . As a result of these more detailed analyses , we are now more confident than before in the findings of [6] , that 4 of the 11 mutations present in individual horses on multiple days were real ( sites 49 , 61 , 231 and 884; the other 7 identified in [6] were present in multiple samples but at low frequencies ) . In addition we picked up a further two mutations using our screening criteria , at sites 134 and 478 . The latter was picked up at one time point in multiple horses in [6] , and the former occurred at one time point in one horse , and so wasn't explicitly reported in [6] . However , it occurred with a high enough frequency of mutations to be detected here . The intention of this work is twofold: first , to screen large data sets for mutations of interest , and second , to focus in more detail on highlighted mutations to elicit information about the change in background population structure across multiple samples . Whilst it is possible to generate classical hypothesis tests to tackle certain questions , we provide a method based on Bayesian model selection , for various reasons . The first is that it is possible to generate evidence in favour of a particular hypothesis , rather than simply weights of evidence against the null hypothesis . Also , it is possible to compare multiple competing hypotheses in a straightforward manner . The Bayesian framework also allows the inclusion of prior information regarding the probability of specific individual nucleotide sites to be linked to the occurrence of non-deleterious or advantageous mutations . When these prior probabilities take the same values for all sites , then they represent the prior proportion of sites thought to be associated in some way , which is similar in principal to the false discovery rate used in classical multiple correction procedures but is invariant to the number of sites being examined . This makes it particularly suitable for analysing long sequences ( i . e . those ones generated by capillary sequencing ) . In many situations this prior information may not be available , and so it is necessary to conduct some form of sensitivity analysis to examine the strength of the posterior association for a range of prior values . This step helps to shed additional light on the robustness of the conclusions in the absence of detailed prior information . Moreover , in this Bayesian approach we integrate over the range of the unknown parameters , which means that the structure of the background population has to be specified , but the proportions do not have to be directly estimated ( as would be necessary in a maximum likelihood framework ) . This allows for alternative hypotheses to be generated that assume that multiple samples can come from either the same , or different background populations or population structures . The Bayesian method produces a posterior probability that a particular hypothesis is true , and can be extended to deal with sequences derived from multiple samples . This means that once a suitable range of competing model structures has been developed , different probabilistic questions can be asked of the data . For example , when analysing the EIV data we originally screened for sites that showed evidence of the phenomena-of-interest in at least one of the samples obtained from one animal . In contrast , in the case of the HIV-1 data , it was possible to apply more stringent criteria , which screened for sites that showed evidence of the phenomena-of-interest in all the samples . An important point is that the question asked will depend highly on the biological context of the problem , but the methodology is flexible enough to allow many probabilistic questions to be posed . It is worth adding at this point that the same framework could be used to screen for other types of change . For example , in the HIV-1 study the population of viruses in the inoculum was highly heterogeneous , and it would be perfectly possible to screen for initially heterogeneous sites that revert to a homogeneous population over time . The only difference would be a change in the definition of “sites-of-interest” . In addition it is worth noting that although the data analysed here have been obtained through experimental studies , this is not necessary for the methodology to be applied , though it may alter the interpretation of the results . It would be perfectly possible to apply the same techniques to observational data as might be obtained in a real-life disease outbreak . What this method does not model explicitly are the underlying mechanisms behind observed systematic mutations . If the amplification and sequencing steps are faultless and therefore introduce no errors , then the identified mutations must exist or occur as part of the replication process in the background viral population . The techniques described here cannot make the distinction between low frequency mutations that may have occurred through viral replication or artefactual error , however they can help to distinguish between likely deleterious mutations or non-advantageous mutations and those that show signs of persistence . It also allows us to compare the distributions of bases at a given site with other populations , such as the inoculum . Furthermore , mutations that occur in more than one animal can happen either de novo within each animal or due to transmission , and one area of future development would be in extending these methods to include information regarding mutations observed in multiple animals explicitly into the PPA calculations . As previously mentioned , different techniques ( i . e . clonal sequencing and SGA ) are commonly used for the study of HIV and influenza within-host evolution . Although it is beyond the scope of this study to argue the relative merits of the two techniques , it has been argued that SGA provides a more realistic representation of the viral populations under study as it avoids the generation of recombinant sequences due to template-switching and facilitates the detection of Taq polymerase errors [14] . However , this seems to be more important for studies of HIV than influenza , and since it is time consuming and expensive other methodologies are normally used to study intra-host evolution of other viruses . Nonetheless , as highlighted in the introduction , the experimental process to generate viral sequences is not fully efficient and so there is a non-zero probability of introducing artefactual errors . Figure 1 provides a simple schematic diagram comparing SGA to clonal sequencing , and highlights areas where errors could be introduced . Recently there have been some methodological developments in estimating true mutation rates that account for bias and selection [27] , and it would be possible to change the value of the overall mutation probability to accommodate this . It is worth noting that in terms of screening for true changes in the distribution of bases at particular sites as defined here , the values of obtained for the within-sample problem will be conservative ( i . e . will have a higher false negative rate ) , since the observed per-nucleotide mutation probability will include both artefact and real mutations . It is also possible to conduct various control experiments to quantify the amount of error that occurs during various steps of the process . The experimental procedure used to generate the sequences in [6] included four sequential steps of DNA synthesis ( generation of cDNA , PCR , DNA replication in bacteria and capillary sequencing ) . The main issue is determining the level of artefact mutations introduced during the reverse transcription , as this is likely to be the principal source of such errors . An issue is that these errors cannot be easily directly estimated experimentally as this will require the synthesis of a template RNA population made of identical RNA molecules , and there is no in vitro transcription system available that uses enzymes with proofreading activity . Moreover , the level of RT errors may vary with different template sequences , intracellular environment , and species origin of the RT enzyme . As a result it is difficult to draw firm conclusions as to the levels and sources of non-systematic error within sequences derived from a single sample without being able to directly quantify this error . Hence mutants that appear multiple times may either arise due to mutation events de novo in each sample , result from transmission from another animal , or be due to systematic errors in the RT-PCR steps ( e . g . if particular sites/mutations are amplified in a highly non-uniform manner ) . However , as we discuss in detail in the Materials and Methods , there are various reasons that we do not think that we are likely to pick up changes in the distributions that are purely artefacts of RT-PCR errors using the screening criteria we introduce here . The probability of a result being a false positive is further diminished if a more stringent criteria is used ( requiring evidence across multiple samples ) , or if similar changes are observed in multiple animals . There is also the issue of sampling bias , however there is no reason to assume that systematic bias should creep into either the swab sampling or in the RNA extraction . Since , by producing a set of sequences , we are effectively taking a small sample from a large population , then the effect of sampling bias ( if any ) is most likely to be that rare mutations will constitute a very low probability of detection and a high probability of being missed during sampling . Therefore if we do identify sites-of-interest using the criteria defined here , then it is even more likely that these mutations would have to be present in reasonably high levels in the background population to be detected in this manner . This is reflected also in the increase in variability observed when smaller numbers of sequences are analysed . Flexible probabilistic methods such as proposed here can help to elicit patterns from these complex and large-scale data sets based on asking intuitive questions about the data . We have described a method that allows improved inference from studies of viral transmission and evolution , in particular regarding the probabilities of observing particular mutations in viral sequence data . These types of study are becoming more common with the advent of deep and affordable sequencing technologies . Although the techniques presented here are based on data generated from capillary sequencing , they form a strong basis for developing algorithms specifically aimed at data generated by next generation sequencing technology . For example , sequences obtained using the Illumina platform can display substantial heterogeneity with regard to the depth of coverage across the genome segments after alignment . This means that more information ( e . g . samples ) will be available at some sites than others . This heterogeneity in information is intrinsically incorporated into the PPAs for individual sites through the Bayesian model specification . However , it will also be necessary to incorporate additional sources of error intrinsic to the specific platform being used , and this is the focus of ongoing work .
|
Characterising genetic variation in viral populations can have important implications in terms of understanding how viruses evolve within infected hosts . Modern sequencing technologies allow genetic information to be obtained faster , more affordably , and in much greater quantities than before . This allows new experimental procedures to be designed to explore aspects of pathogenesis that were previously unattainable , particularly with regard to mutations that occur at particular nucleotide sites that may confer a fitness advantage to the pathogen . This information can be used to study important issues such as the development of antiviral resistance , virulence , and/or changes in host-range specificity . Nonetheless , the experimental procedures used to generate the data can incorporate artefactual errors , and in order to optimise the information obtained from these studies techniques are required to characterise which sites exhibit mutations that may alter viral fitness . As both the depth of sequencing increases and the length of the region sequenced increases ( e . g . moves to whole genomes rather than smaller segments ) , large numbers of sites will exhibit some form of variation , and hence development of a probabilistic method to define and extract these sites-of-interest becomes more important . We tackle this problem here using a Bayesian framework .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"mathematics",
"statistics",
"genetics",
"biology",
"computational",
"biology",
"genetics",
"and",
"genomics"
] |
2011
|
A Bayesian Approach to Analyse Genetic Variation within RNA Viral
Populations
|
Lymphatic filariasis ( LF ) is a neglected tropical disease , and the Global Program to Eliminate LF delivers mass drug administration ( MDA ) to 500 million people every year . Adverse events ( AEs ) are common after LF treatment . To better understand the pathogenesis of AEs , we studied LF-patients from a treatment trial . Plasma levels of many filarial antigens increased post-treatment in individuals with AEs , and this is consistent with parasite death . Circulating immune complexes were not elevated in these participants , and the classical complement cascade was not activated . Multiple cytokines increased after treatment in persons with AEs . A transcriptomic analysis was performed for nine individuals with moderate systemic AEs and nine matched controls . Differential gene expression analysis identified a significant transcriptional signature associated with post-treatment AEs; 744 genes were upregulated . The transcriptional signature was enriched for TLR and NF-κB signaling . Increased expression of seven out of the top eight genes upregulated in persons with AEs were validated by qRT-PCR , including TLR2 . This is the first global study of changes in gene expression associated with AEs after treatment of lymphatic filariasis . Changes in cytokines were consistent with prior studies and with the RNAseq data . These results suggest that Wolbachia lipoprotein is involved in AE development , because it activates TLR2-TLR6 and downstream NF-κB . Additionally , LPS Binding Protein ( LBP , which shuttles lipoproteins to TLR2 ) increased post-treatment in individuals with AEs . Improved understanding of the pathogenesis of AEs may lead to improved management , increased MDA compliance , and accelerated LF elimination .
Lymphatic filariasis ( LF ) is a disabling neglected tropical disease that is caused by the mosquito-borne filarial parasites Wuchereria bancrofti , Brugia malayi and B . timori . Adult worms live in the human host’s lymphatic system and release larval parasites ( microfilariae or Mf ) that circulate in the blood . Infection and host inflammatory responses to the parasite can lead to severe morbidity including lymphedema , hydrocele and elephantiasis [1] . To combat this disease the WHO launched the Global Program to Eliminate Lymphatic Filariasis ( GPELF ) in the year 2000 with the goal of eliminating LF as a public health problem by 2020 . The program uses mass drug administration ( MDA ) , to cure infections , prevent disease , and reduce transmission of new infections . As of 2016 a total of 6 . 7 billion treatments had been delivered to more than 850 million individuals [2] , making GPELF the largest public health intervention for an infectious disease to date based on MDA . Drugs used for LF MDA include albendazole ( ALB ) , ivermectin ( IVM ) and diethylcarbamazine ( DEC ) . MDA with two-drug combinations is usually provided annually for 4–6 years . The combinations used are ALB with IVM in sub-Saharan Africa and ALB with DEC in other regions [1] . New studies have shown that combining all three drugs increases the anti-filarial effect and potentially decreases the number of required treatment rounds [3–7] . This new triple therapy ( IDA ) was recently recommended by the WHO as the preferred regimen for LF elimination in some settings [8] . Although LF treatment is safe , transient mild to moderate systemic adverse events ( AEs ) are common following treatment , and these are especially common in individuals with circulating Mf [3] . Furthermore , the risk of AEs and AE severity are positively correlated with blood Mf counts ( Mf/mL ) [9] . Systemic AEs are not direct effects of the drugs on the host , because they are quite uncommon in uninfected individuals [10] . The pathogenesis of these AEs is not completely understood , but they are believed to be trigged by host responses to dying filarial worms . Post-treatment AEs have been associated with increases in plasma levels of IL-6 , TNF-α and soluble TNF receptor [11 , 12] . We recently reported significant increases in 16 cytokines in persons who experienced AEs after treatment during a clinical trial that was performed in Papua New Guinea [13] . These results were consistent with LPS-like stimulation of cytokines with increases in TNF-α , IL-1β , IL-6 , IL-1RA and IL-10 . Wolbachia are intracellular α-proteobacteria that are present in filarial species that cause LF . The bacteria are hypothesized to trigger AEs when they are released by dying parasites after treatment . One study detected free Wolbachia DNA in blood collected 4–48 hours after LF treatment in individuals with moderate and severe AEs , but bacterial DNA was not detected in blood from most individuals with no or mild AEs [14] . Some features of AEs are consistent with the effects of LPS . A filarial ( Brugia malayi ) antigen with LPS-like characteristics was described some years ago [15] . However , the B . malayi-associated Wolbachia genome [16] does not include orthologues of genes responsible for the biosynthesis of lipid A ( a component of LPS ) [17] . It is therefore unlikely that B . malayi Wolbachia contains LPS in its cell wall . Bioinformatic analysis of the Wolbachia genome predicts the presence of a Wolbachia lipoprotein: peptidoglycan-associated lipoprotein ( PAL ) [18] . A synthetic , lipolated version of the N-terminus of Wolbachia PAL can signal through TLR2-TLR6 and induce pro-inflammatory responses in vitro in murine and human cells and in vivo in mice [18] . Additionally , the diacylated N-terminal polypeptide of the Wolbachia PAL ( WoLP ) was identified as the main trigger for a neutrophil inflammatory response through a TLR2-TLR6 dependent mechanism in vivo in human samples from individuals infected with Onchocerca volvulus [19] . Recently PAL was confirmed by proteomics as one of the most abundant proteins in extracts from adult B . malayi female worms [20] . Besides Wolbachia , post-treatment AEs could also be triggered by immune complexes ( IC ) that develop after treatment of LF . ICs are aggregated antigens , antibodies , and components of the complement cascade that can activate pro-inflammatory pathways [21] . It has been reported that filarial antigen levels increase with a concurrent decrease in filarial specific antibodies post-treatment , and these changes were temporally associated with the development of AEs , suggesting that AEs might be caused by IC [22] . Circulating IC ( CIC ) have also been shown to increase post-treatment [23] , and CIC precipitated from LF-infected individuals can activate granulocytes to release pro-inflammatory cytokines [24] . CIC activate the classical complement pathway . AEs are common after treatment for LF , and fear of AEs reduces population compliance with MDA [25] . Therefore , the goal of this study was to improve understanding of the pathogenesis of AEs after LF treatment . We hypothesized that AEs are caused when filarial worm components are released after treatment and interact with the host innate or adaptive immune systems and that this would be associated with specific biomarker and gene expression profiles . To test this hypothesis we measured filarial antigen , CIC , LPS Binding Protein ( LBP ) and components of the complement cascade in plasma before and after treatment , and we studied host transcriptional responses and cytokine profiles in LF-infected individuals who experienced AEs after treatment .
Buffy coat and plasma samples were collected during an open label filiariasis treatment study in the Agboville District in southeastern Côte d’Ivoire ( Clinicaltrials . gov NCT # 02974049 ) . Written informed consent was obtained from all participants . Adults with W . bancrofti microfilaremia were randomly assigned to one of four treatment arms ( all oral medications ) : the standard LF treatment regimen for Côte d’Ivoire ( 200μg/kg IVM plus 400mg ALB ) , IDA: 200μg/kg IVM plus 6mg/kg DEC and 400mg ALB , a single 400 mg dose of ALB , or a single 800 mg dose of ALB . A subset of ninety-five individuals treated with either IVM/ALB , IDA or 400mg ALB had samples processed for use in the AE study described in this paper . We selected these individuals based on the availability of pre- and post-treatment samples and clinical AE data . Metadata of these 95 individuals is shown in S1 Table . A physical examination was performed shortly before treatment , and vital signs were recorded . A review of systems ( ROS ) questionnaire was also completed to assess subjective symptoms prior to treatment . Venous blood ( 3 to 4 mL in EDTA ) was collected immediately before participants received treatment . Participants were interviewed and examined the next day to assess AEs , and venous blood was collected approximately 24 hours after treatment . Blood samples were centrifuged within an hour of collection , and plasma was removed . The buffy coat ( approximately 500μL ) was carefully aspirated with a pipette and added to 1 . 8mL of RNAlater ( Ambion , Foster City , CA ) . The plasma samples and buffy coat/RNAlater samples were stored at the study site at -20°C , shipped frozen , and later stored at -80°C . AEs were categorized as mild , or moderate . Those with moderate AEs ( n = 9 ) had at least two new or worsening subjective symptoms plus one objectively measured change in their vital signs ( an increase in axillary temperature of ≥ 0 . 8°C to at least 37 . 4°C post-treatment and/or a decrease in sitting systolic blood pressure of at least 20 mm Hg ) . Individuals with subjective or objective AEs that did not fulfill the criteria for moderate AEs were considered to have had mild AEs ( n = 24 ) . Individuals with no new objective or subjective symptoms after treatment were considered to have no AEs ( n = 62 ) . A direct sandwich enzyme immunoassay ( EIA ) was performed as previously described [13] . This assay uses the monoclonal antibody AD12 that binds to a carbohydrate epitope on circulating filarial antigen ( CFA ) . It is important to note that the carbohydrate epitope recognized by AD12 is present in many filarial glycoproteins [26] . However , the high molecular weight CFA is the only filarial antigen that is frequently detected in the blood of W . bancrofti-infected individuals . Pre- and post-treatment plasma samples from 95 individuals were tested in duplicate . The detection range of the CFA EIA was 6 . 3 to 400 ng/mL . CFA was detected in all samples , but two individuals had extremely high CFA levels that were above the upper detection limit of the assay . These samples were retested after dilution to obtain baseline CFA concentrations . The percent change in CFA relative to baseline following treatment was calculated for each participant . Sample pairs with pre-treatment values less than 20 ng/mL ( 7 individuals ) were excluded from the percent calculations , because they were near the lower detection limit of the assay . Kruskal-Wallis H tests were used to compare percent change values and absolute values between the three AE groups and the three treatment arms . Wilcoxon signed-rank tests were used to compare pre- and post-treatment CFA levels within the three AE groups . Nine paired ( pre- and post-treatment ) samples with high CFA levels at baseline were selected for this analysis; six of these participants had moderate AEs , one had mild AEs , and two had no AEs . 15mg of a monoclonal antibody ( DH6 . 5 ) that detects the same carbohydrate epitope as AD12 was directly conjugated to 2mL of agarose Affigel 10 beads ( Bio-Rad , Hercules , CA ) according to the manufacturer’s protocol . Conjugated beads were stored as a 50% solution in PBS . 40μL of conjugated beads were mixed with 50μl of human plasma and 300μl PBS and rocked overnight at 4⁰C . The beads were washed four times with cold PBS and then boiled in 1X NuPAGE LDS sample buffer ( Invitrogen , Carlsbad , CA ) to release bound antigens . Proteins were resolved by SDS-PAGE using a 4–12% bis-tris NuPAGE gradient gel ( Invitrogen ) and transferred to 0 . 45μM nitrocellulose membrane ( Amersham , Piscataway , NJ ) . Membranes were blocked with 5% milk in phosphate buffered saline with tween-20 ( PBS-T ) followed by incubation with a peroxidase-conjugated AD12 antibody ( 1:3000 dilution ) for one hour at room temperature . After washing , membranes were incubated with Clarity Western ECL substrate ( Bio-Rad ) . Chemiluminescence was detected by a ChemiDoc imager ( Bio-Rad ) , and results were analyzed using Image Lab 5 . 2 . 1 software . CIC were measured with a C1q ELISA . C1q was purchased ( Sigma-Aldrich , St . Louis , MO ) , and a previously published protocol was followed [13] . Plasma samples were available from 41 individuals for this assay ( 8 with moderate AEs , and 33 with no AEs ) , and both pre- and post-treatment samples were tested in duplicate . Negative control samples ( plasma samples from healthy North American subjects and deionized water ) were tested on each plate . Values were expressed as μg/mL of AHG ( aggregated human gamma globulin ) ( Invitrogen ) . The range of detection for the CIC ELISA was 0 . 0006 to 6 μg/mL of AHG , and all samples had detectable CIC . Mann-Whitney U tests were used to compare absolute CIC levels between the two AE groups pre- and post-treatment . The Wilcoxon signed-rank test was used to compare post-treatment CIC levels to baseline levels within AE groups . The Kruskal-Wallis H test was used to compare absolute CIC levels between the three treatment arms post-treatment . Nine individuals with moderate AEs were matched to individuals with no AEs following treatment . Matching was based on age , sex , baseline Mf count , and treatment arm ( S2 Table ) . Complement component 3 ( C3 ) , complement component 4 ( C4 ) and Factor B ( FB ) were measured in the 36 samples ( 18 matched case-control subjects pre- and post-treatment ) with ELISA kits ( AssayPro , St . Charles , MO ) . The C3 and C4 assays were competitive enzyme immunoassays , and the FB assay was a sandwich ELISA . Each sample was tested in duplicate and manufacturer’s protocol was followed . Paired t-tests were used to compare pre- and post-treatment complement component levels by AE group . LBP was measured with a sandwich ELISA kit ( Abnova , Taipei , Taiwan ) . Plasma samples from the same 18 matched case-control subjects were included . Each sample was tested in duplicate and manufacturer’s protocol was followed . Paired t-tests were used to compare pre- and post-treatment levels within both AE groups . The range of detection for the LBP ELISA was 5 to 50 ng/mL . Twenty-seven cytokines ( IL-1β , IL-1RA , IL-2 , IL-4 , IL-5 , IL-6 , IL-7 , IL-8 , IL-9 , IL-10 , IL-12 ( p70 ) , IL-13 , IL-15 , IL-17 , basic FGF , eotaxin-1 , G-CSF , GM-CSF , IFN-γ , IP-10 , MCP-1 , MIP-1α , MIP-1β , PDGF-BB , RANTES , TNF-α , and VEGF ) were measured with the MAGPIX system with the Bio-Plex Human 27-Plex Cytokine Panel and Bio-Plex Cytokine Reagent Kit ( Bio-Rad ) . Plasma samples from the same 18 matched-control subjects were included . A previous paper includes the detailed protocol [13] . Briefly , all samples were tested in duplicate , standard curves were calculated using the manufacturer’s software , and the analysis considered mean concentrations ( pg/mL ) from two duplicate wells . Wilcoxon signed-rank tests were used to determine whether cytokine levels changed after treatment in either of the two AE groups . Mann-Whitney U tests were used to determine whether pre- or post-treatment cytokine levels were different between the two AE groups . For graphing , fold changes were calculated for each cytokine by AE groups , and samples with cytokines below the detection limit were assigned a value equal to half the value of the lowest pre-treatment sample concentration measured for that cytokine . RNA was extracted from pre- and post-treatment buffy coat samples from the same 18 matched case-control subjects . Total RNA was extracted using Qiagen RNeasy kits ( Qiagen , Hilden , Germany ) according to the manufacturer’s protocol with an added homogenization step and on-column DNase digestion as follows . For each sample 200μL of the buffy coat/RNAlater mixture was added to 700μL of the kit’s RLT buffer and vortexed . The mixture was then added to a QIAShredder column ( Qiagen ) and centrifuged for 2 min at 16 , 000g . The flow-through was added to 700μL of 70% ethanol , and this mixture was added to a RNeasy column . Bound RNA was eluted in 30μL RNase-free water and stored at -80°C . The quality and quantity of RNA was verified with a Bioanalyzer 2100 ( Agilent Technologies , Cedar Creek , Texas ) . Samples were processed with the TruSeq Stranded Total RNA LT Sample Prep Kit with Ribo-Depletion using the manufacturer’s protocol ( Illumina , San Diego , CA ) . The RNA was high quality ( average RIN value 9 . 3 , range 8 . 5–10 ) . The 36 samples were sequenced in two batches . The first 14 samples were sequenced with the HiSeq2000 ( 2x 100 PE run and Illumina TruSeq Stranded Total RNA ) platform , and the remaining 22 samples were sequenced with HiSeq4000 ( 2x 150 PE run and Illumina TruSeq Stranded Total RNA ) . Between 28–41 million read fragments per sample were mapped to 19 , 864 protein-coding genes . Raw reads were mapped to protein coding genes using HISAT2 ( version 2 . 0 . 5 ) [27] , and the human reference genome GRCh38 . 84 . FeatureCounts [28] was used to count reads per gene . DESeq2 [29] was used to generate normalized read counts and to identify differentially expressed genes between the different comparator groups ( namely AEs vs . no AEs and pre- vs . post-treatment ) . The program “R” with the biocLite package “DESeq2” was used . Gene expression results from individuals before and after treatment were considered to be repeated measures for the analysis . Principal component analysis ( PCA ) was performed for 500 genes with the greatest variability in expression ( based on DESeq2 output , default settings ) , and distance metrics statistics [30] were used to determine whether groupings affected overall expression patterns . A clustering dendrogram ( Euclidean distance , complete linkage ) was also used to illustrate overall expression patterns , and this method considered all genes . A two-tailed binomial distribution with unequal variance ( for categorical data ) , and Mann-Whitney U tests ( for continuous variables ) were used to identify over-represented metadata variables in the different clustering groups such as baseline Mf/mL and treatment group . The online tool WebGestalt [31] was used to identify enriched KEGG pathways within genes that were upregulated post-treatment during AEs . The reference set was a list of all 19 , 864 genes with expression signals in the RNAseq data , and the default values were used except the significance level ( FDR < 0 . 05 ) . The program i-cisTarget [32] was used to identify enriched transcription factor binding sites in the upregulated gene set using default settings and database version 4 . 0 . GeneQuery is an online tool that can search the PubMed GEO database and compare transcriptional signatures to published gene expression profiles [33] . The input for the post-treatment AE profile were 744 genes that were identified by differential gene expression ( DESeq2 ) to be upregulated post-treatment in individuals with moderate AEs . CIBERSORT[34] is an analytical tool that can estimate the abundances of 22 leucocyte subtypes based on RNA-seq data . Pre- and post-treatment DESeq2 normalized read counts were used as input for the program . The standard LM22 ( 22 immune cell types ) was the signature gene file , and all default settings were used . Thirteen cell subtypes had very low representation in this dataset ( totaling less than 4% in all 36 samples ) , so the analysis was limited to the remaining subtypes . Percent change for each cell type post-treatment was calculated for the two AE groups , and Mann-Whitney U tests were used to assess the significance of differences by AE group . Random forest ( RF ) analysis was used to prioritize the 678 genes that were differentially expressed in individuals with moderate post-treatment AEs . RF was performed using the “R” package “randomForest” with 1000 trees and default values to analyse DESeq2 normalized read counts . Differentially expressed genes were ranked based on decreasing Mean Decrease in Accuracy values for 10 separate RF models . The Mean Decrease in Accuracy is the decrease in model accuracy from permuting the values in each feature . This metric is used to compare the impact of the variables in the model , and a large positive value indicates that a variable was closely linked to AE group across the dataset . Additional RNA was extracted from residual buffy coat samples that were available for 34 of the 36 samples that were subjected to expression profiling ( 17 individuals pre- and post-treatment ) as described above . Extracted RNA was treated with DNase I ( Invitrogen ) , and RNA was measured with a NanoDrop 1000 Spectrophotometer ( Thermo Scientific , Waltham , MA ) . Each sample was diluted to approximately 0 . 5ng/μL RNA with RNase-free water . cDNA was prepared with SuperScript II Reverse Transcriptase ( Invitrogen ) and with Oligo ( dT ) 12-18 according to the manufacturer’s protocol . SYBR Green based assays were performed for the top eight genes based on the RF analysis ( DIP2B , ZCCHC6 , RBPJ , PELI1 , FNDC3B , TLR2 , LTBR , NT5C2 ) that were upregulated in peripheral blood leukocytes ( PBL ) after treatment in participants who experienced moderate AEs . Four housekeeping genes ( SDHA , ACTB , HPRT1 and YWHAZ ) were used as controls for these experiments . We chose these based on prior validation as housekeeping genes by others [35] and because our results confirmed their stable gene expression before and after treatment . Pre-validated primer sets for the eight target genes were purchased from KiCqStart SYBR Green Primers ( Sigma-Aldrich ) , and primers for the four housekeeping genes were made using previously published sequences ( IDT , Coralville , IA ) ( S3 Table ) . Real-time PCR reactions were performed with 10μL of SYBR Green Master Mix ( Applied Biosystems , Foster City , CA ) , 450 nmol/L of each primer , and 2μL cDNA ( approx . 1ng RNA ) with a final volume of 20μL . Thermal cycling was performed for 40 cycles with a QuantStudio 7-Plex Real-Time PCR System ( Applied Biosystems ) , and cycle threshold ( Ct ) values were determined using the manufacturer’s software . All samples were tested in duplicate , and each plate included a negative water control and a RNA sample that had not been treated with reverse transcriptase . Delta delta Ct values were calculated [36] , using the geometric mean Ct value of three housekeeping genes ( SDHA , ACTB and YWHAZ ) as a normalization factor [35] . Student’s t-tests were performed to compare baseline and post-treatment delta Ct values by AE group . All statistical analyses were performed with IBM SPSS ( version 23 ) . Shapiro-Wilk tests were used to test for normality in each sample set , and additional tests were performed as described in each section above . Logistic regression analysis was performed with the binary dependent variable AEs ( moderate AEs vs . no AEs ) . The independent variables considered included age , sex , treatment arm , baseline Mf/mL , and baseline CFA level . Separate RF analyses of gene expression and plasma biomarker data were performed 10 times using 1000 trees . The output was the average Mean Decrease in Accuracy over the 10 runs for each variable . Institutional review boards in Cleveland , USA ( University Hospitals Cleveland Medical Center IRB #08-14-13 ) and in Côte d’Ivoire ( Comité National d’Ethique et de la Recherche , CNER , N: 008/MSLS/CNER/-kp ) approved the clinical trial study protocol . Written informed consent was received from all participants prior to inclusion in the study .
This study of the pathogenesis of AEs that occur after LF treatment used human samples that were obtained as part of a clinical trial for LF that was conducted in Côte d’Ivoire . Full results from that study have not yet been published , but early results from the study have been reported in published abstracts [4 , 5] . Briefly , 189 W . bancrofti-infected adults were randomly assigned to one of four treatment arms as described above , and all participants had AE assessments performed 24 hr after treatment . The AE study enrolled a subset of 95 treated participants . S1 Table summarizes the specific analyses that were performed on samples from each of the 95 individuals . Nine of these participants experienced moderate AEs ( Table 1 ) , 24 had mild AEs , and 62 had no AEs . There was no difference in age or sex distribution between the three AE groups ( S4 Table ) . Baseline CFA levels were positively correlated with baseline Mf counts ( Spearman’s rho: 0 . 51 , P < 0 . 001 ) , and absolute CFA levels were significantly higher at baseline in individuals who developed moderate AEs compared to individuals who developed mild or no AEs after treatment ( P = 0 . 012 by Kruskal-Wallis H test ) . Plasma CFA levels increased post-treatment in all three AE groups , but the increases were greater in persons with moderate AEs ( P < 0 . 05 by Kruskal-Wallis H test ) ( Fig 1 ) . Percent changes in CFA levels post-treatment were significantly lower in the individuals treated with only ALB compared to those in individuals treated with IVM/ALB or IDA ( P < 0 . 0001 by Kruskal-Wallis H test ) . Western blot analysis was performed for nine pre- and post-treatment plasma pairs to compare CFA patterns detected in plasma before and after treatment . All nine pre-treatment plasma samples contained only a single high molecular weight parasite antigen as expected , and this antigen was also present in post-treatment plasma samples . However , four of the post-treatment plasma samples contained many parasite antigens that were not present before treatment . Two examples are shown in S1 Fig ( P1 had moderate AEs , and P2 had mild AEs ) , and this pattern was also observed in plasma from two other participants who experienced moderate AEs following treatment . Western blot results obtained with five other post-treatment plasma samples tested ( 3 from persons with moderate AEs , and 2 from persons with no AEs ) were no different from those observed in pre-treatment samples . All pre- and post-treatment samples contained CIC . However , there was no difference in CIC levels between the two AE groups before or after treatment , and CIC levels did not significantly change after treatment in either AE group ( S2 Fig ) . There was also no difference in post-treatment CIC levels between the treatment arms . C3 levels significantly decreased post-treatment in individual with moderate AEs , but this change was not observed in individuals with no AEs ( Fig 2A ) . C4 levels did not change in either AE group ( Fig 2B ) . Factor B levels decreased post-treatment in most individuals with moderate AEs , but three individuals had increases in FB levels , and the group differences were not significant ( Fig 2C ) . LBP was detected in all pre- and post-treatment samples . LBP levels increased post-treatment in individuals with moderate AEs ( P = 0 . 0007 by paired t-test ) , but they did not increase in individuals with no AEs ( Fig 2D ) . Plasma cytokine levels before and after treatment are shown by AE group in Fig 3 . Seven cytokines ( IL-8 , MCP-1 , VEGF , TNF-α , MIP-1β , G-CSF and IFN-γ ) increased post-treatment only in individuals who experienced moderate AEs ( P < 0 . 05 by Wilcoxon signed-rank test ) . Five cytokines ( IL-6 , IL-10 , IL-1RA , IP-10 and MIP-1α ) increased post-treatment in individuals with and without AEs ( P < 0 . 05 by Wilcoxon signed-rank test ) , but three of these ( IL-6 , IL-10 and IL-1RA ) had significantly higher levels post-treatment in individuals with moderate AEs compared to individuals with no AEs ( P < 0 . 05 by Mann-Whitney U tests ) ( Fig 3 ) . The remaining 15 cytokines ( IL-1β , IL-2 , IL-4 , IL-5 , IL-7 , IL-9 , IL-12 ( p70 ) , IL-13 , IL-15 , IL-17 , basic FGF , eotaxin-1 , GM-CSF , PDGF-BB and RANTES ) did not change after treatment in either AE group . There was no difference in pre-treatment cytokine levels between individuals that would develop moderate AEs and individuals that would not develop AEs for any of the 27 cytokines . Raw and processed RNA-seq data are available to the public on NCBI’s Gene Expression Omnibus ( Accession number: GSE110146 ) . We analyzed changes in gene expression in PBL after treatment to further elucidate host responses associated with AEs . Post-treatment gene expression profiles from persons who developed moderate AEs clustered together using a clustering dendrogram based on gene expression profiles across all genes ( Fig 4A ) . Post-treatment AE samples were significantly overrepresented in the fourth group ( bolded in Fig 4A , P-value < 0 . 0001 for enrichment within the cluster , two-tailed binomial distribution with unequal variance ) . Higher levels of baseline Mf/mL were also observed in this group ( P-value = 0 . 038 , Mann-Whitney U test ) , but none of the other metadata categories ( treatment arm or village ) were over-represented . Age also did not affect clustering . A similar pattern was observed by principal components analysis ( Fig 4B ) , where post-treatment moderate AE samples clustered together and were clearly separated from their pre-treatment controls ( P-value = 0 . 005 by PERMANOVA [30] ) . No other differences were significant between the four groups by PCA . We used differential gene expression analysis to identify the genes that were responsible for the clustering of the post-treatment moderate AE samples . At a very stringent significance threshold ( P < 10−5 according to DESeq2 output ) , 783 genes were identified to be upregulated after treatment ( n = 744 ) or before treatment ( n = 39 ) in individuals who experienced moderate AEs ( Fig 4C ) . No differences were observed pre- or post-treatment in individuals with no AEs when this stringent significance threshold was used . However , at a less stringent P-value of 0 . 05 , there were 126 genes upregulated post-treatment and 19 genes upregulated pre-treatment in individuals without AEs ( Fig 4D ) . There was only one overlapping gene in the genes upregulated pre-treatment in individuals with and without AEs , whereas the majority of the genes upregulated post-treatment in individuals with no AEs were also upregulated post-treatment in individuals with AEs ( Fig 4E ) . We then assessed whether there was evidence for functional enrichment in the genes upregulated post-treatment in individuals with AEs . Among the 744 upregulated genes post-treatment in the AE samples a total of 35 enriched biological pathways ( KEGG ) were identified ( S5 Table ) , and these included TLR signaling and downstream pathways such as NF-κB , TNF and Jak/STAT . Many individual genes in the TLR signaling pathway , including TLR2 , TLR6 , STAT1 and STAT2 , were identified by DESeq2 to be significantly upregulated post-treatment in individuals with AEs . A separate analysis ( i-cisTarget ) predicted that six transcription factors were over-represented in the differentially expressed genes ( S6 Table ) , and three of these , STAT1 , STAT2 and IRF1 , are downstream of TLR signaling . Convincingly , STAT1 and STAT2 were therefore identified by two independent analyses , signifying the importance of these two transcription factors in the development of AEs . IRF1 is activated by IFN-γ and is a major transcription factor of IL-8 , and correspondingly both IFN-γ and IL-8 levels significantly increased post-treatment in individuals with AEs . The complete TLR signaling pathway highlighting the individual upregulated genes , KEGG pathways and transcription factors , was constructed with the use of the online database SPIKE [37] ( S3 Fig ) . Finally we wanted to compare our newly identified LF AE transcriptional signature to published gene expression profiles . The post-treatment AE transcriptional signature of the 744 upregulated genes was very similar to multiple published endotoxin exposure gene expression profiles , in addition to many other profiles ( S7 Table ) . After successfully identifying a significant transcriptional signature of post-treatment AEs , we explored whether a pre-treatment transcriptional signature could predict what individuals would go on to develop AEs after treatment . There were no significant differentially expressed genes at baseline between individuals that would develop moderate AEs , and individuals that would not develop AEs ( by DESeq2 ) . Changing cell populations can have a large effect on gene expression profiles . Differential cell counts were unavailable , so cell type proportions were estimated using the RNA-seq data . This analysis suggested that neutrophils increased more and lymphocytes decreased more post-treatment in individuals with AEs compared to individuals with no AEs ( Fig 5A ) . These changes are consistent with stress-type immune responses . For simplicity , B cells ( memory and naïve ) , T cells ( CD8 , CD4 naïve and memory resting ) and NK cells were combined into one category ( lymphocytes ) in Fig 5A , and ungrouped data are presented in S4 Fig . Estimated leucocyte proportions at baseline were very similar between the two AE groups , but individuals who experienced post-treatment AEs had significantly fewer estimated memory B cells compared to individuals who did not develop AEs after treatment ( Fig 5B ) . The genes upregulated post-treatment in individuals with AEs were prioritized by importance for AE development using a machine-learning tool ( RF analysis ) . This was done in order to identify genes with the strongest associations between expression levels and development of AEs and to identify genes of interest for PCR validation . Table 2 shows the top 15 genes that were upregulated in persons who developed moderate AEs after treatment . However , based on this analysis it was not possible to determine whether the gene expression changes were the cause or effect of the AEs that were experienced . qRT-PCR studies were performed to confirm whether expression of genes identified by DESeq2 and RF analyses was actually increased post-treatment in individuals with moderate AEs . Increased expression after treatment was confirmed for seven of the top eight genes ( DIP2B , ZCCHC6 , PELI1 , FNDC3B , TLR2 , LTBR and NT5C2 ) ( Fig 6 ) . Expression of the eighth gene ( RBPJ ) did not change after treatment in either AE group . Expression of the housekeeping gene HPRT1 , did not change with treatment in either AE group ( as expected ) . We were unable to identify any pre-treatment transcriptional signature that could predict moderate AEs . We therefore wanted to assess if any metadata or baseline infection parameter was associated with the development of moderate AEs . A logistical regression was performed to consider effects of age , sex , treatment arm , baseline Mf/mL and baseline CFA on the risk for development of post-treatment moderate AEs . A total of 71 individuals were included in the model ( 9 moderate AEs and 62 no AEs ) . The logistic regression model was statistically significant , X2 ( 6 ) = 22 . 1 , P = 0 . 0012 . The model explained 50 . 2% ( Nagelkerke R2 ) of the variance in AE outcome , and correctly predicted 93% of outcomes . However the model was better at predicting people who did not develop AEs; it correctly predicted only 44 . 4% of the individuals who developed moderate AEs . Increasing baseline CFA levels were associated with increased likelihood of developing AEs ( P = 0 . 022 ) , but the other independent variables did not significantly contribute to the model . RF analysis was performed on the same dataset ( 71 individuals ) , and this also identified the baseline CFA level as the best predictor for subsequent development of AEs . However , treatment arm was also a positive predictor in the RF model , and could therefore be related with the development of AEs ( Table 3 ) . It was surprising that baseline Mf count was not identified by the logistic regression model or RF to significantly contribute to correctly predicting AEs . However , baseline Mf counts were higher in individuals who developed moderate AEs ( geometric mean 343 Mf/mL ) compared to individuals with no AEs ( geometric mean 188 Mf/mL ) , and this difference was significant ( P = 0 . 036 by Mann-Whitney U test ) . RF analysis was also used to identify the variable ( CFA , CIC , C3 , C4 , FB or LBP ) that was best at classifying AE outcome based on post-treatment fold change in the 18 matched case-control subjects . LBP changes after treatment was the only variable that was significantly associated with the development of AEs ( S8 Table ) .
Filarial antigen levels increased in plasma after treatment in individuals with moderate AEs , and this agreed with our recently published results from a separate clinical trial [13] . Western blot results from this study showed that many new filarial antigens with the carbohydrate epitope detected by the monoclonal antibody AD12 appeared in the blood 24 hours after treatment in some individuals . In contrast , only a single high molecular weight antigen circulates in the blood of W . bancrofti-infected individuals without treatment [26] . We postulate that treatment kills or injures worms so that they release internal antigens that are normally concealed inside the parasite . Results from this study also confirmed our previous finding that plasma CIC levels do not increase after treatment of LF in persons who develop moderate AEs [13] . This finding suggests that AEs after LF treatment are not caused by CIC . The complement cascade ( classical pathway ) modulates pro-inflammatory effects of CIC [38] . Activation of the classical complement cascade leads to decreased C3 and C4 , whereas activation of the alternative pathway ( AP ) leads to decreased C3 and factor B ( FB ) . Our results are most consistent with activation of the AP by parasite antigens , as C3 and FB decreased in individuals with moderate AEs while there was no change in C4 levels . The RNA-seq data also supports the AP hypothesis , because expression of CFP ( complement factor properdin- a positive regulator and initiator of the AP ) significantly increased post-treatment in individuals with moderate AEs ( adjusted P-value 0 . 001 , DESeq2 ) . In contrast , expression of C4B ( basic form of C4 , part of the classical pathway ) significantly decreased ( adjusted P-value 0 . 04 , DESeq2 ) post-treatment in individuals with moderate AEs . Additionally , IFN-γ and TNF-α are known to induce FB synthesis [39] . This could account for the inconsistent FB levels post-treatment in individuals with moderate AEs , because both IFN-γ and TNF-α increased in these individuals; the positive stimulus of these cytokines may have counteracted decreases in FB levels as it was used in the AP . In summary , many different filarial antigens were transiently released post-treatment , but they did not appear to form CIC or activate the classical complement cascade . PBL appeared to respond to the release of filarial antigens by releasing cytokines , and the cytokine profiles of post-treatment AEs in LF infected individuals are complex . In our previous study of samples from a treatment trial in Papua New Guinea , we reported that 16 cytokines increased post-treatment in individuals with moderate AEs [13] , and eight of these cytokines were also increased after treatment in this study ( IL-1RA , IL-6 , IL-10 , G-CSF , MCP-1 , MIP-1β , TNF-α , and VEGF ) . It is not surprising that more cytokines increased in the previous study , because more time-points were sampled . Also , participants in the Papua New Guinea study had higher blood Mf counts and higher rates and severity of AEs than participants in the present study , and this may account for their stronger cytokine responses . A new finding in this study was the increase in IFN-γ post-treatment in individuals with moderate AEs . This was supported by the fact that IRF1 ( downstream of IFN-γ ) was identified to be an important transcription factor for AE development . The increase in IL-8 levels paralleled the increase in neutrophils post-treatment in individuals with moderate AEs . Results from this study did not confirm the finding from the prior Papua New Guinea study that high levels of eotaxin-1 pre-treatment are a risk factor for development of post-treatment AEs . Again this discrepancy may be related to differences in infection intensity between participants in the two studies . Levels of IL-6 and TNF-α have previously been shown to be positively correlate to levels of Wolbachia DNA in human plasma 48 hours after treatment of LF [40] . RNA-seq was performed to better understand changes in leukocyte gene expression that occur in persons who experienced moderate AEs after treatment . We identified a distinctive transcriptional signature associated post-treatment moderate AEs with 783 genes that were differentially expressed ( at the P < 10−5 level ) in persons who experienced moderate AEs . In contrast , no gene was differentially expressed at that level of significance before or after treatment in individuals who did not experience AEs . 95% of the 783 genes associated with AEs exhibited increased expression post-treatment . Thus , moderate AEs were primarily associated with upregulation of gene expression and not with gene suppression . A total of 126 genes were upregulated post-treatment in individuals with no AEs at the low stringency P < 0 . 05 level , but 83% of these genes were also upregulated post-treatment in individuals with moderate AEs . Thus changes in gene expression after treatment did not always lead to clinically evident AEs . The transcriptional signature results are consistent with the hypothesis that Wolbachia lipoprotein activates TLR2-TLR6 [18 , 41] , as bacterial lipoproteins can induce pro-inflammatory responses through TLR2 signaling and NF-κB and STAT1 activation [42] . The finding that TLR2 was one of the genes most highly associated with the development of moderate AEs also supports this hypothesis . Furthermore , LBP was found to increase post-treatment in plasma from individuals with moderate AEs , and RF analysis identified LBP ( fold change post-treatment ) as the best variable for classifying AE outcome . LBP is an acute-phase protein that is mostly known for its function of shuttling LPS to TLR4 via CD14 . However , it can also shuttle lipoproteins to TLR2 also via CD14 [43] as would be the case with PAL . CD14 expression was upregulated post-treatment in individuals with moderate AEs , and it had one of the most significant adjusted P-values ( 6 . 4e-26 ) in the dataset . CD36 is another accessory receptor for the TLR2-TLR6 heterodimer [44] , and this gene was also upregulated post-treatment in individuals with moderate AEs ( adjusted P-value 0 . 004 ) . The LF AE transcriptional signature was similar to previously published endotoxin exposure gene expression profiles further supporting the Wolbachia lipoprotein hypothesis . Since multiple filarial antigens that are normally not accessible to the immune system are released post-treatment , it is possible that Wolbachia components are released into the host’s circulation in a similar fashion . Thus we cannot be certain that Wolbachia lipoprotein is the only or prime trigger for AEs . It is possible that other filarial components can signal through TLRs and contribute to the development of AEs , and this might explain why expression of many other TLRs ( TLR1 , TLR4 , TLR5 , TLR6 and TLR8 ) were upregulated post-treatment in individuals with moderate AEs . Many different ligands can activate TLR2 , including protozoan ligands such as GPI anchors from Trypanosoma cruzi [45] and Leishmania major [46] . Wolbachia-independent activation of the immune system causing severe AEs is seen in Loa loa ( a filarial worm that lacks Wolbachia [1] ) infected individuals post-treatment suggesting that Wolbachia is not the sole cause of AEs after anti-filarial treatment . TLR signaling is clearly associated with the development of AEs , but complement activation has similar downstream effects , and there is considerable crosstalk between these two pathways [47] . Thus both TLR signaling and the complement AP could be actively involved in the pathogenesis of AEs . Another possible mechanism for AEs following treatment of LF is that treatment abrogates the normally dominant Th2 immune responses stimulated by helminth infections that interfere with the expression and function of TLRs [48] . Increased TLR expression and signaling after treatment may then induce pro-inflammatory Th1 responses causing AEs . Indeed , classic Th1 cytokines TNF-α and IFN-γ were increased after treatment in people with moderate AEs . We did not detect a pre-treatment transcriptional signature that was a significant risk factor for development of post-treatment AEs . Baseline CFA levels were the best predictor for moderate AEs in this study , and this was the only variable that was a significant predictor by logistic regression . CFA levels are correlated with Mf counts and likely with adult worm numbers , and this result suggests that CFA levels are related to the total parasite biomass that can potentially contribute to the development of AEs after treatment . An important finding from this study was that post-treatment AEs in LF-infected individuals are associated with upregulation of hundreds of genes . A prioritized list of the top 15 genes important for the development of AEs is listed above in Table 2 . These genes and their associated pathways may provide insight into the pathogenesis of AEs . In addition to the TLR pathway , Notch , NF-κB and IL-1 signaling were common themes . Four of the top 15 genes ( RBPJ , TLR2 , ALDH1A2 and APLP2 ) are involved in TLR/Notch signaling . The Notch pathway is involved in development and is conserved from Drosophila to mammals . RBPJ is involved in the crosstalk between TLR and Notch signaling that is thought to help fine-tune the immune response through negative and/or positive feedback ( S5 Fig ) [49] . NF-κB is another downstream pathway of Notch signaling , and three of the top 15 genes ( PELI1 , TLR2 and LTBR ) are involved in NF-κB signaling . TLR2 is a receptor for the canonical NF-κB pathway , and PELI1 is involved in intracellular downstream signaling . The canonical pathway results in the release of pro-inflammatory cytokines , such as IL-6 and TNF-α . Thus , activation of this pathway is consistent with the cytokine profiles individuals who experienced moderate AEs after treatment . Interestingly , lymphotoxin beta receptor ( LTBR ) stimulation has been shown to enhance the LPS-induced expression of IL-8 via the combined action of NF-κB and IRF1 [50] , and this is consistent with our results . Finally , IL-1 is clearly associated with the development of AEs . IL-1RAP was one of the top 15 genes identified by RF , and this gene is one of two co-receptors for IL-1 . Additionally , expression of the second co-receptor IL-1R1 was increased post-treatment in individuals with AEs . Both expression ( this study ) and plasma levels ( our prior study [13] ) of IL-1β increased after treatment in individuals with AEs . Inhibitors of the IL-1 pathway were also upregulated post-treatment in individuals with AEs . This included both increased expression and protein levels of IL-1RA and increased expression of the IL-1β decoy receptor IL-1R2 . These results illustrate the importance of the balance between the pro-inflammatory effects of IL-1β and the anti-inflammatory effects of IL-1RA for AE development . One limitation of this study was that the RNA-seq was performed on mixed PBL samples . This makes it difficult to separate the effects of altered gene expression from the effects of changing cell type proportions . Separating different types of leukocytes was not feasible , because the samples were collected in rural Côte d’Ivoire and processed in a simple field lab . On the other hand , this study provides insight into the pathogenesis of post-treatment AEs . Additionally , it was possible to estimate the different cell subtypes present in PBLs using the RNA-seq data , so we could associate changes in gene expression with altered cell types to decrease the chance that the former was a directly result of the latter . For example , if the post-treatment AE transcriptional signature had been similar to a neutrophil gene expression profile it could have been caused by the increasing proportion of neutrophils post-treatment and not due to specific neutrophil activation during AEs . Another limitation was that we did not study an untreated control group , because it would have been unethical to withhold treatment from infected individuals . For the cytokine analysis we did not correct for multiple comparisons , however , the results are generally consistent with our previous findings [13] , and this increases our confidence in the results . Based on the standard significance level of 0 . 05 , approximately 3 differences would be expected to be significant by chance for 54 tests ( 27 cytokines measured pre- and post-treatment in two AE groups ) , whereas 17 comparisons were significant at the 0 . 05 level in this study . The semi-quantitative Filariasis Test Strip ( FTS ) was used to assay CFA in the field , whereas a quantitative ELISA was used to measure CFA levels in the laboratory setting in this study . Newer studies have demonstrated that tests that detect LF CFA ( including FTS ) can cross-react with L . loa antigens that circulate in blood from a subset of individuals with heavy infections [51] , and with biological samples from animals infected with L . loa and Onchocerca ochengi [52 , 53] . This cross-reactivity was not a concern for this study because L . loa is not endemic to Côte d’Ivoire , and the area of Côte d’Ivoire where the study was conducted is non-endemic for O . volvulus . In future studies it would be interesting to measure Wolbachia DNA in pre and post-treatment samples from individuals with AEs after treatment of LF and onchocerciasis to try to correlate bacterial DNA release with host expression profiles post-treatment . Wolbachia DNA has been shown to increase post-treatment in individuals with AEs after treatment of LF [14] . Additionally , peak Wolbachia DNA levels have been shown to be correlated with AE reaction scores in individuals treated with DEC or IVM for onchocerciasis [54] . This study included first global RNA-seq analysis of PBLs from LF-infected individuals , and it has provided novel insights into the pathogenesis of a clinically relevant problem . The samples were ideal for studying AEs after LF treatment , because each post-treatment sample was paired with a pre-treatment sample from the same individual . This internal control improved our ability to study the AE phenotype in humans . AEs represent a significant challenge for the global program to eliminate LF , and the fear of AEs in communities receiving MDA is a main factor that reduces compliance [25] . Minimizing the impact of AEs has therefore been identified as a key component for successful MDA programs [25] . More than 850 million individuals have been treated as part of GPELF , and a significant percentage of these individuals experience AEs . This study has also provided a framework for investigating the host responses associated with severe AEs that occur after treatment of other filarial worms such as O . volvulus and L . loa . Treatment of other , more familiar infections can also result in severe AEs that are caused by host responses to dying pathogens . This Jarisch-Herxheimer reaction occurs after antibiotic treatment of spirochetal infections such as syphilis , Lyme disease , leptospirosis , and relapsing fever , and it is also hypothesized to be caused by the release of bacterial lipoproteins that activate TLR2 [55] . The transcriptomic response during the Jarisch-Herxheimer reaction has not been studied , so the dataset from this study could provide a valuable starting point for research on this related clinical problem . To recap our major findings , this study has provided new insights regarding the pathogenesis of post-treatment AEs in LF-infected individuals . Our results are consistent with the hypothesis that a Wolbachia lipoprotein triggers AEs by binding to TLR2-TLR6 , but other uncharacterized filarial antigens might also play a role . Since TLR , NF-κB , and TNF pathways are involved , these pathways could potentially be targeted to prevent or treat AEs after LF treatment . We also found that high pre-treatment CFA levels were the best predictors of post-treatment AEs . This finding could be relevant for treatment-naïve areas with high LF infection prevalence and intensities . Individuals with high CFA levels pre-treatment ( assessed with the FTS [56] ) could be offered non-steroidal anti-inflammatory medications together with anti-filarial medications for home management of moderate or severe AEs . However , a positive FTS from an individual who resides in or has traveled to an area that is also endemic for L . loa needs to be interpreted with caution due to the issues of cross-reactivity mentioned above . Information from this study should allow program managers and drug distributors to reassure populations and communicate to them that AEs experienced after LF treatment are transient and caused by host responses to dying or injured parasites .
|
Lymphatic filariasis ( LF ) is a disabling parasitic disease that affects millions of people in the developing world . The Global Programme to Eliminate Lymphatic Filariasis ( coordinated by the World Health Organization ) uses mass administration of antifilarial medications to cure infections , prevent disease , and reduce transmission . Some individuals develop adverse events ( AEs ) after treatment , and this can reduce willingness of persons in endemic areas to accept treatment . The purpose of this study was to improve understanding of the cause of AEs following treatment . We hypothesized that parasite antigens released into the blood following treatment trigger inflammatory responses that lead to AEs . To test this hypothesis we collected blood from LF-infected individuals before and after treatment and clinically assessed them for AEs . We measured parasite antigens , cytokines and other components of the immune system in blood samples and compared post-treatment changes in persons with and without AEs . We also assessed changes in transcription profiles in peripheral blood leukocytes that were associated with post-treatment AEs . Post-treatment changes in transcription profiles and in immune proteins and parasite components in plasma suggest that systemic AEs are triggered by death of the parasites following treatment with release of parasite antigens and Wolbachia bacteria into the circulation . Improved understanding of the pathogenesis of post-treatment AEs may help to improve messaging related to mass drug administration programs and lead to improved AE management .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"body",
"fluids",
"clinical",
"research",
"design",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"wolbachia",
"research",
"design",
"developmental",
"biology",
"molecular",
"development",
"bacteria",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"animal",
"cells",
"gene",
"expression",
"adverse",
"events",
"pathogenesis",
"blood",
"plasma",
"immune",
"system",
"blood",
"anatomy",
"cell",
"biology",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2019
|
Systems analysis-based assessment of post-treatment adverse events in lymphatic filariasis
|
The evolution of drug resistance has a profound impact on human health . Candida glabrata is a leading human fungal pathogen that can rapidly evolve resistance to echinocandins , which target cell wall biosynthesis and are front-line therapeutics for Candida infections . Here , we provide the first global analysis of mutations accompanying the evolution of fungal drug resistance in a human host utilizing a series of C . glabrata isolates that evolved echinocandin resistance in a patient treated with the echinocandin caspofungin for recurring bloodstream candidemia . Whole genome sequencing identified a mutation in the drug target , FKS2 , accompanying a major resistance increase , and 8 additional non-synonymous mutations . The FKS2-T1987C mutation was sufficient for echinocandin resistance , and associated with a fitness cost that was mitigated with further evolution , observed in vitro and in a murine model of systemic candidemia . A CDC6-A511G ( K171E ) mutation acquired before FKS2-T1987C ( S663P ) , conferred a small resistance increase . Elevated dosage of CDC55 , which acquired a C463T ( P155S ) mutation after FKS2-T1987C ( S663P ) , ameliorated fitness . To discover strategies to abrogate echinocandin resistance , we focused on the molecular chaperone Hsp90 and downstream effector calcineurin . Genetic or pharmacological compromise of Hsp90 or calcineurin function reduced basal tolerance and resistance . Hsp90 and calcineurin were required for caspofungin-dependent FKS2 induction , providing a mechanism governing echinocandin resistance . A mitochondrial respiration-defective petite mutant in the series revealed that the petite phenotype does not confer echinocandin resistance , but renders strains refractory to synergy between echinocandins and Hsp90 or calcineurin inhibitors . The kidneys of mice infected with the petite mutant were sterile , while those infected with the HSP90-repressible strain had reduced fungal burden . We provide the first global view of mutations accompanying the evolution of fungal drug resistance in a human host , implicate the premier compensatory mutation mitigating the cost of echinocandin resistance , and suggest a new mechanism of echinocandin resistance with broad therapeutic potential .
The emergence of drug resistance is an evolutionary process with a profound impact on human health . The widespread deployment of antimicrobial agents in medicine and agriculture exerts strong selection for organisms with enhanced capacity to survive and reproduce in the presence of drug , which has led to the rapid emergence of drug resistance in diverse pathogen populations [1]–[4] . The evolution of drug resistance compromises the efficacy of drugs that we depend on critically for a myriad of therapeutic interventions , and has striking economic consequences . The annual cost attributable to the evolution of drug resistance in the United States alone exceeds $33 billion to cover treatment of patients with drug-resistant infections , additional pesticides required to manage resistant pests , and loss of crops to resistant pests [5] . The emergence of drug resistance in fungal pathogens is of particular concern given the increasing incidence of invasive fungal infections , and the limited number of antifungal drugs . Fungi can cause life-threatening infectious disease in immunocompromised hosts , as well as in healthy humans , and the incidence of fungal bloodstream infections has increased by 207% in recent decades [6]–[8] . Fungi are eukaryotes and share close evolutionary relationships with their human hosts , which limits the number of drug targets that can be exploited to selectively kill fungal pathogens yet minimize host toxicity [1] , [3] . Even with current treatment options , mortality rates due to invasive fungal infections can reach 50–90% depending on the pathogen and patient population [7] , [8] , demanding new strategies to prevent the evolution of drug resistance and enhance the efficacy of antifungal drugs . The evolution of drug resistance is contingent on genetic variability , the ultimate source of which is mutation . One of the most fundamental questions of central importance to predicting and preventing the evolution of drug resistance is which mutations accompany the evolution of drug resistance in the human host . Developments in sequencing technology [9]–[11] now enable this question to be addressed on a genome-wide scale to reveal the identity of mutations that either confer drug resistance in a clinically relevant context or that modify the fitness consequences of resistance mutations . Whole genome sequencing has been applied to bacteria and has revealed principles underpinning the evolution and transmission of drug-resistant pathogens [12] , risk factors for the evolution of drug resistance [13] , and population dynamics during the evolution of drug resistance in vitro [14] . In fungi , changes in genome-wide gene expression and chromosomal alterations that accompany the evolution of drug resistance have been monitored in experimental populations that evolved resistance in vitro [15] , [16] , and targeted sequence and expression analysis of specific genes has been implemented to identify mechanisms of resistance that evolve in the human host [1] , [17] . However , a global approach to mapping mutations that underpin the evolution of fungal drug resistance has yet to be achieved . Candida glabrata is a leading fungal pathogens of humans and provides a particularly powerful system for studying the evolution of drug resistance in a human host . Candida species are the fourth most common cause of hospital acquired blood-stream infections and are the most prevalent cause of invasive fungal infection worldwide , with mortality rates approaching 50% [7] , [18] . C . glabrata is now second to C . albicans as the most prevalent Candida species isolated from clinical specimens [7] , [8] , [19] . This is due in part to both intrinsic and rapidly acquired resistance of C . glabrata to the azoles , which are the most widely used class of antifungal drugs and inhibit the biosynthesis of the key sterol in fungal cell membranes , ergosterol [20] . As a consequence , the echinocandins are the front-line therapeutic agent for C . glabrata infections [19] . C . glabrata is closely related to the model yeast Saccharomyces cerevisiae and is placed within the Saccharomyces clade rather than the Candida clade to which the leading cause of candidiasis , C . albicans , belongs [21]–[23] . It is thought that C . glabrata emerged as a human pathogen independently from other Candida species . Notably , gene families associated with pathogenicity in C . albicans including iron acquisition and host cell adhesion and invasion are absent from C . glabrata [24] . C . glabrata is an obligate haploid and mating has never been reported , although two mating types , mating type switching , and other mating and meiotic machinery have been described [25]–[27] . To increase genetic diversity C . glabrata undergoes chromosomal translocations and variation in gene copy number [28] , [29] , mechanisms that also contribute to C . albicans resistance to azoles [16] , [30] , [31] . As a haploid , analysis of C . glabrata genome sequence is more facile than in diploids such as C . albicans , where mitotic recombination and gene conversion can inflate the number of polymorphisms that accrue and obscure the signal of those functionally associated with drug resistance or adaptation to the host . Compared to the most widely used antifungal drugs in clinical use for the treatment of systemic infections , the azoles , mechanisms of resistance remain more limited for echinocandins . The echinocandins are the only novel class of antifungal drugs approved for clinical use in decades and target the biosynthesis of the key fungal cell wall component , 1 , 3-β-D-glucan [3] , [20] , [32] . The 1 , 3-β-D-glucan synthases are encoded by FKS1 , FKS2 , and FKS3 in S . cerevisiae , C . glabrata , and C . albicans , and require a regulatory subunit encoded by RHO1 for activity [3] , [20] , [32] , [33] . It is thought that the echinocandins bind to and inhibit the Fks protein; however , the exact mechanism of inhibition remains unknown [33] . Although the echinocandins have been in clinical use since only 2001 , there have been numerous reports of C . glabrata echinocandin resistance in patients [34]–[37] . Thus far , the only echinocandin resistance mechanism described is mutation in the drug target , Fks , particularly in highly conserved hot spots regions [33] , [36]–[39] . Such mutations can reduce echinocandin susceptibility of 1 , 3-β-D-glucan synthase by 2 to 3 log orders relative to the wild-type enzyme [36] . Additional resistance mechanisms may remain to be described given that Fks mutations have not been identified in some echinocandin-resistant isolates [20] , [37] , [40] , and that isolates with identical Fks mutations exhibit different resistance phenotypes with distinct responses to cellular perturbations [41] . Even with azoles , for which resistance mechanisms have been studied for decades , new resistance mechanisms and modulators of resistance continue to be discovered , expanding the repertoire of strategies employed by fungi to survive drug exposure to include mutation in the drug target , overexpression of multidrug-efflux transporters , and metabolic alterations that minimize drug toxicity [3] , [20] , [32] . In addition to these canonical resistance mechanisms where mutations in relevant genes confer resistance , there is also an emergent paradigm in which regulators of cellular stress responses are crucial for enabling the evolution and maintenance of drug resistance [3] , [20] , [32]; while mutations in these regulators have not been identified as a cause of resistance , stress response regulators are key resistance modulators critical for enabling the phenotypic effects of resistance acquired by diverse mechanisms . Beyond mapping mutations that confer resistance , there is pressing clinical need to elucidate strategies to block the evolution of drug resistance and abrogate resistance once it has evolved . One of the most well studied examples of a protein that governs the emergence and maintenance of fungal drug resistance is the molecular chaperone Hsp90 . Hsp90 regulates the folding and function of diverse client proteins , including many signal transducers [42] , [43] . In C . albicans , compromise of Hsp90 function reduces basal tolerance and resistance of clinical isolates to both the azoles and the echinocandins [41] , [44] , [45] . Hsp90 enables crucial responses to drug-induced stress by orchestrating signaling through the protein phosphatase calcineurin and the protein kinase C ( PKC ) cell wall integrity signaling cascade [41] , [46] . Hsp90 stabilizes the catalytic subunit of calcineurin and the terminal mitogen-activated protein kinase ( MAPK ) in the Pkc1 cell wall integrity pathway . Genetic or pharmacological compromise of Hsp90 function can enhance the efficacy of antifungals against C . albicans in multiple metazoan models of infection [41] , [45] . Notably , Hsp90's role in governing cellular responses to azoles in C . albicans is conserved in S . cerevisiae [44] , [47] . In contrast , Hsp90 and calcineurin play a key role in crucial cellular responses to echinocandins in C . albicans , but not in S . cerevisiae [41] . Whether Hsp90 influences drug resistance in C . glabrata remains entirely unknown . In C . glabrata , both calcineurin and PKC signaling have been implicated in basal tolerance to echinocandins [48] , [49] , though the role of Hsp90 remains unknown as does the impact of any of these regulators on bona fide echinocandin resistance . Here , we provide the first global analysis of mutations accompanying the evolution of fungal drug resistance in a human host . We report on a series of C . glabrata isolates that evolved echinocandin resistance in a patient undergoing treatment with the echinocandin caspofungin for recurring C . glabrata candidemia over a 10-month period . Whole genome sequencing revealed that a mutation occurred in the gene encoding the drug target , FKS2 , accompanying a major increase in resistance , as well as 8 other non-synonymous mutations in genes not previously implicated in echinocandin resistance , including CDC6 and CDC55 . The FKS2-T1987C ( S663P ) mutation was sufficient to confer echinocandin resistance in a susceptible laboratory strain; however , the mutant allele also imparted a growth defect in clinically relevant conditions using RPMI medium at 37°C . The fitness cost of resistance was mitigated with further evolution , and this trend was also observed in a murine model of disseminated infection . A CDC6-A511G ( K171E ) mutation acquired prior to the FKS2-T1987C ( S663P ) mutation was sufficient to confer a small increase in resistance . Elevated dosage of CDC55 , which acquired a C463T ( P155S ) mutation after FKS2-T1987C ( S663P ) , ameliorated the fitness cost imparted by the FKS2 mutation . To uncover mechanisms that abrogate echinocandin resistance , we turned to Hsp90 and found that genetic or pharmacological compromise of C . glabrata Hsp90 function reduced basal tolerance and resistance of clinical isolates . Compromising calcineurin function pharmacologically or genetically phenocopied compromising Hsp90 function . Caspofungin induced FKS2 expression in a manner that depended upon Hsp90 and calcineurin , providing a molecular mechanism by which Hsp90 and calcineurin regulate echinocandin resistance . Furthermore , one of the clinical isolates in the series is a petite mutant based on morphology and inability to respire; although the petite phenotype was not intrinsically involved in echinocandin resistance , it imparted resistance to the combination of echinocandins and inhibitors of Hsp90 or calcineurin . In a mouse model of candidemia , the petite mutant was rapidly cleared and completely avirulent while infection with the HSP90-repressible strain yielded a reduced fungal burden compared to wild type . Thus , our results provide the first global view of mutations that accompany the evolution of fungal drug resistance in a human host , implicate the first compensatory mutation that mitigates the cost of echinocandin resistance , and suggest a new molecular mechanism regulating echinocandin resistance , with broad therapeutic potential .
While numerous echinocandin-resistant isolates have been recovered from patients [34]–[37] , in most cases , there has not been adequate sampling over the course of drug treatment to identify related fungal lineages with which to study the evolution of drug resistance in the human host . The most detailed sampling includes a clinical isolate pre-caspofungin treatment and two isolates taken serially post-caspofungin treatment , however , the mechanism by which resistance evolved was not investigated [50] . In contrast , series of clinical isolates recovered over time from patients undergoing treatment with azoles have proven instrumental for dissecting mechanisms of azole resistance in C . albicans [44] , [51]–[55] . The more detailed sampling and analysis of the evolution of resistance to azoles likely reflect the fact that azoles have been used clinically for a longer duration . We report here on a series of C . glabrata isolates recovered over a 10-month period from a 45-year old female patient with Crohn's disease who suffered from recurrent C . glabrata candidemia and underwent multiple rounds of caspofungin treatment before ultimately succumbing to the infection . A case report on the details of the patient history and therapeutic interventions is provided in the Materials and Methods . Given that multiple blood cultures were negative prior to stopping treatment at any interval and that susceptibility testing was not routinely performed at the time of treatment , caspofungin remained the main therapeutic intervention for candidemia . The 7 blood culture isolates analyzed are labeled alphabetically in the order in which they were recovered , such that isolate A was recovered prior to caspofungin treatment and isolate G was recovered 10 months after recurrent infection and several rounds of caspofungin treatment . The isolates were determined to be related based on molecular typing analysis including pulsed-field gel electrophoresis ( PFGE ) - karyotype analysis ( Figure S1 ) , as well as restriction enzyme PFGE using SfiI ( data not shown ) . Antifungal susceptibility of the 7 isolates in the series was determined for caspofungin , as well as for numerous azoles ( fluconazole , ketoconazole , itraconazole , and voriconazole ) and for amphotericin B , which binds to ergosterol and disrupts membrane integrity [20] , using broth microdilution with RPMI 1640 and following the standard CLSI M27-A3 protocol [56] ( Table S1 ) . The major trend observed was an increase in caspofungin resistance over the course of treatment . There were only minimal changes in susceptibility to the other antifungal drugs tested , with the exception of an increase in resistance to azoles that peaked at isolate F and returned to intermediate levels at isolate G . Because the patient was not treated with azoles , the changes in azole susceptibility may be due to mutations that arose in the lineage due to genetic drift rather than selection , or may reflect additional phenotypic consequences of mutations associated with echinocandin resistance , mutations associated with adaptation to the bloodstream , or mutations that were not directly selected for but simply hitch-hiked along with mutations under selection in this predominantly clonal system . The genomic spectrum of resistance mutations that emerge during the evolution of a fungal pathogen in its human host can be addressed for the first time using next-generation sequencing technology and the series of C . glabrata clinical isolates we describe here . As a haploid , genome analysis of C . glabrata is simplified relative to diploids such as C . albicans , where mitotic recombination and gene conversion can amplify the number of polymorphisms observed between early and late isolates ( changes from heterozygous to homozygous states are also classified as nucleotide changes ) and thereby hinder functional analysis of mutations conferring resistance . Further , C . glabrata shares a recent common ancestor with the model yeast S . cerevisiae with a large number of orthologs between the two species , facilitating bioinformatic analysis [21] , [22] . Given the clinical importance of echinocandin resistance , and that to date resistance has only been attributed to mutations in the echinocandin target [33] , [36]–[39] , with evidence that additional resistance determinants and modulators may remain to be discovered [20] , [37] , [40] , we sought to identify the mutations that accompany the evolution of echinocandin resistance in the human host on a genome-wide scale . We performed whole genome sequencing of the C . glabrata isolate recovered prior to caspofungin treatment ( isolate A ) and the last isolate recovered after multiple rounds of treatment ( isolate G ) using the Illumina Genome Analyzer II platform . We obtained 5 . 1 and 3 . 8 million 76 base pair single-end reads for isolate A and isolate G , respectively , resulting in 22 to 30× genome coverage . Reads were aligned against the reference genome sequence of CBS138 [57] . Single nucleotide variants were identified using a machine learning approach . A total of 45 , 797 single nucleotide variants were identified between late clinical isolate G and CBS138 . Of these single nucleotide variants , 39 , 146 had sufficient sequencing depth in isolate A to be reliably assigned . Overall , 26 single nucleotide variants were uncovered between isolate A and G , with only 17 of these within open reading frames and only 9 resulting in non-synonymous changes ( Table 1 ) . Although the 9 mutations that were not within open reading frames ( Table 2 ) could include mutations that affect regulation of genes important for drug resistance , we focused our analysis on mutations within open reading frames given that silent mutations can more readily be distinguished from those with functional consequences in coding sequences and given their potential impact on gene product function . All 9 non-synonymous changes were verified and then mapped across isolates B to F using Sanger sequencing to determine when each mutation arose in the series ( Figure 1 ) . Genes are named based on homology to S . cerevisiae genes [57] . Mutations in the MOH1 , GPH1 , CDC6 , and TCB1/2 genes accompanied the first modest increase in resistance in the series at isolate C ( Figure 1 ) . The function of MOH1 in S . cerevisiae is largely unknown except that it is required for survival in stationary phase [58] , [59] , and it was found to genetically interact with Hsp90 in a genome-wide chemical-genetic screen [60] . Expression of C . albicans MOH1 is induced by alpha pheromone in filament-inducing Spider medium , by weak acid stress via Mnl1 , and in biofilm conditions [61]–[63] . GPH1 encodes a glycogen phosphorylase regulated by the high osmolarity glycerol ( HOG ) mitogen-activated protein ( MAP ) kinase pathway in S . cerevisiae [64] , [65] , and is induced upon fluconazole treatment in C . albicans [66] . The CDC6 gene product is involved in DNA replication initiation by forming and maintaining the pre-replicative complex and serving as a loading factor for the Mcm2–7 proteins onto chromatin [67] , [68] . TCB1 and TCB2 encode proteins containing both calcium and lipid binding domains and appear to be involved in membrane trafficking in S . cerevisiae [69] , [70] . None of these genes or proteins have been previously implicated in echinocandin resistance . Additional mutations emerged accompanying the subsequent major increase in echinocandin resistance , and in the last isolate of the series . Mutations in FKS2 , DOT6 , MRPL11 , and SUI2 accompanied the largest increase in echinocandin resistance at isolate D ( Figure 1 ) . Of these genes , only FKS2 has been implicated in echinocandin resistance . FKS2 encodes the catalytic subunit of 1 , 3-β-D-glucan synthase , the target of the echinocandins , and the Fks2 S663P mutation identified is in mutational hot spot 1 and has been found in other echinocandin-resistant C . glabrata clinical isolates [36] , [37] , [71] . Dot6 is a subunit of the RPD3L histone deacetylase complex in S . cerevisiae and is involved in both pseudohyphal morphogenesis as well as silencing at telomeres [72]–[74] . Mrpl11 is a mitochondrial protein and part of the large ribosomal subunit [75] , [76] . SUI2 plays a role in translation initiation and encodes the alpha subunit of the translation initiation factor eIF2 in S . cerevisiae [77] . Finally , a non-synonymous mutation occurred in CDC55 in the latest clinical isolate G , despite no further increase in echinocandin resistance ( Figure 1 ) . CDC55 encodes the regulatory B subunit of protein phosphatase 2A and has a number of functions , including roles in spindle assembly during meiosis , mitotic exit , pseudohyphal morphogenesis , and chromosome disjunction [78]–[81] . Thus , in addition to a mutation in the known echinocandin target , whole genome sequencing revealed the acquisition of 8 additional non-synonymous mutations in 8 genes not previously implicated in echinocandin resistance or adaptation to host conditions during the evolution of echinocandin resistance in a human host . The genome sequence analysis further confirms clonality of the lineage given the very limited number of single nucleotide variants genome-wide compared to large number observed between the late clinical isolate G and the reference genome CBS138 . Further , that each of the mutations identified persisted throughout the lineage once it emerged suggests that there may have been strong selective sweeps in the population such that polymorphisms rapidly reached near fixation . To determine which of the mutations identified by whole genome sequencing contributes to echinocandin resistance we first turned to the most likely candidate resistance gene , FKS2 . The FKS2 mutation that emerged in isolate D and was maintained throughout the rest of the series ( T1987C ) results in substitution of a serine to proline at amino acid 663 in the target of the echinocandins . This FKS2 T1987C ( S663P ) mutation has been previously associated with high levels of caspofungin resistance in C . glabrata clinical isolates [36] , [37] . In vitro biochemical studies established that this mutant Fks2 enzyme displays reduced sensitivity to inhibition by echinocandins , as indicated by a higher kinetic inhibition parameter , as well as decreased catalytic capacity , and reduced enzyme velocity , compared to the wild-type enzyme; notably the binding affinity of the mutant enzyme for echinocandins remains unchanged [36] . While there is an association of FKS mutations with resistance , and biochemical data support the resistance mechanism , it has only been conclusively demonstrated that such mutations are sufficient to confer echinocandin resistance in S . cerevisiae [82] , [83] . To test whether Fks2 S663P is sufficient to confer echinocandin resistance , we introduced the T1987C mutation into the sensitive laboratory strain BG2 using a strategy involving single-stranded DNA containing the mutation and a silent marker , followed by selection of transformants on medium containing caspofungin . No echinocandin-resistant colonies were obtained following control transformations with single-stranded DNA containing the equivalent wild-type sequence or with a water control , while many resistant colonies were obtained with the sequence containing the T1987C mutation . We assessed resistance of four sequence-verified transformants that harboured both the T1987C and silent mutation via minimum inhibitory concentration ( MIC ) assays . The transformants displayed resistance similar to clinical isolate G ( Figure 2A ) . Thus , Fks2 S663P is sufficient to confer echinocandin resistance in a susceptible laboratory strain of C . glabrata . Given that specific Fks amino acid substitutions that decrease sensitivity of the 1 , 3-β-D-glucan synthase enzyme to echinocandins also reduce the enzyme catalytic capacity , the Fks2 S663P mutation may confer a fitness cost in terms of reduced growth or viability in the absence of the drug [36] , [39] . To determine if the FKS2 mutations compromises fitness , we monitored growth kinetics of the lab strain BG2 and the progeny harbouring the Fks2 S663P substitution . In clinically relevant conditions , RPMI at 37°C , we found that strains harbouring the Fks2 S663P substitution displayed significantly reduced fitness relative to the parental wild-type laboratory strain BG2 ( P<0 . 01 , area under the curve followed by ANOVA , Bonferroni's Multiple Comparison Test , Figure 2B , left panel ) . Furthermore , we observed a significant reduction in fitness between isolate A and isolate D ( P<0 . 001 ) , which is rescued to some extent in isolate G ( P<0 . 01 , Figure 2B , right panel ) . Thus , the FKS2 mutation that emerged in clinical isolate D is sufficient to confer a high level of caspofungin resistance equivalent to that observed in the late isolate G , however , it is also associated with a cost in terms of reduced fitness in the absence of the drug . Compensatory mutations that mitigate the cost of drug resistance have been well established in bacterial systems [84]–[87] , but remain enigmatic in fungi . Given the fitness cost associated with echinocandin resistance due to mutation in the drug target observed in this study , and in others [88] , one would anticipate selection to favour the emergence of compensatory mutations that mitigate the fitness cost of the resistance mutation . The series of C . glabrata isolates studied here provide the ideal opportunity to identify compensatory mutations given that a single non-synonymous mutation occurs between isolate D and G , CDC55 C463T ( P155S ) , associated with an increase in fitness ( Figure 1 ) . The CDC55 C463T allele from late clinical isolate G was cloned into a plasmid under the control of its native promoter , as was the CDC55 allele from the early clinical isolate A , and introduced into the BG2 laboratory strain harbouring the FKS2 T1987C mutation . Monitoring growth kinetics of two independent transformants harbouring each of the CDC55 plasmids relative to two transformants with the empty vector control demonstrates that the additional copy of CDC55 ameliorates fitness ( P<0 . 01 , ANOVA , Bonferroni Multiple Comparison Test , Figure 3 ) . This suggests that perhaps the CDC55 C463T mutation might be a gain-of-function mutation , such that increased fitness could be achieved either by mutation causing increased Cdc55 activity or by increased dosage of CDC55 . These results identify the premier genetic alteration that mitigates the fitness cost of echinocandin resistance . Despite the fact that the FKS2 T1987C mutation was sufficient to impart the full level of caspofungin resistance of isolate G on an otherwise susceptible laboratory strain ( Figure 2A ) , there is evidence for additional mutations affecting resistance in the evolved lineage . The initial small increase in echinocandin resistance observed at isolate C ( Figure 1 ) occurred prior to the FKS2 mutation and thus other mutations identified at this early transition may be responsible for this increase in resistance . Notably , isolate C showed a fitness defect in the absence of drug ( Figure 2B ) , suggesting that the mutations imparting this small increase in resistance are also costly . Additional mutations that accumulated could mitigate the fitness cost of resistance mutations . To prioritize mutations for functional analysis , we addressed the prevalence of mutations in any of the 8 genes found to harbour mutations in our whole genome sequence analysis in addition to FKS2 in other C . glabrata echinocandin-resistant mutants . To do so , we obtained 10 additional unrelated C . glabrata clinical isolates harbouring the Fks2 S663P mutation [71] , and sequenced across the 8 genomic regions that were mutated in clinical isolate G via Sanger sequencing . We discovered non-synonymous changes in MOH1 in two out of the 10 clinical isolates sequenced and non-synonymous changes in CDC6 in 7 out of 10 of the clinical isolates ( Figure 4 ) . Given the prevalence of polymorphisms in CDC6 and MOH1 among these echinocandin-resistant clinical isolates , we prioritized these genes for functional analysis . We cloned plasmids with the gene sequences found in isolate A or in isolate G , and expressed these in a susceptible laboratory strain . While the MOH1-T13C ( Y5H ) allele did not confer any increase in echinocandin resistance ( data not shown ) , the CDC6-A511G ( K171E ) allele did confer a small increase in resistance ( Figure 5 ) . Thus , we establish a novel mutation in CDC6 that contributes to reduced echinocandin susceptibility in C . glabrata clinical isolates . Given the importance of echinocandins in the therapeutic front line against C . glabrata infections and the acute problem imposed by the evolution of echinocandin resistance , there is a pressing need to discover strategies to abrogate drug resistance . We focused on Hsp90 due to its critical role in enabling basal tolerance and acquired antifungal drug resistance in pathogenic fungi such as C . albicans and the most lethal mould Aspergillus fumigatus [41] , [44] , [45] , [47] . In the context of the echinocandins , C . albicans Hsp90 orchestrates crucial cellular responses to survive echinocandin exposure by stabilizing the catalytic subunit of the protein phosphatase calcineurin , while in S . cerevisiae Hsp90 and calcineurin do not modulate echinocandin susceptibility under any of the standard conditions tested [41] . To date , no studies have examined the consequences of pharmacological or genetic compromise of Hsp90 function on cellular responses to echinocandins in C . glabrata , and thus whether this pathogen shows resistance circuitry more akin to the pathogen C . albicans or its closer relative S . cerevisiae remains unknown . We first implemented a pharmacological approach to determine if inhibition of Hsp90 modulates echinocandin resistance of C . glabrata . We monitored growth across a gradient of the widely used echinocandin caspofungin relative to a drug-free growth control in the presence or absence of the two structurally unrelated Hsp90 inhibitors , geldanamycin and radicicol , that bind to the adenosine triphosphate ( ATP ) binding pocket of Hsp90 and thereby compromise ATP-dependent chaperone function [89] , [90] . In the absence of geldanamycin , there was a small increase in caspofungin resistance between isolate B and isolate C and a large increase in resistance between isolate C and isolate D ( Figures 1 and 6A ) . Pharmacological inhibition of Hsp90 with geldanamycin or radicicol decreased tolerance of the early clinical isolates A , B , and C , and reduced resistance of the late clinical isolates D , E , and G ( Figure 6A ) . Synergy between caspofungin and geldanamycin was also observed in RPMI , a medium used for clinical susceptibility testing ( Figure S2 ) . Notably , isolate F was refractory to the synergy between caspofungin and Hsp90 inhibitors , as discussed in more detail below . Next , we validated our pharmacological findings genetically . We engineered a strain of C . glabrata in which the only HSP90 allele was expressed under the control of the MET3 promoter , which is repressed in the presence of methionine and/or cysteine . We monitored the impact of a gradient of methionine concentrations on growth across a gradient of caspofungin concentrations for both isolate G and its derivative in which HSP90 expression is driven by the MET3 promoter . Methionine had no impact on caspofungin resistance of isolate G ( Figure 6B ) . In contrast , methionine reduced echinocandin resistance of the MET3p-HSP90 derivative of isolate G in a dose-dependent manner , providing genetic validation of the importance of Hsp90 for echinocandin resistance ( Figure 6B ) . The highest concentrations of methionine tested ( ≥1 µg/ml ) blocked growth of the MET3p-HSP90 strain , consistent with Hsp90's essentiality in all eukaryotes tested . Thus , targeting Hsp90 provides the first and much-needed strategy to abrogate echinocandin resistance of C . glabrata . Hsp90 enables resistance to both the azoles and echinocandins , in large part via the protein phosphatase calcineurin in C . albicans [3] , [41] , [44] , [47] . Hsp90 stabilizes the catalytic subunit of calcineurin in both S . cerevisiae and C . albicans [41] , [91] , thereby enabling calcineurin-dependent responses to drug-induced cellular stress [3] , [92] . While calcineurin has been implicated in basal tolerance to echinocandins in C . glabrata [49] , whether calcineurin affects bona fide resistance remains unknown . We used both pharmacological and genetic approaches to determine if calcineurin is a key mediator of Hsp90-dependent echinocandin resistance in C . glabrata . First , we took a pharmacological approach and assessed growth across a gradient of caspofungin concentrations in the presence or absence of two structurally unrelated calcineurin inhibitors , cyclosporin A and FK506 . Cyclosporin A and FK506 inhibit calcineurin function in different ways . Cyclosporin A binds to cyclophilin A , a peptidyl-prolyl cis-trans isomerase , and it is this drug-protein complex that inhibits calcineurin function [93] . FK506 binds a structurally unrelated peptidyl-prolyl cis-trans isomerase , FKBP12 , and this distinct drug-protein complex also inhibits calcineurin function [93] . We used concentrations of each calcineurin inhibitor that abolished echinocandin resistance in C . albicans but did not inhibit growth on their own [41] . We found that inhibition of calcineurin with cyclosporin A or FK506 reduced caspofungin resistance of the late resistant C . glabrata clinical isolate G ( Figure 7A ) . Next , we validated our pharmacological findings genetically by deletion of the regulatory subunit of calcineurin , encoded by CNB1 , which is required for calcineurin function . Loss of calcineurin function reduced resistance of clinical isolate G in three independent mutants ( Figure 7A ) . Thus , calcineurin is a key mediator of Hsp90-dependent echinocandin resistance in C . glabrata . Since the clinical isolate G harbours multiple mutations relative to its echinocandin-susceptible counterpart , we next specifically assessed the dependence of Fks2-mediated echinocandin resistance on Hsp90 and calcineurin . Pharmacological inhibition of Hsp90 with geldanamycin , or calcineurin with cyclosporin A reduced echinocandin resistance of the laboratory strain harbouring the FKS2 T1987C allele ( Figure 7B ) , confirming that both Hsp90 and calcineurin are required for echinocandin resistance acquired by mutation of the drug target and providing the first circuitry that governs echinocandin resistance in this pathogen . The specific mechanism by which calcineurin governs echinocandin resistance remains unknown in any system . In S . cerevisiae , FKS2 expression is induced during high temperature growth via calcineurin , and deletion of both FKS1 and FKS2 is synthetically lethal [94]–[97] . Based on these findings , we propose a model in which calcineurin regulates echinocandin resistance in C . glabrata by controlling expression of the resistance determinant FKS2 . Given the functional dependence of calcineurin on Hsp90 in other systems [41] , [44] , [91] , one might expect Hsp90 to also influence expression of FKS2 . According to our model , inhibition of calcineurin or Hsp90 would compromise expression of the echinocandin-resistant 1 , 3-β-D-glucan synthase and reduce resistance . Compromising calcineurin or Hsp90 function would also reduce basal tolerance of susceptible strains by reducing FKS2 expression , thereby decreasing the cellular pool of 1 , 3-β-D-glucan synthase and enhancing susceptibility to a given concentration of echinocandin . To test this model , we used quantitative RT-PCR to measure transcript levels of FKS2 , encoding the catalytic subunit of 1 , 3-β-D-glucan synthase , in the echinocandin-resistant clinical isolate G and derivatives in which we deleted the regulatory subunit of calcineurin required for its function , encoded by CNB1 , or in which we reduced HSP90 levels . Transcript levels of CNB1 , HSP90 , and FKS2 were monitored after growth in rich medium for one hour with or without caspofungin treatment . We found that caspofungin induced expression of both CNB1 ( P<0 . 001 , ANOVA , Bonferroni's Multiple Comparison Test , Figure 8A ) and FKS2 ( P<0 . 001 , Figure 8A ) . Importantly , deletion of CNB1 blocked caspofungin-induced upregulation of FKS2 ( P<0 . 001 , Figure 8A ) . To reduce HSP90 levels in the MET3p-HSP90 strain , we included methionine in the medium for both this strain and the wild-type control at a concentration that had minimal impact on growth ( 0 . 25 µg/ml ) . HSP90 levels were reduced in the MET3p-HSP90 strain relative the wild-type control ( P<0 . 001 , Figure 8B ) . Caspofungin-dependent induction of FKS2 was significantly reduced in the MET3p-HSP90 strain ( P<0 . 001 , Figure 8B ) , suggesting that Hsp90 enables FKS2 expression . Taken together , these results establish that FKS2 is induced upon echinocandin exposure , and that full induction of the resistance determinant FKS2 is dependent on both calcineurin and Hsp90 . This provides a novel mechanism by which calcineurin and Hsp90 regulate echinocandin resistance in pathogenic fungi . Respiratory deficient mutants with loss of mitochondrial function , referred to as petite mutants , are associated with azole resistance in S . cerevisiae , C . albicans , and C . glabrata [98]–[102] . To date , there have been no reports of the petite phenotype contributing to echinocandin resistance , although C . glabrata is able to produce petite mutants at high frequency in vitro as well as in vivo [98] , [101] , [103] . Isolate F in our C . glabrata series was isolated from the patient at the same time point as isolate E , however , it was morphologically distinct and thus archived separately ( Text S1 ) . When cultured on rich medium containing dextrose as the carbon source , this isolate had a reduced growth rate relative to the other isolates and produced smaller more transparent colonies ( Figure 9A ) , despite possessing the same karyotype ( Figure S1 ) and complement of polymorphisms in the nuclear genome as isolate E ( Figure 1 ) . When cultured on rich medium containing glycerol as the carbon source , which is non-fermentable , isolate F was unable to grow ( Figure 9A ) . These results suggests that isolate F is a petite mutant as it behaves morphologically as a petite and is unable to respire based on failure to grow on glycerol . Notably , isolate F had a distinct echinocandin resistance phenotype from the other resistant isolates in the series in that its resistance was maintained even in the presence of the Hsp90 inhibitors geldanamycin or radicicol ( Figure 6 ) , or the calcineurin inhibitor cyclosporin A ( Figure 9 ) . To determine if the petite phenotype is intrinsically involved in echinocandin resistance of C . glabrata clinical isolates , petite mutants were generated from clinical isolate A and their resistance profiles were tested via MIC assays ( Figure 9B ) . If the petite phenotype is sufficient to impart echinocandin resistance , then petite mutants generated from isolate A should acquire resistance , however , they do not ( Figure 9B ) . This suggests that the petite phenotype is not intrinsically involved in C . glabrata echinocandin resistance . To determine if petite mutants of C . glabrata are intrinsically refractory to the synergy between echinocandins and inhibitors of Hsp90 or calcineurin , petite mutants were generated from late clinical isolate G and their resistance profiles were tested via MIC assays ( Figure 9C ) . The petite mutants generated from isolate G are indeed resistant to the combination of caspofungin and geldanamycin or cyclosporin A , suggesting that they either no longer require calcineurin or Hsp90 for FKS2-mediated resistance , or that they are simply resistant to geldanamycin and cyclosporin A . Notably , petite mutants are known to up-regulate multidrug efflux transporters [98] , which may confer resistance to these pharmacological inhibitors by their increased efflux from the cell . To distinguish between these two possibilities , petite mutants were generated from the isolate G derivative in which calcineurin function was genetically compromised due to deletion of CNB1 . If petite mutants no longer require calcineurin for echinocandin resistance , then deletion of CNB1 should have no impact on resistance , however , this was not the case ( Figure 9D ) . Petite mutants generated from isolate G cnb1Δ mutants were no longer resistant to caspofungin , suggesting that petite mutants are able to bypass the effects of cyclosporin A , and likely geldanamycin , potentially due to up-regulation of efflux pumps . We next turned to a murine model of disseminated candidemia to evaluate fitness of the series of clinical isolates and the impact of reduction of Hsp90 levels in vivo . Mice were infected by tail vein injection , and fungal burden in the kidney and spleen was assessed 7 days post infection . Infection with the initial clinical isolate A led to kidney fungal burden greater than that observed with the reference strain CBS138 ( P<0 . 001 , Kruskal-Wallis Test , Dunn's Multiple Comparison , Figure 10A ) . There was a trend towards reduced fitness , as measured by kidney fungal burden , associated with the early acquisition of echinocandin resistance ( Figure 10A ) , consistent with the trend observed in vitro ( Figure 2B ) . Aside from the petite mutant , isolate C showed the greatest reduction in fitness of all the clinical isolates in the series relative to isolate A ( P<0 . 05 , Figure 10A ) . This fitness cost was mitigated with further evolution as shown by an increase in kidney fungal burden of isolate G compared to isolate C ( P<0 . 01 , Figure 10A ) . Kidneys recovered from mice infected with the petite mutant ( isolate F ) were completely sterile ( Figure 10A ) , consistent with the fitness deficit observed in vitro ( Figure 2B ) and findings of reduced virulence of some C . glabrata petite mutants [102] , [104] . Methionine levels in the mouse are sufficient to repress gene expression from the MET3 promoter in C . albicans leading to avirulence of conditional mutants [105] , allowing us to test fungal burden of mice infected with the MET3p-HSP90 strain . One of the two MET3p-HSP90 derivatives of isolate G showed reduced kidney fungal burden relative to isolate G ( P<0 . 05 , Figure 10A ) , consistent with the importance of Hsp90 for growth in vitro and in vivo in a C . albicans murine model of systemic infection [106] . The second MET3p-HSP90 derivative also showed a trend towards reduced kidney fungal burden relative to isolate G ( Figure 10A ) . There were no significant differences observed among any of the isolates tested in the spleen with the exception of isolate F , for which the recovered spleens were sterile ( Figure 10B ) , suggesting that most of the mutations that accumulated in the lineage have negligible impact on fitness in this environment , and that methionine levels are likely not sufficient to achieve substantial reduction of Hsp90 levels or that Hsp90 has a less important role for fungal proliferation in the spleen compared to the kidney . Thus , analysis of kidney fungal burden in a mouse model of disseminated infection reveals a trend towards reduced fitness accompanying early stages of echinocandin resistance that is ameliorated with further evolution , avirulence associated with a petite mutant , and the importance of Hsp90 for fitness in the host .
Our results provide the first global view of mutations that accompany the evolution of fungal drug resistance in a human host , implicate the premier compensatory mutation that ameliorates the fitness cost of echinocandin resistance , and suggest a new molecular mechanism regulating echinocandin resistance , with broad therapeutic potential . We report on C . glabrata bloodstream isolates that evolved increased resistance to the echinocandin caspofungin over a 10-month period during which the patient underwent multiple rounds of caspofungin treatment for recurrent candidemia ( Figure S1 and Table S1 ) . This case demonstrates that echinocandin resistance can evolve during treatment , and that an undetected nidus of infection may contribute to fungal persistence and drug resistance despite apparent adequate therapy and documentation of negative blood cultures after treatment , emphasizing the importance of routine antifungal susceptibility testing . Whole genome sequencing of the susceptible isolate recovered prior to drug treatment and the last resistant isolate revealed that 9 non-synonymous mutations accumulated during evolution in the human host ( Table 1 and Figure 1 ) . A mutation in FKS2 , encoding the drug target , accompanied the largest increase in echinocandin resistance in the lineage; this mutation was sufficient to confer echinocandin resistance in a susceptible C . glabrata strain , but was associated with a fitness cost with reduced growth rate in the absence of drug ( Figure 2 ) . The fitness cost of resistance was ameliorated with further evolution , based on observations in vitro and in a murine model of systemic infection ( Figures 2 and 10 ) . The 8 additional mutations in genes not previously implicated in echinocandin resistance ( Table 1 ) provide candidates for novel resistance determinants as well as mutations that mitigate the cost of resistance . Consistent with these possibilities , increased dosage of CDC55 , which acquired a mutation accompanying the increase in fitness , mitigated the fitness cost of the FKS2 mutation ( Figure 3 ) , while a mutation in CDC6 that arose prior to the FKS2 mutation was sufficient to confer a small increase in echinocandin resistance ( Figure 5 ) . Further , we establish that Hsp90 governs both basal tolerance to the echinocandins and bona fide resistance of clinical isolates ( Figure 6 ) , and that Hsp90 is important for C . glabrata proliferation in the mouse kidney ( Figure 10 ) . We found that calcineurin is a key mediator of Hsp90-dependent resistance ( Figures 7 ) . Hsp90 and calcineurin regulate echinocandin resistance by controlling expression of the resistance determinant FKS2 ( Figure 8 ) , providing a novel mechanism via which Hsp90 and calcineurin govern echinocandin resistance in pathogenic fungi . The whole genome sequence analysis yields powerful insights into the evolutionary dynamics of adaptation in the host , as well as novel mutations associated with resistance or with ameliorating the fitness cost of resistance . To date , mechanisms of echinocandin resistance remained restricted to mutations in the drug target . In the C . glabrata series , mutations in 4 genes not previously associated with echinocandin resistance ( MOH1 , GPH1 , CDC6 , and TCB1/2 ) accompanied an early and small increase in echinocandin resistance ( Figure 1 ) . Mutations in these genes could confer the small increase in resistance , or could create a genetic background in which the FKS2 mutation is less detrimental; in the latter case , the mutation would have to confer a fitness benefit on its own in order to be selected for in advance of the FKS2 mutation , or it could be selectively neutral and fixed by genetic drift . Given that a reduction in fitness accompanied the acquisition of these 4 mutations ( Figure 2 ) , it is likely that they contribute to resistance or were fixed by genetic drift . Consistent with the former possibility , polymorphisms in one of the genes that acquired a mutation associated with the small increase in resistance , CDC6 , were common in an unrelated set of echinocandin-resistant C . glabrata isolates with Fks2 S663P ( Figure 4 ) , and the CDC6-A511G ( K171E ) mutation identified by whole genome sequencing was sufficient to confer a small increase in resistance ( Figure 5 ) . Mutations in 3 additional genes not previously implicated in echinocandin resistance ( DOT6 , MRPL11 , and SUI2 ) coincided with the FKS2 mutation , and a last mutation in CDC55 arose in the last isolate , without any associated change in resistance ( Figure 1 ) . Given that the FKS2 mutation is sufficient for the full resistance phenotype of the late clinical isolate ( Figure 7 ) , it is likely that these other mutations are unrelated to echinocandin resistance or that they mitigate the fitness cost of the FKS2 mutation . It is notable that multiple mutations accumulated at the two major transitions in resistance ( Figure 1 ) . This is consistent with strong selection favouring the rapid accumulation of mutations in the lineage . That each of the mutations identified persisted throughout the lineage is consistent with the occurrence of selective sweeps , such that each mutation rose to near fixation . Selective sweeps in response to drug selection are also observed in experimental populations of C . albicans during the evolution of azole resistance in vitro [107] . The fate of drug-resistant mutants in nature depends on their fitness relative to drug-susceptible counterparts . While resistance mutations are expected to confer a fitness benefit in the presence of the drug , they may also confer a cost in terms of reduced fitness in the absence of the drug . This model is consistent with the impact of the FKS2 T1987C mutation observed in the C . glabrata lineage , and reported for target-mediated echinocandin resistance in C . albicans [88] . This mutation confers a major increase in growth in the presence of echinocandin ( Figure 2A ) , but also confers reduced growth in the absence of the drug ( Figure 2B ) . The deleterious impact on fitness is likely due to the reduced catalytic capacity commonly observed among 1 , 3-β-D-glucan synthase enzymes that acquire amino acid substitutions that reduce their sensitivity to echinocandins [36] . C . glabrata may upregulate FKS2 expression to compensate for its decreased catalytic capacity [36] , or may acquire additional mutations that mitigate the cost of the resistance mutation . The fitness effects of antibiotic resistance mutations have been studied extensively in bacteria , where most resistance mechanisms are associated with a fitness cost that manifests in reduced growth rate [84] . In the vast majority of cases , the fitness cost is mitigated by the acquisition of compensatory mutations [84]–[87] . Consistent with these patterns , any cost of resistance in experimental populations of C . albicans that evolved azole resistance in vitro was mitigated was further evolution [108] , as many changes in gene expression observed in the less fit , resistant population were restored to the ancestral state [15] . In the C . glabrata lineage studied here , a mutation in CDC55 accompanied an increase in fitness of isolate G relative to isolate D ( Figures 1 and 2 ) . Increased dosage of CDC55 ameliorated fitness of an independent FKS2 T1987C mutant , independent of whether it was the C463T ( P155S ) allele ( Figure 3 ) . This suggests that the C463T ( P155S ) mutation may confer increased Cdc55 function and that fitness can be ameliorated by either elevated activity or dosage of Cdc55 . This C463T ( P155S ) mutation was not identified in other C . glabrata echinocandin-resistant FKS2 T1987C mutants ( Figure 4 ) , suggesting that distinct compensatory mutations might be favoured in vivo or that they may harbor duplications of CDC55 . The mechanisms by which alterations in CDC55 ameliorates fitness of the FKS2 mutant and the scope of beneficial effects in other backgrounds remains to be determined . Morphological variants that emerge in an evolutionary lineage can reveal important features of mechanisms of drug resistance or drug synergy . Isolate F is a petite mutant based on morphology and inability to grow on a non-fermentable carbon source ( Figure 9A ) . Such respiratory deficient mutants with loss of mitochondrial function are associated with azole resistance in S . cerevisiae , C . albicans , and C . glabrata [98]–[102] . The azole resistance of petites is attributed to increased expression of multidrug transporters of the ATP binding cassette family [98] , [99] , [101] . The petite phenotype has not been linked to echinocandin resistance to date , consistent with the limited evidence that multidrug transporters are involved in resistance to this drug class [109]–[112] . Indeed , induction of petite mutants in the early C . glabrata clinical isolate A does not confer echinocandin resistance , confirming that petite mutants are not intrinsically resistant to echinocandins ( Figure 9B ) . In S . cerevisiae , several mitochondrial proteins have been identified as required for echinocandin tolerance [113] , consistent with the slight reduction in tolerance we observe in C . glabrata petite mutant derivatives of isolate A ( Figure 9 ) . A striking feature of the isolate F is that its echinocandin resistance phenotype is recalcitrant to the impact of the Hsp90 inhibitor geldanamycin or calcineurin inhibitor cyclosporin A , unlike that of all other isolates in the series ( Figure 9C ) . Induction of petite mutants in late clinical isolate G confirms that the petite phenotype is intrinsically recalcitrant to the impact of geldanamycin or cyclosporin A ( Figure 9C ) . Genetic compromise of calcineurin function in petite mutants abrogates echinocandin resistance , suggesting that petites are simply resistant to cyclosporin A ( Figure 9D ) , and likely geldanamycin; this may be attributable to overexpression of multidrug transporters in petites that remove these inhibitors from the cell . Whether the original isolate F petite mutant arose during evolution in the human host or during sampling remains unknown , although our finding that kidneys and spleens from mice infected with the petite mutant were completely sterile ( Figure 10 ) , suggests that petite mutant arose shortly before or during its isolation . In contrast , one C . glabrata petite mutant was reported to have enhanced virulence relative to an isolate recovered from the same patient at an earlier time point , however , an isogenic control was lacking [102] . Consistent with our findings , most C . glabrata petite mutants have been reported to have attenuated virulence relative to isogenic controls [104] . Our findings of reduced kidney fungal burden in mice infected with the late clinical isolate G with conditional expression of HSP90 driven by the MET3 promoter support a role for Hsp90 in proliferation in the host ( Figure 10 ) . That there was only a modest reduction in fungal burden is surprising in light of Hsp90's essentiality in vitro , as observed by methionine-mediated transcriptional repression ( Figure 6 and 8 ) . It is likely that methionine levels in the mouse were not sufficient to repress the C . glabrata MET3 promoter and deplete Hsp90 , and that methionine levels were even lower in the spleen , leading to no significant differences in fungal burden ( Figure 10 ) . Notably , methionine levels in the mouse were sufficient to cause avirulence of an ino1Δ/ino1Δ itr1Δ/MET3p-ITR1 conditional mutant of C . albicans [105] , but only partial attenuation of virulence of a C . albicans conditional mutant of the essential gene FBR1 ( fbr1Δ/MET3p-FBR1 ) [114] . The impact on virulence using the MET3 system to deplete essential genes may depend on the level of depletion required to observe phenotypic effects . Using a tetracycline-repressible promoter system and delivery of tetracycline to the mice in a systemic model of infection , Hsp90 has been shown to be required for proliferation of C . albicans [106] , suggesting that this promoter system may be more suitable for in vivo studies given that doses of tetracycline can be titrated . Our results further establish Hsp90 and calcineurin as the first regulators of bona fide echinocandin resistance in C . glabrata , and reveal that resistance circuitry has been rewired over evolutionary time . The molecular chaperone Hsp90 and its client protein calcineurin govern basal tolerance and resistance to both the azoles and the echinocandins in C . albicans [41] , [44] , [47] . Pharmacological inhibition of Hsp90 can enhance the efficacy of azoles against C . glabrata [115] , and calcineurin plays an important role in both azole and echinocandin tolerance [49] . The roles of Hsp90 and calcineurin in azole resistance are conserved in S . cerevisiae [44] , [47] . However , compromise of Hsp90 or calcineurin function does not alter echinocandin susceptibility in S . cerevisiae under any of the standard conditions tested where echinocandin susceptibility of C . albicans is affected [41] . Here , we find that Hsp90 and calcineurin are required for basal tolerance to echinocandins in C . glabrata as well as for resistance that evolves in a human host ( Figures 4 and 5 ) , suggesting that despite the closer evolutionary relationship of C . glabrata to S . cerevisiae , the C . glabrata cellular circuitry governing resistance to drugs that target the cell wall shares more similarity to that of its more distant pathogenic relative , C . albicans . While conditions may exist in which Hsp90 and calcineurin influence echinocandin susceptibility in S . cerevisiae , it is clear that the C . glabrata phenotypic response more closely resembles that of C . albicans than S . cerevisiae . Notably , signaling pathways governing cell wall integrity have been rewired between C . albicans and S . cerevisiae [116] . The cell wall is essential for fungal viability and is an elaborate structure , components of which are recognized by vigilant immune cells in the human host [117] . As commensals and opportunistic pathogens , C . glabrata and C . albicans are likely to harbour circuitry governing cell wall architecture that is subject to strong selection in response to host immune system challenge . This work establishes that targeting Hsp90 or calcineurin has broad therapeutic potential for infections caused by one of the leading fungal pathogens of humans , and complements the expanding repertoire of therapeutic applications for inhibitors of Hsp90 and calcineurin in the treatment of infectious disease . Inhibition of Hsp90 or calcineurin transforms echinocandins from ineffective to highly efficacious against echinocandin-resistant C . glabrata ( Figures 6 and 7 ) . Notably , a human recombinant antibody against Hsp90 also has synergistic activity with echinocandins against C . glabrata in a mouse model [118] , though the mechanism by which this antibody works remains entirely unknown as it is unlikely to enter intact fungal cells to influence function of the cytosolic Hsp90 chaperone . Genetic compromise of Hsp90 function enhances the efficacy of azoles and echinocandins in a mouse model of systemic C . albicans infection [41] , [45] . Genetic or pharmacological compromise of Hsp90 also transforms fluconazole from ineffective to highly efficacious against C . albicans biofilms in a mammalian model of biofilm infection [119] . Beyond Candida species , inhibition of Hsp90 also enhances antifungal efficacy against the most lethal mould , A . fumigatus , in biofilms and in a metazoan model of infection [45] , [119] . Consistent with the functional relationship between Hsp90 and calcineurin , calcineurin inhibitors also have therapeutic potential and are synergistic with azoles against C . albicans endocarditis , keratitis , and biofilms in mammalian models [120]–[122] . Beyond their utility in the treatment of fungal infections , Hsp90 and calcineurin are promising targets for treating infections caused by protozoan parasites including Plasmodium falciparum , Trypanosoma evansi , and Leishmania major [123]–[126] . Supporting their clinical relevance , Hsp90 inhibitors have advanced in clinical trials for the treatment of cancer and other diseases [9] , [127] , [128] , and calcineurin inhibitors are widely used in the clinic as immunosuppressants [92] . Given the potential for toxicity upon inhibition of key cellular regulators in the host during infection [45] , the challenge for further development of Hsp90 and calcineurin as therapeutic targets for infectious disease lies in developing pathogen-selective inhibitors or drugs that target pathogen-specific components of the cellular circuitry governing drug resistance and virulence .
Dr . Susan M . Poutanen discussed this study and specifically highlighted the inclusion of the case history and the case isolates in this study with Nushrat Sultana , Research Ethics Board Coordinator for the Research Ethics Board at Mount Sinai Hospital in Toronto , Canada , the hospital in which the case patient had been hospitalized . Nushrat Sultana confirmed that completing and publishing a case study that involves only a single case does not require review by the Research Ethics Board . Written confirmation has been provided by Dr . Ronald Heslegrave , Chair , Research Ethics Board , Mount Sinai Hospital . Oral and written consent was provided by the case patient's mother , as the patient is deceased . Animals studies conducted in the Division of Laboratory Animal Resources ( DLAR ) facilities at Duke University Medical Center ( DUMC ) were handled with good practice as defined by the United States Animal Welfare Act and in full compliance with the guidelines of the DUMC Institutional Animal Care and Use Committee ( IACUC ) . The murine systemic infection model was reviewed and approved by the DUMC IACUC under protocol number A238-09-08 . A 43 year-old female had four episodes of Candida glabrata candidemia between April 2004 and October 2005 . Her past medical history was significant for severe fistulizing Crohn's disease diagnosed at the age of 9 years . She had been on total parenteral nutrition ( TPN ) since 2003 for short gut syndrome due to numerous small bowel resections over the 30-year course of her disease . Archives of C . glabrata strains were maintained at −80°C in 25% glycerol . Strains were grown in either YPD ( 1% yeast extract , 2% bactopeptone , 2% gucose ) , YPG ( 1% yeast extract , 2% bactopeptone , 2% glycerol ) , synthetic defined medium ( 0 . 67% yeast nitrogen base , 2% glucose ) supplemented with required amino acids , or RPMI medium 1640 ( Gibco , 3 . 5% MOPS , 2% glucose , pH 7 . 0 ) . 2% agar was added for solid media . Strains used in this study are listed in Table S2 . Strain construction is described in Text S1 . Recombinant DNA procedures were performed according to standard protocols . Plasmids used in this study are listed in Table S3 . Plasmid construction is described in the Text S1 . Plasmids were sequenced to verify the absence of any nonsynonymous mutations . Primers used in this study are listed in Table S4 . PFGE-karyotyping and restriction enzyme-PFGE using SfiI were performed following previously reported methods [129] . Genomic DNA was extracted from clinical isolate A and G , and sequencing libraries were prepared using the Illumina genomic DNA library preparation kit according to the manufacturers recommendations ( Illumina , CA ) with several modifications . In brief , DNA was sheared by sonication to an average fragment length of 200 base pairs . Illumina adapters were blunt-end ligated and libraries were amplified by PCR and purified using Ampure ( Agencourt ) beads at a DNA:bead ratio of 1:0 . 9 . Each sample was sequenced together in a single lane on an Illumina Genome Analyzer II platform , yielding 5 . 1 and 3 . 8 million 76 base pair single-end reads for isolate A and isolate G , respectively , resulting in 22 to 30× genome coverage . Reads were aligned using SOAP2 ( PMID: 19497933 ) against the reference genome sequence of CBS138 [57] . Single nucleotide variants were identified using a machine learning approach as described previously [130] . All non-synonymous mutations identified were validated independently using Sanger sequencing . Raw data is available for download from the Short Read Archive under accession SRA047280 . 2 . The 9 non-synonymous mutations identified by whole genome sequencing were mapped across clinical isolates B , C , D , E , and F using Sanger sequencing . CgFKS2 was amplified using oLC1344/1345 , CgDOT6 with oLC1559/1560 , CgMOH1 with oLC1561/1562 , CgGPH1 with oLC1563/1564 , CgMRPL11 with oLC1565/1566 , CgCDC6 with oLC1567/1568 , CgCDC55 with oLC1569/1570 , CgSUI2 with oLC1571/1572 , and CgTCB1/2 with oLC1573/1574 . CgFKS2 was sequenced with oLC1344 , CgDOT6 with oLC1559 , CgMOH1 with oLC1561 , CgGPH1 with oLC1563 , CgMRPL11 with oLC1565 , CgCDC6 with oLC1567 , CgCDC55 with oLC1569 , CgSUI2 with oLC1571 , and CgTCB1/2 with oLC1573 . Antifungal susceptibility was determined in flat bottom , 96-well microtiter plates ( Sarstedt ) using a modified broth microdilution protocol , as described [44] . Minimum inhibitor concentration ( MIC ) tests were set up in a total volume of 0 . 2 ml/well with 2-fold dilutions of caspofungin ( CF , generously provided by Rochelle Bagatell ) . Echinocandin gradients were from 16 µg/ml down to 0 with the following concentration steps in µg/ml: 16 , 8 , 4 , 2 , 1 , 0 . 5 , 0 . 25 , 0 . 125 , 0 . 0625 , 0 . 03125 , 0 . 015625 , and 0 . Cell densities of overnight cultures were determined and dilutions were prepared such that ∼103 cells were inoculated into each well . Geldanamycin ( GdA , A . G . Scientific , Inc . ) and radicicol ( RAD , A . G . Scientific , Inc . ) were used to inhibit Hsp90 at the indicated concentrations , and cyclosporin A ( CsA , CalBiochem ) and FK506 ( A . G . Scientific , Inc . ) were used to inhibit calcineurin at the indicated concentrations . Dimethyl sulfoxide ( DMSO , Sigma Aldrich Co . ) was the vehicle for GdA , RAD , CsA , and FK506 . Sterile water was the vehicle for CF . Plates were incubated in the dark at 30°C for the time period indicated , at which point plates were sealed and re-suspended by agitation . Absorbance was determined at 600 nm using a spectrophotometer ( Molecular Devices ) and was corrected for background from the corresponding medium . Each strain was tested in duplicate on at least two occasions . MIC data was quantitatively displayed with colour using the program Java TreeView 1 . 1 . 3 ( http://jtreeview . sourceforge . net ) . Clinical antifungal MICs were determined using broth microdilution with RPMI 1640 broth for amphotericin , fluconazole , ketoconazole , itraconazole , voriconazole , and caspofungin following Clinical and Laboratory Standards Institute document M27-A3 [56] . Visual MIC endpoints were read after 24 hours of incubation at 35°C for caspofungin and after 48 hours of incubation for all other drugs . Complete inhibition was used to determine amphotericin endpoints; 50% inhibition ( compared to growth control ) was used for caspofungin and 80% inhibition was used for the other drugs . To measure gene expression changes in response to caspofungin treatment in C . glabrata , cells were grown overnight in YPD at 30°C . Cells were diluted to OD600 of 0 . 2 in SD and grown for 2 hours in duplicate for each strain at 25°C . After 2 hours of growth 120 ng/ml CF was added to one of the two duplicate cultures and left to grow for one additional hour at 25°C . Cells were centrifuged and pellets were frozen at −80°C immediately . RNA was isolated using the QIAGEN RNeasy kit and RNAse-free DNase ( QIAGEN ) , and cDNA synthesis was performed using the AffinityScript cDNA synthesis kit ( Stratagene ) . PCR was performed using SYBR Green JumpStart Taq ReadyMix ( Sigma-Aldrich Co ) with the following cycling conditions: 94°C for 2 minutes , 94°C for 15 seconds , 60°C for 1 minute , 72°C for 1 minute , for 40 cycles . All reactions were performed in triplicate , using primers for the following genes: CgACT1 ( oLC1500/1501 ) , CgCNB1 ( oLC1502/1503 ) , and CgFKS2 ( oLC1498/1499 ) . Data were analyzed using iQ5 Optical System Software Version 2 . 0 ( Bio-Rad Laboratories , Inc ) . Statistical significance was evaluated using GraphPad Prism 5 . 0 . To monitor gene expression changes in response to reduction of HSP90 levels , strains CgLC751 and CgLC2121 were grown overnight at 30°C in synthetic defined medium supplemented with 0 . 25 µg/ml methionine . Stationary phase cultures were diluted to an OD600 of 0 . 2 and grown for 2 hours at 30°C in synthetic defined medium with methionine . Following incubation , cultures were split and 120 ng/ml caspofungin added to one set . Cells were harvested after one hour and RNA was isolated using the QIAGEN RNeasy kit and cDNA synthesis was performed using the AffinityScript cDNA synthesis kit ( Stratagene ) . PCR was carried out using the SYBR Green Fast Mix ( Applied Biosystems ) with the following cycle conditions: 95°C for 20 seconds , and 95°C for 3 seconds , 60°C for 30 seconds , for 40 cycles . All reactions were done in triplicate using the following primer pairs: CgACT1 ( oLC1500/1501 ) , CgHSP90 ( oLC2155/2156 ) , CgFKS2 ( oLC1498/1499 ) . Data were analyzed in the StepOne analysis software ( Applied Biosystems ) . Growth kinetics were measured in C . glabrata strains by inoculating cells from an overnight culture grown in YPD at 30°C to an OD600 of 0 . 0625 in 100 µl of RPMI with 2% glucose in flat bottom , 96-well microtiter plates ( Sarstedt ) . Cells were grown in a Tecan GENios microplate reader ( Tecan Systems Inc . , San Jose , USA ) at 37°C with orbital shaking . Optical density measurements ( OD600 ) were taken every 15 minutes for 48 hours . Statistical significance was evaluated using GraphPad Prism 4 . 0 . For assays involving plasmids , selection was maintained with 150 µg/ml nourseothricin ( Werner BioAgents ) . C . glabrata strains were inoculated from solid YPD medium to liquid YPD medium containing 10 µg/ml ethidium bromide ( EtBr ) . The culture was grown overnight , shaking at 30°C in the dark . Approximately 100 cells were plated on YPD agar to isolate single colonies . After 2 days of incubation at 30°C , single colonies were tested for growth on YPD agar and YP-glycerol agar . Colonies able to grow on glucose as the sole carbon source but not on glycerol as the sole carbon source were selected . Four- to five-week-old male CD1 mice from The Jackson Laboratory ( n = 10 for each group ) were utilized in this study . C . glabrata strains were grown in 10 ml liquid methionine-free minimal medium ( 6 . 7 g yeast nitrogen base without amino acids and 20 g glucose in 1 liter ) overnight at 30°C . Cultures were washed twice with 10 ml of phosphate buffered saline ( PBS ) , and the cells were then resuspended in 2 ml of PBS . Cells were counted with a hemocytometer and resuspended in an appropriate amount of PBS to obtain an infection inocula concentration of 1×108 cells/ml . Two hundred microliters ( 2×107 cells ) were used to infect mice by lateral tail vein injection . Appropriate dilutions of the cells were plated onto solid methionine-free minimal medium and incubated at room temperature for 48–96 hours to confirm cell viability . C . glabrata infected mice were sacrificed and dissected on day 7 post-infection . The kidney and spleen tissues were removed , weighed , transferred to a 15 ml Falcon tube filled with 5 ml PBS , and homogenized for 10 seconds at 17 , 500 rpm ( Power Gen 500 , Fisher Scientific ) . Tissue homogenates were serially diluted , and 100 µl was plated onto YPD solid medium ( except CgLC2121 and CgLC2122 strains which were plated onto methionine-free minimal medium containing 100 µg/ml chloramphenicol ) . The plates were incubated at room temperature for 48–96 hours to determine CFUs per gram of organ . We confirmed that organ-recovered CgLC2121 and CgLC2122 cells only grew on methionine-free minimal medium but not on medium containing methionine . All experimental procedures were carried out according to NIH guidelines and Duke IACUC protocols for the ethical treatment of animals . C . glabrata: ACT1 ( 2890423 ) ; CDC55 ( 2890752 ) ; CDC6 ( 2890231 ) ; CNB1 ( 2890566 ) ; DOT6 ( 2886442 ) ; FKS1 ( 2888318 ) ; FKS2 ( 2890040 ) ; FKS3 ( 2891236 ) ; GPH1 ( 2887861 ) ; HSP90 ( 2891108 ) ; MET3 ( 2886500 ) ; MOH1 ( 2887590 ) ; MRPL11 ( 2889720 ) ; RHO1 ( 2888920 ) ; SUI2 ( 2886561 ) ; TCB1/2 ( 2889714 ) . S . cerevisiae: CDC55 ( 852685 ) ; DOT6 ( 856822 ) ; FKS1 ( 851055 ) ; FKS2 ( 852920 ) ; FKS3 ( 855353 ) ; GPH1 ( 856289 ) ; MOH1 ( 852231 ) ; MRPL11 ( 851325 ) ; RHO1 ( 856294 ) ; SUI2 ( 853463 ) ; TCB1 ( 854253 ) ; TCB2 ( 855637 ) .
|
The evolution of drug resistance poses a severe threat to human health . Candida glabrata is a leading cause of mortality due to fungal infections worldwide . It can rapidly evolve resistance to drugs such as echinocandins , which target the fungal cell wall and are front-line therapeutics for Candida infections . We harness whole genome sequencing to provide a global view of mutations that accumulate in C . glabrata during the evolution of echinocandin resistance in a human host . Nine non-synonymous mutations were identified , including one in the echinocandin target . A mutation in an additional gene conferred a small resistance increase and another was in a gene whose dosage mitigated the fitness cost of resistance . We further discovered that compromising function of the molecular chaperone Hsp90 abrogates drug resistance and reduces kidney fungal burden in a mouse model of infection . Hsp90 and its downstream effector calcineurin are required for induction of the drug target in response to drug . Thus , we reveal the first global portrait of antifungal resistance mutations that evolve in a human host , identify the first compensatory mutation that mitigates the cost of echinocandin resistance , and suggest a new mechanism of echinocandin resistance that can be exploited to treat life-threatening fungal infections .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genome",
"sequencing",
"mycology",
"genomics",
"fungal",
"evolution",
"microbial",
"evolution",
"microbial",
"pathogens",
"biology",
"evolutionary",
"biology",
"genomic",
"evolution",
"microbiology",
"host-pathogen",
"interaction",
"genetics",
"and",
"genomics"
] |
2012
|
Global Analysis of the Evolution and Mechanism of Echinocandin Resistance in Candida glabrata
|
Differences in the level , timing , or location of gene expression can contribute to alternative phenotypes at the molecular and organismal level . Understanding the origins of expression differences is complicated by the fact that organismal morphology and gene regulatory networks could potentially vary even between closely related species . To assess the scope of such changes , we used high-resolution imaging methods to measure mRNA expression in blastoderm embryos of Drosophila yakuba and Drosophila pseudoobscura and assembled these data into cellular resolution atlases , where expression levels for 13 genes in the segmentation network are averaged into species-specific , cellular resolution morphological frameworks . We demonstrate that the blastoderm embryos of these species differ in their morphology in terms of size , shape , and number of nuclei . We present an approach to compare cellular gene expression patterns between species , while accounting for varying embryo morphology , and apply it to our data and an equivalent dataset for Drosophila melanogaster . Our analysis reveals that all individual genes differ quantitatively in their spatio-temporal expression patterns between these species , primarily in terms of their relative position and dynamics . Despite many small quantitative differences , cellular gene expression profiles for the whole set of genes examined are largely similar . This suggests that cell types at this stage of development are conserved , though they can differ in their relative position by up to 3–4 cell widths and in their relative proportion between species by as much as 5-fold . Quantitative differences in the dynamics and relative level of a subset of genes between corresponding cell types may reflect altered regulatory functions between species . Our results emphasize that transcriptional networks can diverge over short evolutionary timescales and that even small changes can lead to distinct output in terms of the placement and number of equivalent cells .
Transcriptional programs specify and elaborate cell identity during animal development , as a single cell gives rise to the hundreds of cell types that comprise the adult animal . Accordingly , variation in the timing , spatial location , and level of transcription is thought to be a major source of molecular variation for morphological changes during evolution [1]–[3] . Gene expression during animal development is highly dynamic in space and time and occurs in the context of a gene regulatory network; the expression of any given gene is dependent on the spatiotemporal expression patterns of many others . This poses a fundamental problem for comparing gene expression patterns between species . Any measured expression differences for a given gene could be due to multiple non-mutually exclusive factors including changes in embryo geometry , changes in the activity , timing or location of expression for upstream regulators or altered regulatory logic , such as no longer responding to a particular regulator . Attributing gene expression differences to these different sources is a fundamental hurdle to employing a comparative approach; overcoming it would allow new types of systematic analyses to address how changes to gene regulation can contribute to new organismal phenotypes . What strategy can we use to disentangle the potential sources of expression differences ? One possibility is to look specifically for regulatory differences in a way that controls for differences in embryo morphology . In the developing embryo , each nucleus must make a decision about whether to express a gene , and to what level . This decision is based on integrating local information about the concentration of upstream regulators , usually DNA binding proteins termed transcription factors ( TFs ) . This regulatory function , termed the input function [4] or gene regulatory function ( GRF ) [5] , relates the concentration of regulators ( inputs ) to the concentration of their targets ( the outputs ) . If an input function is conserved , we expect to find cells in multiple species with similar concentrations of inputs and outputs , even if they occur in different positions in the respective embryos . We can therefore attempt to identify conserved input functions by identifying cells with similar multi-gene expression profiles . This strategy also provides an embryo-scale view on the output of the gene regulatory network , namely sets of cells distinguished by their transcriptional profile and therefore primed to differentiate into different cell types . Analyzing the set of expression profiles for all cells in the embryo thus reveals how the patterning system allocates cells to different cell types during development . Comparing cellular gene expression profiles between species requires high-resolution data: specifically , expression measurements for an entire gene regulatory network at cellular resolution in multiple species . Imaging technology now makes it possible to collect quantitative spatiotemporal expression data at cellular resolution for several genes at once . Previously , we developed high-resolution microscopy and image analysis methods to measure gene expression quantitatively in blastoderm embryos of Drosophila melanogaster ( D . melanogaster ) , in roughly 10 minute time intervals during the hour prior to gastrulation [6] , [7] . These data are integrated into a gene expression atlas that presents the average expression of many genes in a unified cellular resolution morphological framework . In contrast , most previous comparative gene expression studies in Drosophila have either sacrificed spatial information to obtain quantitative data on many genes using genomic technologies such as arrays or RNA-seq , or used imaging to obtain qualitative spatially resolved data on few genes ( for examples in early development of Drosophila see [8]–[14] ) . Here , we apply our high-resolution imaging methods to measure gene expression patterns for 13 genes from the segmentation network in blastoderm embryos of two closely related species of Drosophila , Drosophila yakuba ( D . yakuba ) and Drosophila pseudoobscura ( D . pseudoobscura ) , and compare our data to a similar pre-existing dataset for D . melanogaster . The segmentation network ( Figure 1 ) comprises a small number of well-characterized TFs that interact to generate increasingly complex patterns of gene expression during a short window of early development . The output of the network prefigures the position of the larval segments and associated morphological structures [15] . The TFs and topology of the segmentation network are assumed to be conserved throughout Drosophila , but vary in higher Diptera and other insects including wasps , beetles , mosquitos and bees [13] , [16]–[23] . We therefore anticipated that the network output in three closely related Drosophila species would be at least qualitatively similar , but we could not predict a priori what type of quantitative differences we would find . We report our findings on quantitative differences in embryo morphology and expression patterns between D . melanogaster , D . yakuba and D . pseudoobscura , our method for comparing cellular gene expression profiles while accounting for changing embryo morphology , and our comparative analysis of cell types at the blastoderm stage of development .
We used high-resolution microscopy and RNA in situ hybridization to image the expression of 13 genes in 616 embryos of D . yakuba and 933 embryos of D . pseudoobscura ( Table S1 ) . This set of 13 genes consists of major determinants of anterior/posterior patterning including the maternal genes bicoid ( bcd ) and caudal ( cad ) ; the gap genes giant ( gt ) , Krüppel ( Kr ) , knirps ( kni ) and hunchback ( hb ) ; the terminal genes forkead ( fkh ) , huckebein ( hkb ) and tailless ( tll ) ; three primary pair-rule genes , even-skipped ( eve ) , fushi-tarazu ( ftz ) , and odd-skipped ( odd ) ; and one secondary pair-rule gene , paired ( prd ) [24] . Staining for cad in D . pseudoobscura was consistently low-level and uniform , and is therefore not included in the dataset . While antibodies are available for some of the components of the network , they vary widely in quality and work with different efficiencies in different species . Where protein levels have been measured in D . melanogaster , they correlate well with RNA levels except in notable cases , such as hunchback , where translational control is known to play a role [6] . Each embryo was stained for the gene of interest , a DNA dye and a second gene serving as a fiduciary marker . Embryos were manually staged into 6 time intervals spanning the hour prior to gastrulation by assessing the extent of cell membrane invagination under phase contrast illumination . Embryos were then imaged using 2-photon microscopy , and the resulting image stacks were segmented to generate individual pointcloud files , which record the 3D location and gene expression values associated with each nucleus [6] . Pointcloud files for individual embryos were registered together to produce gene expression atlases for D . yakuba and D . pseudoobscura . In these atlases , average expression for all of the genes in our dataset are present in a species-specific dynamic morphological framework based on cellular density patterns . Expression levels within an atlas are normalized per gene with expression levels scaled so that the time point with the highest expression value takes on a value of 1 . For a detailed description of atlas building methods , see [7] . Average patterns for each gene for the six time intervals in our dataset are shown in Figure 2 alongside the corresponding genes in the reference D . melanogaster dataset [7] . We assessed the quality of the data by two measures , the range of intensities measured for a given gene , which reflects the ratio of signal to noise ( Figure S1 ) , and the average standard deviation in expression after registration ( Table S2 ) . The atlases for D . yakuba and D . pseudoobscura are of similar quality to the previously assembled D . melanogaster dataset . Though qualitatively similar , our data revealed several quantitative morphological differences between D . melanogaster , D . yakuba and D . pseudoobscura embryos including differences in blastoderm shape , size and the number of nuclei ( Figure 3 ) . These differences required us to build species-specific atlases to account for the different embryo morphologies , rather than register all data into a single morphological framework . Comparison of the eggs of the three species revealed that they vary both in their anterior/posterior shapes ( compare D . yakuba to D . melanogaster and D . pseudoobscura , Figure 3a ) , and their circumferences ( compare D . yakuba and D . melanogaster to D . pseudoobscura , Figure 3a ) . Ordering the embryos in terms of average egg length or surface area , D . pseudoobscura embryos are the smallest , followed by embryos of D . melanogaster and D . yakuba ( Figure 3b , Table 1 ) . Notably , the number of nuclei scales linearly with surface area within each species with the same relationship ( slope ) ( Figure 3b ) . However , this doesn't completely explain changes in nuclear number between species , as even some embryos with the same surface area have different numbers of nuclei ( note in particular differences between D . pseudoobscura and D . melanogaster embryos ) . Nuclear density patterns prefigure movements during gastrulation [25] , [26] . We found that the spatial patterns of local nuclear density are similar between the three species ( Figure S2 ) , though the average density of nuclei on the surface of D . yakuba embryos is lower than that for the other two species ( Figure 3c ) . The overall similarities between the species' nuclear density patterns , including lower density around the cephalic furrow and along the ventral midline , indicate that nuclear density patterns likely reflect conserved developmental processes . During cellularization , nuclei move from the poles towards the center and this can contribute to shifts in gene expression patterns . We call this “cell flow” to distinguish it from “expression flow” [26] . The overall direction and magnitude of cell flow movements are similar between these 3 species ( Figure S3 ) . A species-specific model of cell flow based on changing density patterns is used to find corresponding cells across time points during atlas construction . As a result , comparison of cellular gene expression profiles over time between the expression atlases removes the effect of differences in cell flow [7] , [27] . To systematically analyze expression differences in this transcriptional network , we developed a method to compare gene expression profiles on a cell-by-cell basis . Each cell's gene expression profile can be represented as a vector whose entries are defined by the average expression level for a given gene at a given time point . We used the squared Euclidean distance between such vectors to score the difference between any two cells; we call this the expression distance score . We used the squared distance ( rather than the Euclidian distance ) because it is additive across genes and time points which makes interpretation of the contributions of each gene to the overall expression distance simple to interpret . The expression distance score can be calculated based on any subset of genes in the dataset including single genes , groups of specific interacting genes , or the entire dataset simultaneously . These analyses are possible because our dataset contains expression levels measured for multiple genes in the same cellular resolution framework . Comparing gene expression in this way has several advantages over standard methods where gene expression patterns are compared individually in terms of morphological features of the embryo such as relative egg length . First , this method doesn't rely on choosing an arbitrary threshold for deciding whether a cell is “expressing” or not . Choosing thresholds is particularly problematic for genes with graded expression patterns such as the gap genes . Second , the expression distance score makes use of the whole expression level time course while factoring out the effects of morphological movements ( i . e . cell flow ) . Additionally , the expression distance score can be used as a natural criterion for selecting cells amongst a set . For example , to find cells with similar expression profiles near to a given query cell , one could first define a set of nearby cells to search , then calculate the expression distance score for the query cell compared to each cell in the set . The best match will have the lowest expression distance score ( Figure 4 ) . We use the expression distance score to compare the expression profiles of cells that are spatially nearby both within and between species to determine how expression patterns differ in terms of their output , relative location in the embryo , and the relative number of expressing cells . Assessing statistical significance of the expression distance score directly is difficult since it is based on multivariate quantities whose correlations we have not measured; this would require co-staining every pair of genes in our dataset . We provide two methods to gauge significance . First , we constructed two atlases of D . mel expression from disjoint sets of embryos . Each of these atlases was assembled from approximately the same number of individual embryos per gene as the D . yakuba and D . pseudoobscura atlases . Expression distances between cells in these two atlases provide a baseline for what should be insignificant expression distance scores with respect to measurement error and intra-strain variability . Second , we analyzed differences in expression level for each gene and time point independently using a two-sample t-test ( see Materials and Methods ) . For a pair of cells in two different atlases , we can determine whether a gene's expression is significantly different relative to the variance across measurements of that gene and time point . We can then declare a pair of corresponding cells to be different if they have significantly different expression levels of one or more genes at one or more time points , applying a suitable correction for multiple hypothesis testing . We tallied the number of entries in the expression profile that are statistically different , and call this the t-test score . It is more conservative than the expression distance since it doesn't detect the sum of many small differences across multiple genes or time points . However , it does provide a simple model of statistical significance , validating that the average expression differences we observed are significant relative to error in our measurements . Subtle differences in the dynamics of expression patterns are detectable from inspecting the averaged , normalized expression patterns of all genes in the dataset ( Figure 2 ) . The peak of expression varies between species for multiple genes ( note Kr , fkh , hkb and ftz ) . For some patterns with multiple domains , such as eve , the relative level of the different stripes varies between species . Finally , some patterns also vary differently over the dorsal/ventral axis ( note the longest anterior stripe of gt in later time points ) . To systematically analyze variation in each gene's expression pattern , we calculated the expression distance score for each cell in D . melanogaster compared to its spatially nearest cell in D . yakuba or D . pseudoobscura , for each gene in our dataset , one at a time . Because embryos are of different sizes , we scaled each embryo to the same relative egg length and aligned atlases by their centers of mass before determining spatial relationships between cells . To determine if there are positional shifts in expression patterns , we then performed a local search amongst the 30 spatially nearest cells for the cell with the best match to the D . melanogaster expression profile . This corresponds to movement by 3–4 cells in any direction . We did do not require a one-to-one match; instead we allowed multiple query cells to match the same cell in the target species . This flexibility was necessary because of the differing numbers of cells between the species; forcing a one-to-one match would give misleadingly large expression differences for cells that have clear counterparts in the target embryo , but too few of them . To visualize the results of the search , we assigned the query cell the score of its best match . The breadth of our dataset prevents us from presenting all of these results in the main text of this paper . For this analysis , and the others described below on single gene expression profiles , we show representative data from even-skipped in Figure 5 and the remaining data is presented in Figure S4 . This data can also be viewed using our interactive visualization tool , MulteeSum ( see Materials and Methods ) . A local search improved the expression distance score for most D . melanogaster cells as compared to a direct spatial mapping , indicating that eve expression patterns have shifted in space ( Figure 5 ) . This also holds for the t-test score ( Figure S5 ) . More generally , this is true of all other genes in our dataset , where the mean expression distance score decreases 2 to 5-fold using local search , indicating widespread shifts in relative position ( Figure S4 , Table S3 ) . To visualize the direction of positional shifts in expression , we determined the distance and direction to the average position of each cell's top 10 hits ( Figure 5 , Figure S4 ) . Expression of eve is shifted anteriorly for some stripes in D . yakuba , while it is shifted posteriorly for all stripes in D . pseudoobscura . This is consistent with more conventional representations such as plotting stripe boundaries for specific time points , which also show significant differences in the relative position of eve stripe boundaries ( Figure S6 ) . The direction of movement is roughly similar across most genes , with the exception of the terminal genes , where the movement is towards the poles; there is a partial anterior shift for many D . yakuba genes and a pronounced posterior shift for nearly all D . pseudoobscura genes ( Figure S4 ) . Not all cells have a perfect match in the other species , as indicated by higher expression distance scores even after a local search . For eve , cells in the middle of some stripes differ in their dynamics and relative level ( Figure 5 , cells labeled b ) . Differences of this sort are apparent at all tiers of the network ( Figure S4 ) . This analysis is an underestimate of expression differences because we do not force one to one matching; there are thus some cells in D . yakuba and D . pseudoobscura that are not matched . We analyzed the number of matched cells in D . yakuba and D . pseudoobscura ( Figure S7 and MulteeSum , see Materials and Methods ) , and found that most ( >85% ) D . yakuba and D . pseudoobscura cells appeared in the top 10 matches to at least one D . melanogaster cell . Furthermore , unmatched cells were distributed spatially almost exclusively in areas where eve is not expressed , indicating that there are not large populations of unmatched cells in D . yakuba and D . pseudoobscura that are significantly different than their matched neighbors . From the analysis of individual genes , we learned that the relative position of many genes has shifted and that there are some differences in relative levels and dynamics . To assess whether these differences are due to positional shifts in the expression of multiple genes or changes in input functions , we compared gene expression profiles for multiple genes in our dataset simultaneously . Consider the case where the expression pattern of one gene has shifted in space . If this change in expression ( the output ) is due to a change in the position of an upstream regulator , we would expect the cell's gene expression profile to remain the same . If it is due to a change in the gene's input function ( i . e . it is responding to an upstream input differently ) , we would expect a difference in the concentration of inputs relative to outputs; in other words , a change to the cell's gene expression profile . For cases where the regulatory relationships between inputs and output are well defined , the relation between expression patterns and the input function can be modeled and tested directly . We have undertaken this type of analysis for expression of the hunchback posterior stripe in a parallel study ( Z . Wunderlich et al . , submitted ) . However , the segmentation network is highly interconnected [28] and not all regulatory relationships have been identified . We therefore calculated the expression difference score for all genes in our dataset simultaneously to assess the extent of regulatory differences across the segmentation network in an unbiased , exploratory manner . Cells with differences in cellular gene expression profiles reveal potential regulatory differences . However , these differences are not attributable to any particular input function without further analysis . Figure 6 shows the expression distance metric calculated using all genes in our dataset except for bcd and cad ( see Materials and Methods ) . As we did for matching cells based on single gene expression profiles , we searched locally amongst the nearest 30 cells for the best match to the query cell , and did not require a one-to-one match . We confirmed that our matching protocol is not missing large numbers of cells in D . yakuba and D . pseudoobscura ( >99 . 5% matched ) , and that unmatched cells are intermingled with matched cells ( Figure S8 and MulteeSum , see Materials and Methods ) . We again found that a local search significantly decreases the expression distance score relative to direct spatial mapping for many cells , with the mean decreasing by 2-fold ( Figure 6 , Table S3 ) . This reflects the shifted relative position of expression for many genes in both species . Visualizing the direction of each pair-wise match reveals areas where corresponding cells are shifted along the anterior/posterior or dorsal/ventral axis; frequently they are shifted along both axes , and the patterns are consistent with those observed for individual genes . These positional shifts are particularly uncoordinated in the ends of the embryos , where corresponding cells are found both closer and further away from the ventral midline . This may be in part because the atlases are assembled by registration using pair-rule genes whose expression is confined to the trunk , and hence our expression data is less accurate at the poles [7] . In the trunk region , there is a pronounced anterior shift of D . yakuba cells relative to D . melanogaster in the anterior , and a pronounced posterior shift of D . pseudoobscura cells relative to D . melanogaster throughout the eve expressing region . The genes that are expressed in the trunk are highly interconnected; most regulate one another and would therefore be expected to move together . From the variety of positional shifts observed , we conclude that these expression differences are not likely to result from simple changes in the maternally driven morphogens bicoid or caudal , in which case we would expect coordinated positional shifts along a single axis . Instead , our data is consistent with many small-scale changes throughout the network . After searching locally , the majority of D . melanogaster cells do not differ from their corresponding cells in D . yakuba in more than five of their expression profiles' entries , or in D . pseudoobscura by more than seven ( Figure S9 ) . This indicates that most expression differences we observe for individual genes are attributable to coordinated positional changes in the network as a whole . For some D . melanogaster cells , the best match still exhibits some expression differences according to the expression distance metric ( Figure 6 ) . The expression distance metric could be high in these cases due to large differences in expression for a single gene , or small differences in many genes . By examining the underlying gene expression profiles for the cells with the highest expression distance metric , we find the latter to be true; the differences that remain after local matching are due to quantitative changes in dynamics and relative levels of expression for multiple genes , rather than the presence or absence of a particular gene product ( Figure 6 and MulteeSum , see Materials and Methods ) . Because the expression distance score is additive , we can assess which genes contribute to the overall score by calculating the expression distance score for each individual gene as well as relevant subsets ( Figure S10 ) . We find that differences are widespread; they are not confined to a single gene or tier of the network . The t-test score reveals good concordance between those cells that have large expression distance and those that differ significantly in many individual expression measurements ( Figure S9 ) . In both analyses , D . yakuba is more similar to D . melanogaster in terms of gene expression profiles . This is plausible as D . melanogaster and D . yakuba are more closely related than D . melanogaster and D . pseudoobscura [29] . Some of the small expression differences we identify using the expression distance metric undoubtedly represent experimental noise , but some may represent bona fide regulatory differences between these species . Notably , our expression distance metric identifies cells with differing expression of odd and prd between D . melanogaster and D . yakuba , and both odd and prd exhibit differential binding of hb in D . yakuba , as measured by Chip-Seq [30] . Together with our data , this indicates potentially altered input functions for these genes . Verifying candidate regulatory differences will require assigning them to specific input functions , and functional studies to determine the mechanistic basis of the regulatory change . At this stage of development , cells are still morphologically similar and yet are committed to their future fates as components of larval structures [15] . Their fate is highly correlated with their spatial position in the embryo and is determined by the set of genes that they express . Therefore , we consider gene expression profile to be equivalent to cell type at this stage . Even if all cell types had precisely equivalent gene expression profiles , they could give rise to morphological differences between embryos if they occur in different relative locations , or in different proportions between embryos . In the previous section , we established that equivalent cell types occur in different relative locations in these three species . Because the embryos also have different numbers of nuclei , a natural question is whether they allocate cell types proportionally . One possible solution to analyzing cell types would be to cluster the cells based on expression profile and count the number of cells within each cluster . However , expression is changing in a graded way at almost every point in the embryo making it difficult to decide how many clusters there should be . Instead of using an arbitrary clustering of cells , we determined how many adjacent cells are similar to a given query cell by counting how many adjacent cells are within a given expression distance score . This connected set of “expression neighbors” is therefore a group whose expression profile is quantitatively similar to the chosen cell . For any fixed threshold imposed on the expression distance score , the size of this neighborhood captures how quickly expression levels change in the vicinity of a cell . We visualize how this neighborhood size varies over the surface of the embryo ( see Figure 7 , left ) . Large neighborhoods correspond to regions of roughly constant cell type . Thus , the number of expression neighbors per cell provides a means to compare allocations of cells between different species on a cell-by-cell basis . If variations in expression between different species reflected a simple uniform scaling , then the neighborhood size for every nucleus would also be proportionally smaller or larger by the same scale factor . On the other hand , if the patterning network of one species allocates a relatively larger population of cells to a given type in some region of the embryo , then the local neighborhood for each of the cells in that region will grow larger . We calculated the relative expansion or shrinkage of each cell neighborhood between corresponding best-matched cells in D . melanogaster , D . yakuba and D . pseudoobscura . The most important observation from this analysis is that the relative proportion of cells in these expression neighborhoods varies both up and down by as much as 5 fold ( Figure 7 ) . These three species allocate cell types quite differently; there are discrete areas of relative expansion and contraction . For example , there are relatively more D . melanogaster cells than their equivalents in D . yakuba and D . pseudoobscura in the posterior trunk , ( roughly corresponding to the position of the last 3 stripes of even-skipped expression ) , but fewer immediately posterior , on the border of terminal gene expression . We conclude that small changes to the behavior of the patterning network , achieved either through quantitative regulatory changes , or by initiating patterning in a new morphological context , or both , can result in different proportions of cells allocated to conserved cell types . These could serve as the initial basis for downstream morphological changes .
Identifying the genetic differences that cause variation in gene expression is a major goal not only for evolutionary developmental biologists , but also for those interested in human disease . An increasing number of disease associated variants have been mapped to regulatory regions of the genome [31]; to contextualize their effects we must learn which sequence variants are likely to alter gene expression and which will not . The approach we describe identifies candidate regulatory differences from cellular resolution data on a network of interacting genes . To obtain data for all of the relevant genes over time , we built averaged atlases of gene-expression using high-resolution imaging and registration techniques . This type of data is likely to become increasingly common as these technologies continue to improve . For example , live in-toto imaging techniques such as SPIM have been successfully applied to blastoderm embryos and are likely to provide a view of the behavior of the network at much higher temporal resolution [32] , [33] . As sequencing methods become more sensitive , they may also be able to generate spatially resolved data by either separating cells for biochemical analysis or using imaging-based methods to sequence transcripts in situ [34] . We therefore anticipate that increasing numbers of studies will involve comparing spatially resolved cellular resolution gene expression profiles between different samples from different species , different populations , or from the same individual under different conditions . The expression distance metric is a useful tool to focus attention on subsets of interacting components that are likely to show different behavior between species . Such methods may be applied to less well-characterized gene regulatory networks , where unbiased methods for reconstructing gene regulatory networks and mapping expression differences onto a network from a combination of genomic and functional data will be required . A grand challenge in the post-genomic era is how to move from broadly identified expression differences to precise identification of mechanistic differences in the underlying gene regulatory networks . Regulatory divergence across multiple scales , from the topology of the network to fine-scale changes in input functions , has been observed in comparative studies of the Ascoycota fungi [35] and animals [36] . In principle , the quantitative differences in gene expression we observe could result from many non-mutually exclusive components of the gene regulatory network , including changes in trans-acting TFs ( either in their DNA binding affinity or in their interaction with other components ) , cis-regulatory modules ( CRMs ) , chromatin structure , promoter architecture , or transcript stability . Even at the short evolutionary distances studied here , genetic changes are observed in multiple tiers of this developmental network . DNA binding domains of TFs are highly conserved , with only single amino acid changes in some lineages , but the remainder of the protein diverges more rapidly [29] . cis-regulatory sequences that interpret the concentrations of these TFs differ substantially in terms of the number , affinity and arrangement of TF binding sites [37] . Notably , a recent comparative study of the dorsal/ventral patterning network in blastoderm embryos of Drosophila showed that changes in the arrangement of TF binding sites in CRMs leads to quantitative gene expression differences between species by altering input functions [38] . Other relevant features such as chromatin structure , promoter architecture and miRNAs have recently been systematically functionally characterized in D . melanogaster [39]–[41] , laying the foundation for future comparative studies . Attributing expression differences to these features will require a model of the system to generate experimentally verifiable hypotheses . There are an increasing number of models that take advantage of spatially resolved expression data and knowledge of TF binding sites to predict CRM output [42]–[45] . However , these models do not predict expression accurately enough to capture the quantitative differences we observe between species . Our high-resolution expression data are well suited to the development of new types of models for ascertaining the source of expression differences , a clear line of future experimentation . Despite the quantitative differences in cellular gene expression patterns that we measure , the segmentation network produces remarkably similar cell type output in the face of substantial genetic and morphological perturbation . This implies that formation of these cell types is under strong selective constraint . D . melanogaster embryos can tolerate variation in the proportion of cell types , though there is an upper limit on how much the patterning system can be compressed [46] . The differences we observe may reflect neutral drift within these limits . For example , there may be restrictions on nearly neutral processes of binding site turnover , where small sequence changes cause quantitative variation in output , and subsequently require fine-tuning of expression to stay within acceptable limits [47] . This sort of process would result in fine-scale expression changes as the acceptable limits are explored . Because many expression patterns are qualitatively conserved between closely related species , the prevalent model in the field is that CRMs operating in these species are functionally equivalent , as has been shown for some test cases [12] . However , a recent study found patterns of variation in Drosophila blastoderm CRMs that are inconsistent with a nearly neutral process [48] . As Kreitman and colleagues point out in that paper , “the assumption of CRM functional stasis , which is the main argument for the neutral ( i . e . , compensatory ) view is not well supported experimentally . ” Though not attributable to differences in CRMs without further study , we do provide evidence of quantitative differences in expression for many genes in the segmentation network between closely related Drosophila species . Alternatively , the proportion of cell types may be selected upon directly , as they could contribute to organismal phenotypes by propagating through later stages of development to create fine-scale differences between these species . This would represent selection on a quantitative intermediate developmental trait , likely mediated by the type of small scale differences in expression for multiple genes we observe in our dataset . This scenario would differ from selection on macroscopic terminal organismal phenotypes such as changes in pigmentation , bristle number and skeletal structures , where small numbers of loci or even single loci , of large effect have been identified [49]–[52] . Finally , it is possible that the differences in gene expression are a consequence of selection on egg size and morphology . Egg size is known to be a selectable trait and to vary significantly across populations [53]–[55] . The expression differences we see would then reflect how the segmentation network has been fine-tuned to operate in different morphological contexts while maintaining the proper allocation of cell types . This idea was also recently put forth by Kreitman and colleagues to account for evidence of positive selection on Drosophila blastoderm CRMs , as mentioned above [48] . They term this the “moving target” hypothesis , and posit that input functions must constantly adapt to changing conditions within the embryo . We favor this hypothesis as well . It remains a future challenge to identify both the target of selection for this network , and the design principles that confer its robustness to genetic and morphological perturbation .
Embryos were collected , fixed and prehybridized according to standard protocols , which are available at http://depace . med . harvard . edu/links . html , and described in [6] , [26] . Briefly , D . yakuba and D . pseudoobscura cages were maintained at 23°C . D . yakuba embryos were collected for 3 hours , and aged for 2 hours prior to fixation . D . pseudoobscura embryos were collected for 3 hours , and aged for 3 hours prior to fixation . Embryos were dechorionated in 50% bleach for 3 minutes , washed , and fixed in a 1∶4 solution of 10% formaldehyde ( Polysciences #04018 ) to heptane for 20 minutes with vigorous shaking . The vitelline membranes were removed by shaking with MeOH and washed 3X with 100% MeOH . Fixed embryos were stored at −20°C in 100% ethanol . Embryos were pooled for prehybrization , rehydrated in PBT + Tx ( PBS pH 7 . 2 , 0 . 05% Tween20 and 0 . 2% Triton X-100 ) , post-fixed for 20 minutes in 5% formaldehyde in PBT+Tx , washed in hybridization buffer ( 50% formamide , 5X SSC pH 5 . 2 , 0 . 2% Triton X-100 , 40 µg/ml heparin , and 250 µg/ml salmon sperm DNA ) and incubated at 55°C for 1 to 5 hours in hybridization buffer . Prehybridized embryos were stored in hybridization buffer at −20°C . There were two modifications to the staining protocol developed for D . melanogaster [6] . First , species-specific RNA probes were made using cDNA or genomic DNA as a template , whereas cDNA probes were used exclusively for the D . melanogaster data . Probes ranged in size from 531 bp to 2771 bp , and either encompassed the majority of the coding sequence or overlapped large exons ( Table S4 ) . Variation in probe length did not significantly affect our measurements ( Figure S11 ) . Second , two different haptens are required for our imaging pipeline , one for the registration gene and one for the gene of interest . While dinitrophenol ( DNP ) - labeled probes gave consistently clean results in all species , digoxygenin ( DIG ) - labeled probes yielded variable levels of background . Another commonly used hapten , biotin , was even worse . Because DIG stains were strong enough to reliably distinguish stripes , we chose to use it for the registration channel , but not include the data in the final gene expression atlases . Probe templates were cloned by PCR amplification using either genomic DNA or cDNA libraries as a template , ligated into pGEM-Teasy , and sequence verified . Cloning primers are listed in Table S4 . Probe templates were generated by PCR with M13 forward and reverse primers . Anti-sense digoxygenin ( DIG ) or dinitrophenol ( DNP ) probes were synthesized using in vitro transcription from DNA templates using either SP6 or T7 polymerase , depending on the orientation of the clone . Probes were not carbonate-treated as this did not improve stain quality . All probes were diluted to 200 ng/µl . For in situ hybridizations , approximately 100 µl of embryos were incubated for up to 48 hours at 55–57°C in 300 µl of hybridization buffer with 2–10 µl each of a DIG and DNP probe . Embryos were then washed extensively with hybridization buffer at 55–57°C , and probes were detected sequentially using horseradish-peroxidase ( HRP ) conjugated antibodies ( anti-DIG POD , Roche 11207733910 at 1∶250 or 1∶500; anti-DNP Perkin Elmer NEL747 A001KT at 1∶100 ) and either coumarin or Cy3 tyramide amplification ( Perkin-Elmer NEL703 001KT , SAT 704B ) . To disable the HRP in the first signal detection reaction , embryos were washed in hybridization buffer at 55°C and incubated in 5% formaldehyde in PBT+Tx for 20 minutes . All remaining RNA was removed by incubation with 0 . 18 µg/ml RNAse A in 500 µl PBT+Tx overnight at 37°C . Nuclei were detected by staining with Sytox Green ( Molecular Probes #S7020 , 1∶5000 in 500 µl overnight at 4°C ) . Embryos were dehydrated in an ethanol series and mounted in xylene-based DePex ( Electron Microscopy Service #13514 ) on a slide with 2 bridging coverslips to prevent flattening of the embryos . Detailed protocols are available at http://depace . med . harvard . edu/links . html . Three-dimensional image stacks of individual embryos were acquired semi-automatically on a Zeiss LSM 710 using a plan-apochromat 20X 0 . 8NA objective . Embryos were located , staged using phase contrast optics , and the imaging parameters such as the height of the image stack and gain settings for each fluorophore were recorded . A custom built macro then acquired all marked embryos [7] . All three fluorophores ( Sytox Green , coumarin and Cy3 ) were excited simultaneously at 750 nm , using a Coherent Chameleon 2-photon laser at 4–7% power . The emission was spectrally split into 3 channels: 462–502 nm ( coumarin ) , 514–543 nm ( sytox ) , 599–676 nm ( Cy3 ) . Images were 1024×1024 , and slices were taken every 1 µm . Resulting image stacks were processed by previously described algorithms to unmix channels [56] , and segment individual nuclei [6] , resulting in individual pointcloud files for each embryo . These were housed in a custom-built database . Gene expression atlases were assembled using the registration algorithms previously described in [7] . For each species , a morphological model was constructed that contained an average number of nuclei . The 3D positions of the nuclei in the model were chosen to match the average egg-length , shape and density pattern measured for each of the 6 temporal cohorts . Motions of nuclei between time points in the model were constrained to be as small and smooth as possible while still recapitulating the observed changes in density and shape ( see [27] for details ) . Pointcloud data extracted for each embryo in a given cohort were aligned to the morphological template by a rigid-body transformation and isotropic scaling . For each time point , a registration template was constructed by finding average boundary locations of a registration marker gene ( eve or ftz ) with respect to the egg-length of the morphological model . Fine registration of individual embryo pointclouds was then carried out by non-rigid warping of the embryo to align marker gene boundaries with the template . Finally , expression values were computed for each nucleus and time point in the model by averaging measurements across those nuclei in individual pointclouds that were closest after spatial registration . Prior to averaging , gains and offsets were estimated for expression measurements within each embryo pointcloud in order to minimize the expression variance across the cohort and to match smoothed estimates of the total change in expression level between temporal cohorts ( see [7] for details ) . Surface area was computed as the sum of areas of the triangles defined by the neighbor relation information in the Pointclouds [6] . Local density was computed by defining a disk of 15 µm radius on the surface around each nucleus , and dividing the number of nuclei in this disk by its area [6] . These density maps were then averaged over a cohort of embryos by resampling the cylindrical projections onto a regular grid . Cell-to-cell comparisons within and between species were made by looking at the squared distance between vectors of average expression measurements for the cell at all 6 time points and 11 genes . For a pair of nuclei i and j we computed the distance:where eigt is the expression of the gth gene recorded in the atlas for the ith cell at time point t . We used squared distance since it is additive across genes and time-points which makes the contribution of individual genes more interpretable . Prior to computing the distance , expression levels for each gene in the atlas were scaled so that the maximum expression at each time point was 1 . 0 . In order to determine relevant cells to compare between species , only cells that were nearby were considered . Corresponding locations were estimated by scaling each atlas to unit egg length and nearby nuclei were specified as those nuclei in the target embryo that were within the 30 nearest to the cell to be matched . Since there are often several cells that are good matches , the displacement direction to the best matching nearby cell is noisy . In order to visualize displacement , we used a weighted average of the locations of the top 10 matching cells ( smallest expression distance ) . The 3D locations of these 10 matching cells were averaged using weights inversely proportional to the expression distance ( i . e . 1/dij ) . These 3D displacement vectors were then visualized on a cylindrical projection . We chose not to include bcd and cad in calculation of the expression distance score for the entire dataset . We excluded bcd because its expression increases over the first two time points in the D . melanogaster dataset; this is likely an experimental artifact and leads to artificially high expression distance scores in the anterior . We excluded cad because data was not available for D . pseudoobscura and we wished to compare results between the D . melanogaster/D . yakuba and D . melanogaster/D . pseudoobscura analyses . As an alternative to expression distance scores , we also considered a hypothesis-testing framework in which two cells are declared to have different expression profiles if the expression of some gene at some time point is significantly different relative to variance in our measurements . This comparison was carried out independently for each cell , gene and time-point using a two-sample t-test with unequal sample sizes and variances . In all tests we used the Bonferroni correction to assure a family-wise error rate of less than 0 . 01 . Visualizations and histograms in Figures S5 and S9 show the number of expression profile measurements for which a given cell was significantly different under this significance threshold . We defined an expression neighborhood Ni for a given nucleus i in the following way . Choose a threshold t and find all nuclei in the atlas for which dij<t . Of these nuclei with similar spatio-temporal expression profiles , let Ni be the largest connected component on the embryo surface that contains cell i . For areas where the expression pattern varies rapidly in space , this neighborhood of similar cells is small . In areas where the pattern changes slowly , the neighborhood is large . To determine how these neighborhoods might expand or contract between different species , we consider the neighborhood size around corresponding nuclei . Let j be the nucleus in the target atlas whose expression profile best matches i in the source atlas . We compare the relative sizes of the two neighborhoods in order to gauge the degree of expansion or contraction measured by the log ratio of neighborhood sizes: The log ratio is symmetric about zero with positive values indicating an expansion and negative values indicating a contraction . One concern is that the choice of neighbor threshold may affect this analysis since there is not a meaningful way to scale the measured fluorescence levels between atlases of different species . To resolve this , we choose the threshold for each target atlas adaptively . Given a fixed threshold for the query atlas , we searched over thresholds for the target atlas in order to find a threshold in which the average expansion ratio R across all cells matched the log-ratio of the number of nuclei in the two atlases . Choosing the threshold in this way entails that ratios are visualized relative to a null hypothesis of uniform scaling between species . Figures similar to Figure 5 , Figure 6 , Figure S4 , Figure S7 , and Figure S8 can easily be generated using MulteeSum , a custom software tool for visualizing comparative analysis of cellular gene expression profiles [57] . Our datasets are complex and best viewed interactively . We therefore have made MulteeSum and the analyses presented here available for download at http://depace . med . harvard . edu/downloads/MulteeSum . zip . We have released MulteeSum open source , and it was developed using the Processing programming language ( http://www . processing . org ) , an open-source language for visualization . Executables for running on Mac OSX , Windows and Linux and instructions ( see README . txt ) are included in the download . A full description of the usage and features of MulteeSum can be found at http://www . multeesum . org .
|
For a gene to function properly , it must be active in the right place , at the right time , and in the right amount . Changes in any of these features can lead to observable differences between individuals and species and in some cases can lead to disease . We do not currently understand how the position , timing , and amount of gene expression is encoded in DNA sequence . One approach to this problem is to compare how gene expression differs between species and to try to relate changes in DNA sequence to changes in gene expression . Here , we take the first step by comparing gene expression patterns at high spatial and temporal resolution between embryos of three species of fruit flies . We develop methods for comparing gene expression in individual cells , which allow us to control for variation in the size , shape , and number of nuclei between embryos . We find measurable quantitative differences in the patterns for all individual genes that we have examined . However , by considering all genes in our dataset at once , we show that many genes are changing together , leading to largely equivalent types of cells in these three species .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"systems",
"biology",
"developmental",
"biology",
"gene",
"expression",
"genetics",
"regulatory",
"networks",
"biology",
"computational",
"biology",
"pattern",
"formation",
"genetics",
"and",
"genomics",
"evolutionary",
"developmental",
"biology"
] |
2011
|
A Conserved Developmental Patterning Network Produces Quantitatively Different Output in Multiple Species of Drosophila
|
Lethal toxin ( LT ) is a critical virulence factor of Bacillus anthracis , the etiological agent of anthrax , whose pulmonary form is fatal in the absence of treatment . Inflammatory response is a key process of host defense against invading pathogens . We report here that intranasal instillation of a B . anthracis strain bearing inactive LT stimulates cytokine production and polymorphonuclear ( PMN ) neutrophils recruitment in lungs . These responses are repressed by a prior instillation of an LT preparation . In contrast , instillation of a B . anthracis strain expressing active LT represses lung inflammation . The inhibitory effects of LT on cytokine production are also observed in vitro using mouse and human pulmonary epithelial cells . These effects are associated with an alteration of ERK and p38-MAPK phosphorylation , but not JNK phosphorylation . We demonstrate that although NF-κB is essential for IL-8 expression , LT downregulates this expression without interfering with NF-κB activation in epithelial cells . Histone modifications are known to induce chromatin remodelling , thereby enhancing NF-κB binding on promoters of a subset of genes involved in immune response . We show that LT selectively prevents histone H3 phosphorylation at Ser 10 and recruitment of the p65 subunit of NF-κB at the IL-8 and KC promoters . Our results suggest that B . anthracis represses the immune response , in part by altering chromatin accessibility of IL-8 promoter to NF-κB in epithelial cells . This epigenetic reprogramming , in addition to previously reported effects of LT , may represent an efficient strategy used by B . anthracis for invading the host .
Pulmonary infection by B . anthracis , the etiologic agent of anthrax disease [1] , has been shown to be the most life-threatening form of the disease as compared to gastrointestinal and cutaneous forms of anthrax . Whatever the infection route used by this bacterium , spores are taken up by macrophages and/or dendritic cells , and subsequently migrate in the draining lymph nodes where they germinate [2] , [3] . Germination leads to bacterial multiplication and dissemination through the whole organism . Despite a therapeutic intervention , all forms of infection may progress to fatal systemic anthrax , which is characterized by shock-like symptoms , sepsis , and respiratory failure [4] . Therefore , investigation of the immune response triggered by B . anthracis may help to better understand the pathophysiology of anthrax and to develop efficient therapeutic approaches for the treatment of this disease . The innate immune response , including the inflammatory reaction that is the first line of host defense against invading pathogens , is characterized by upregulation of various inflammatory genes due to the activation of transcription factors . IL-8 ( Interleukine-8 , CXCL8 ) is a human chemokine that induces the recruitment of polymorphonuclear ( PMN ) neutrophils from the blood to the injured tissue [5] . This allows PMN to eradicate the invading pathogen within the site of infection . In healthy tissues , IL-8 is barely detectable , but is rapidly induced by nuclear factor κB ( NF-κB ) and activator protein 1 ( AP-1 ) -dependent inflammatory stimuli such as TNFα , bacteria or virus [6] . Recently , significant advances in the understanding of signaling pathways , which regulate IL-8 transcription as well as mRNA stabilization in response to external stimuli , have been made [6] . It is now clearly established that induction of inflammatory genes is associated with local changes in histone modification that plays a major role in epigenetic regulation of gene expression . The core histones including H2A , H2B , H3 and H4 are the basic components of the nucleosomes that organize the cellular DNA in chromatin [7] , [8] . Covalent modifications of the histones play a major role in gene regulation by affecting chromatin compaction and thereby DNA accessibility . These modifications can result in either transcriptional repression ( like for H3me3K9 and H3me1K27 ) or activation ( H3me3K4 , H3pS10 and acetylation at various positions ) [9] , [10] . The exact mechanisms by which B . anthracis induces anthrax disease are not fully understood; however , it is clearly established that this bacterium spreads rapidly into the host at the early stages of infection with low or no detectable immune response [2] , [11] , [12] . This is likely due to the ability of B . anthracis to subvert the host immune response [11]–[13] via , at least in part , the action of tow major toxins , edema toxin and lethal toxin ( LT ) . B . anthracis secretes a transporter called protective antigen ( PA ) which acts in pair with lethal factor ( LF ) leading to the formation of lethal toxin ( LT ) . LT cleaves MAPKK thereby interfering with MAPK cascade [14] , [15] a process known to play a major role in pathogenesis of anthrax . However , the molecular mechanisms that may link MAPK inhibition by LT and modulation of target gene expression remained unclear . We report here that LT impaired the inflammatory response in a mouse model of pulmonary anthrax . Remarkably , LT repressed IL-8 expression without interfering with NF-κB activation , an essential transcription factor for IL-8 expression . Investigation of the molecular mechanisms involved in this inhibition revealed that LT blocked histone ( H3 ) phosphorylation at serine 10 , a modification normally induced by activation of the MAPK pathway and that is also involved in chromatin accessibility for NF-κB . This epigenetic reprogramming is probably a key mechanism by which B . anthracis circumvents the immune response during the earlier stage of infection .
We examined the effect of B . anthracis on lung inflammation in a mouse model of pulmonary anthrax and the contribution of LT to this effect . All strains used in this study were in bacilli form . The degree of inflammatory reaction was assessed by measuring the recruitment of polymorphonuclear ( PMN ) leukocyte , and the levels of proinflammatory cytokines interleukin-6 ( IL-6 ) and KC ( mouse functional orthologue of human IL-8 ) in broncho-alveolar lavage fluids ( BALFs ) . Intranasal instillation of the RPLC2 strain ( producing inactive LT ) induced an increase of IL-6 and KC levels ( Figure 1A ) and PMN recruitment ( Figure 1B ) , in contrast to the RP9 strain ( producing active LT ) which had no effect . Intratracheal instillation of LT to mice before bacterial administration impaired RPLC2-induced increases of IL-6 and KC levels ( Figure 1A ) and PMN recruitment ( Figure 1B ) . Intranasal instillation of LPS ( instead of bacteria ) induced an increase in concentrations of MIP-2 , KC and IL-6 levels ( Figure 1C ) and PMN recruitment ( Figure 1D ) . All these responses were abrogated by intratracheal instillation of LT before LPS . These findings led us to examine the effect of LT on the in vitro secretion of cytokines by primary cells isolated from mouse lungs . LT blocked LPS-induced KC and IL-6 secretions by primary epithelial cells ( Figure 2F ) and MIP-2 , KC and IL-6 secretions by alveolar macrophage ( AMs ) ( data not shown ) . We next investigated in more details the effects of LT on cytokine production in a human epithelial cell line Beas-2B . The RPLC2 strain induced IL-8 secretion by Beas-2B cells in contrast to the RP9 strain ( Figure 2A ) . The RP9 strain inhibited TNFα-induced IL-8 secretion , whereas the RPLC2 strain had no effect ( Figure 2B ) . LT blocked TNFα-induced IL-8 mRNA expression ( Figure 2C ) , IL-8 ( Figure 2D ) and IL-6 ( Figure 2E ) secretions . Inhibition of IL-8 secretion by LT was also observed when Beas-2B cells were stimulated by IL-1β or LPS instead of TNFα ( data not shown ) . As NF-κB is a key transcription factor of inflammatory gene expression including IL-8 [8] , we examined its implication in IL-8 expression in our cell system . We analyzed IL-8-promoter activation in Beas-2B cells transfected with the ( −133 pb ) IL-8-promoter or an [ΔNF-κB] IL-8 promoter in which the responsive elements of NF-κB were mutated . TNFα induced the activation of the ( −133 pb ) IL-8-promoter but failed to stimulate the activation of the [ΔNF-κB] IL-8 promoter ( Figure 3A ) . Preincubation of cells with BAY , an NF-κB inhibitor , abrogated TNFα-induced IL-8 secretion ( Figure 3B ) . These findings indicated that NF-κB is essential for IL-8 expression in Beas-2B cells . The use of the [ΔAP-1] and [ΔNF-IL6] IL-8 promoter showed that AP-1 and NF-IL6 are only marginally involved in the induction of the IL-8 expression by TNFα ( data not shown ) . These findings led us to examine whether inhibition of IL-8 expression by LT is due to an alteration of NF-κB activation . Our results showed that neither LT nor MAPK inhibitors inhibited TNFα induced NF-κB activation or translocation . MAPK inhibitors even enhanced this activation ( Figure 3C and 3D ) . MAPK cascade is known to modulate IL-8 expression in various cell systems [8] . This was confirmed in the present study using SB202190 and PD98059 , specific inhibitors of p38-MAPK and ERK , respectively . Indeed , pre-treatment of Beas-2B cells with these compounds , significantly reduced TNFα-induced IL-8 expression ( Figure 4B ) . We also showed that LT cleaves MEK1 , MEK2 , MEK3 in these cells ( Figure 4C ) , a finding in agreement with the know ability of LT to cleave MAPKKs [14] , [15] . LT inhibited TNFα-induced ERK and p38-MAPK phosphorylation , but had no effect on JNK phosphorylation ( Figure 4A ) . Studies on Beas-2B cells transfected with the IL-8 promoter reporter plasmid showed that LT and MAPK inhibitors had no effect on TNFα-induced IL-8 promoter activity ( Figure 4D ) , which contrasted with the ability of these compounds to inhibit the expression of endogenous IL-8 . We next examined the effect of LT on the activation of mitogen and stress-activated kinases ( MSK ) , located down-stream MAPKK pathways , and involved in histone phosphorylation . Incubation of Beas-2B cells with TNFα increased the level of MSK2 phosphorylation which was reduced by pre-treating cells with LT ( Figure 4E ) . No detectable expression of MSK1 or phospho-MSK1 was observed in these cells ( data not shown ) . The results above suggested that LT inhibited IL-8 expression , at a site probably located down-stream of NF-κB activation/translocation and depending on MAPK cascade . This led us to examine the effect of LT on the phosphorylation of H3 , known to promote the accessibility of NF-κB to target promoters [5]–[8] . Western blot and immunofluorescence analyses showed that LT suppressed TNFα-induced H3 phosphorylation at Ser10 and acetylation at Lys14 ( Figure 5A and 5B ) in Beas-2B cells , suggesting that LT can impair the chromatin accessibility of NF-κB at the IL-8 promoter . We stimulated Beas-2B by TNFα with or without LT and then carried out chromatin immunoprecipitation ( ChIP ) assays using an antibody against p65 subunit of NF-κB , RNAPII antibodies or using an irrelevant immunoglobulin G ( IgG ) of the same isotype as a control . ChIP analyses showed that LT suppressed TNFα-induced p65 and RNAPII recruitment to the IL-8 promoter while no effect was observed in the recruitment at the GAPDH promoter ( Figure 6A ) . In addition , the results showed that TNFα induced phosphorylation at Ser10 and acetylation at Lys14 of histone H3 on the IL-8 promoter whereas the other modifications on H3 were not significantly regulated . LT abrogated TNFα-induced presence of di-modified histone H3 phosphorylated at Ser10 and acetylated at Lys14 at the IL-8 promoter ( Figure 6B ) . Acetylation at Lys9 and tri-methylation at Lys4 ( known to reflect positive modulation of transcription ) were not modified on the IL-8 promoter in our experimental conditions . Tri-methylation at Lys9 , which generally correlates with inhibition of transcription , occurred in parallel to tri-methylation at Lys4 ( Figure 6B ) . We next performed ChIP analyses on a mouse epithelial cell line MLE-15 to examine the effect of LT on the KC promoter . We verified that LT inhibited TNFα-induced KC expression in these cells ( data not shown ) . We observed almost similar regulations of histone modifications in this promoter ( Figure 7 ) . The weaker effects on KC versus IL-8 of the TNFα treatment on the H3pS10 and AcK14 modifications could be explained partially by the fact that the KC promoter appears to contain more H3me3K4 than the IL-8 promoter ( 160% of H3 versus 40% of H3 , respectively ) , suggesting that the KC promoter chromatin could be in a more opened configuration than the IL-8 one .
Pulmonary anthrax is an acute disease of animals and humans caused by inhalation of spores of pathogenic strains of B . anthracis [1]–[3] . Upon infection , the spore germinates and begins to outgrow into toxin-producing replicating bacilli , which can ultimately kill the infected host [2] , [11] , [12] . LT , a major virulence factor of B . anthracis , has been shown to play a key role in the pathophysiology of anthrax and associated mortality [11] , [12] . We demonstrated here that B . anthracis represses the mouse pulmonary inflammatory response ( PMN recruitment and cytokine production ) and that LT plays a key role in this inhibition . Moreover , in the absence of LT , B . anthracis induces this response , which suggests that component ( s ) of this bacterium is ( are ) able to induce an inflammatory reaction , and that their action is masked by the inhibitory effect of LT . This is consistent with previous studies , which have reported that B . anthracis spores stimulate cytokine production in various cells [16]–[18] . Our recent study showed that peptidoglycan , a constitutive cell wall component of B . anthracis , induces type-IIA secreted phospholipase A2 expression in AMs through a NF-κB dependent process [13] . In the earlier stages of infection , inhaled spores trigger an inflammatory response that is subsequently repressed as soon as LT is produced by outgrowing bacilli . Therefore , it was of a great interest to identify the molecular mechanisms by which LT represses the expression of genes involved in the inflammatory response . Early studies demonstrated that phosphorylation of histones at specific amino acid residues regulates gene transcription [7]–[10] . The present study suggested that B . anthracis targets chromatin access for transcription factors , which may alter gene transcription . LT has been shown to suppress cytokine production by a process involving the disruption of MAPKKs [11] , [12] , yet the molecular mechanisms by which this disruption inhibits the expression of inflammatory genes remained to be elucidated . We reported here for the first time that inhibition of cytokine production in epithelial cells by LT occurs via a process involving an alteration of histone H3 phosphorylation/acetylation . Histone modification such as phosphorylation of H3 and acetylation of lysine on histones H3 or H4 can be associated with transcriptional activation [19]–[22] . Indeed , gene expression occurs when the chromatin is in the opened conformation that is promoted in part by histone phosphorylation . Moreover , phosphorylation of H3 is mediated , in part , by mitogen and stress-activated kinase ( MSK ) , which is activated by MAPK ( p38-MAPK and ERK ) signaling pathways [22] , [23] . Our findings showed that LT blocked the activation of p38-MAPK and ERK ( as a consequence of MAPKK cleavage by LT ) and that this inhibition led to the alteration of histone H3 phosphorylation through a process involving an inhibition of MSK2 phosphorylation . Indeed , MSK2 has been shown to mediate histone H3 phophorylation [22] , [23] . The present study shows that both MAPK and NF-κB pathways are essential for IL-8 expression in Beas-2B cells . These results are consistent with previous reports showing that these pathways play a major role in the induction of IL-8 expression in various cell systems [6] . Our results showed that NF-κB activation in Beas-2B cells occurred via a MAPK-independent process . This led us to suggest that MAPK act down-stream of NF-κB activation to modulate IL-8 expression . Our results suggested that epigenetic remodeling initiated by MAPK-induced H3 phosphorylation promotes the accessibility of IL-8 promoters to NF-κB , thereby enhancing IL-8 expression . This process seems also to play a role in the modulation of KC promoter activity , the mouse functional orthologue of human IL-8 promoter . This epigenetic modulation , which is in part under the control of MAPKs , may occur only in the chromatin context whose remodeling favors the recruitment of transcription machinery components including NF-κB . This may explain why MAPK inhibition failed to interfere with the activation of the transfected IL-8 promoter reporter construct , which is not in the context of chromatin and therefore cannot be modulated via a MAPK-dependent epigenetic remodeling . Our results are in agreement with the recent study of Arbibe et al [24] showing that OspF , a virulence factor of Shigella flexneri endowed with phosphatase activity , reduced H3 phosphorylation on a limited subset of gene promoters , including IL-8 promoter , which then become inaccessible to NF-κB . Although both S . flexneri and B . anthracis downregulate inflammatory gene expression , they clearly interfere with H3 phosphorylation at different levels: S . flexneri acts downstream H3 phosphorylation by dephosphorylating ERK p38-MAPKs once they translocate into nucleus whereas B . anthracis blocks this process upstream these MAPKs by a cleavage of MAPKKs in the cytosol . However , it should be noted that , in addition to its epigenetic effect on the transcriptional activity of the IL-8 promoter , LT can also inhibit IL-8 expression via destabilization of IL-8 mRNA [25] . Recent studies have shown that lung epithelial cells interact with B . anthracis suggesting that these cells may play a role in host defense against pulmonary anthrax . Indeed , B . anthracis spores are taken up by lung epithelial cells either in vitro [26] and in vivo [27] and this uptake is followed by entry and germination of spores , though with low efficiency [26] . AMs and dendritic cells are also known to participate in host defense against inhaled B . anthracis , and play a role in spore uptake and germination [2] , [3] . Interestingly , recent reports showed that spore germination and bacterial growth occurred at the site of entry , i . e . nasopharynx and alveoli [28]–[31] . Thus , available data suggest that spore germination , initial bacterial multiplication and lung epithelium crossing may occur along the respiratory tract . The respective role of each cell population still remains to be evaluated in vivo , in particular whether efficiency in spore uptake is paralleled with efficient dissemination from the lung . On the other hand , inhibition by toxins of the inflammatory response initiated by lung epithelial cells may favor the implication of AMs and dendritic cells in subsequent dissemination at later stages of the disease . Moreover , induction of cytokine production by epithelial cells has been shown to occur through a MAPK-dependent pathway [32] . Another report [33] showed that bronchial epithelial cells highly expressed TEM8 , a receptor of B . anthracis toxins , thus confirming the implication of these cells in host response to B . anthracis . Lung epithelial cells have been shown to contribute to the innate immune response after interaction with pathogens such as virus [34] and bacteria [35] . Given the role of lung epithelial cells as an important source of chemokines ( including IL-8 ) , the inhibition of their activation by LT may have a critical impact on PMN recruitment , and therefore would compromise host defense against inhaled B . anthtacis . Indeed , a previous study showed that PMN play a role in host defense in pulmonary anthrax since PMN-depleted mice exhibited higher mortality by inhalation of B . anthracis [36] . Although lung epithelial cells could play a role during the activation of immune response in the early steps of anthrax infection , it should be , however , kept in mind that other pulmonary cell types such AMs , whose cytokine production is repressed by LT ( our unpublished data ) , may contribute to the inhibitory effect of LT on lung IL-8/KC production . In conclusion , we report here that B . anthracis represses the expression of cytokines such as IL-8 in epithelial cells . This inhibition occurs through a process involving LT-mediated epigenetic reprogramming , and might represent a novel mechanism by which this bacterium can evade the host innate immune response . This reprogramming would take place following spore germination and may compromise pulmonary host defense leading to bacterial proliferation and ultimate host death . Therefore , the use of pharmacological approaches to inhibit LT may represent a potential therapeutic strategy for the treatment of anthrax .
F-12K nutrient mixture , antibiotics , glutamine , Hanks' Balanced Salt Solution , Alexa Fluor® 488 F ( ab′ ) fragment of goat anti-rabbit IgG ( H+L ) were from Invitrogen ( Cergy-Pontoise , France ) . Fetal calf serum was from PAA ( Etobicoke , USA ) . ( TLRgrade™ ) LPS from E . coli ( Serotype 0111:B4 ) was from Alexis Biochemicals ( San Diego , USA ) . Human β-actin antibody was from Sigma Aldrich ( St . Louis , USA ) . SB203580 was from Tebu-bio ( Le Perray , France ) . Antibodies against histone H2B ( ab1790 ) , histone H3 ( ab1791 ) , acetyl-histone H3 ( H3AcK9 , ab4441 ) , methyl-histone H3 ( H3me3K4 , ab8580 ) and ( H3me3K9 , ab8898 ) and MSK-2 ( ab42101 ) were from Abcam ( Cambridge , USA ) . Antibodies against phospho-histone H3 ( H3pS10 , 05-817 ) and phospho-acetyl-histone H3 ( H3pS10-AcK14 , 07-353 ) were from Upstate Cell Signaling ( New York , USA ) . Antibodies against RNA Polymerase II ( RNAPII , sc-899 ) , MEK-1 ( sc-219 ) , MEK-3 ( sc-959 ) and the p65 subunit of NF-κB ( sc-372 ) were from Santa-Cruz ( Santa-Cruz , USA ) . Antibodies against MEK-2 ( 9125 ) , p38-MAPK ( 9212 ) , phospho-p38-MAPK ( p38pT180-Y182 , 9211 ) , ERK ( 9102 ) , phospho-ERK ( ERKpT202-Y204 , 9101 ) , JNK ( 9252 ) and phospho-JNK ( JNKpT183-Y185 , 4668 ) and PD98059 were from Cell Signaling ( Danvers , USA ) . Antibody against phospho-MSK-2 ( Ser196 ) , human and mouse recombinant TNFα were from R&D systems ( Minneapolis , USA ) . Recombinant Lethal Factor ( LF ) and Protective Antigen ( PA ) from B . anthracis were from List Biological lab ( Campbell , USA ) . Human bronchial epithelial cell line Beas-2B was from the American Type Cell Collection ( Rockville , USA ) . Mouse alveolar epithelial cell line MLE-15 was a gift from Dr . J . A . Whitsett ( Cincinnati Children's Hospital Medical Center , Cincinnati , USA ) . 7 weeks old C57/BL6 mice were supplied by the Centre d'Elevage R . Janvier ( Le Genest Saint-Isle , France ) . Mice were cared for in accordance with Pasteur Institute guidelines , in compliance with the European Animal Welfare regulations . The following isogenic non-capsulated B . anthracis strains were studied: the single mutant RP9 that produces active LF and inactive EF , and the double-mutant RPLC2 producing inactive LF and inactive EF . Both mutants produce functional PA [37] . Mice were anesthetized by i . p . injection of a mixture of ketamine-xylazine as previously described [38] before intranasal instillation of 50 µl of bacilli ( 107 cfu ) or LPS ( 330 µg/kg ) . In certain experiments , anesthetized mice received intratracheal instillation of 10 µl of LT ( 550 µg/kg = 550 µg PA+550 µg LF , per Kg ) 1 h before instillation of LPS or bacilli . Nascent bacilli were recovered from a 90 min growth in R medium as previously described [28] . Epithelial cells isolated from lung homogenates as previously described [39] , [40] , Beas-2B and MLE-15 cell lines were pretreated with LT ( PA+LF ) , PD98059 , SB203580 , or BAY for 1 h before incubation with LPS or TNFα at the concentrations and durations indicated in the figure legends . In other experiments , 1 h after addition of drugs or LT , Beas-2B cells were infected with bacilli for 3 h in the absence of antibiotics , washed and re-incubated for additional 24 h in a medium supplemented with antibiotics . IL-8 , KC , IL-6 and MIP-1 concentrations were measured in culture medium or in BALs using DuoSet ELISA kits from R&D systems ( Minneapolis , USA ) . Total cell counts in BALs were determined with a Coulter Counter from Coulter Electronics ( Margency , France ) , and differential cell counts were determined after cytospin centrifugation and staining with Diff-Quick products . Proteins were extracted from Beas-2B cells using RIPA and then 20 µg of total proteins were subjected to SDS-PAGE . Proteins were transferred to membrane from Millipore ( Billerica , USA ) , which were blocked and probed for 1 h with the indicated antibodies . After washing , the immunoreactive bands were visualized with a specific peroxidase-conjugated anti-immunoglobulin G ( IgG ) antibody and using an ECL Plus Western Blotting Detecting System ( Amersham , Biosciences ) . Nuclear proteins were extracted from Beas-2B cells , as previously described [41] . The NF-κB double-stranded oligonucleotides ( Santa Cruz Biotechnology ) 5′-AGT TGA GGG GAC TTTT CCC AGG C-3 γ-32P-labeled with T4 polynucleotide kinase ( Biolabs ) were incubated for 20 min at room temperature , with 5 µg of nuclear extract , 10 µl of binding buffer ( 40 mM HEPES ( pH 7 ) , 140 mM KCl , 4 mM DTT , 0 . 02% Nonidet P-40 , 8% Ficoll , 200 µg/ml BSA , 1 µg of poly ( dI:dC ) . Migration was performed on a 5% polyacrylamide gel in 0 . 5% Tris/borate/EDTA buffer at 150 V for 2 h . Gels were dried and exposed for 2 to 12 h . Beas-2B cells were fixed for 10 min with 3 . 7% ( weight/volume ) paraformaldehyde in PBS and then were permeabilized with 0 . 5% ( volume/volume ) Triton-X 100 in PBS . Fixed cells were treated 1 h with the indicated primary antibodies . Beas-2B cells were cultured in 24 well plates , at 70% confluence and transfected with an IL-8 promoter construct [133]-Luc , an [Δ NF-κB] IL-8 promoter construct mutated on NF-κB site or a NF-κB construct-Luc . After indicated treatment , luciferase activity was measured using a luciferase reporter assay kit , with signal detection for 12 s by a luminometer ( Berthold , Pforzheim , Germany ) , Total RNA was extracted using an RNeasy kit from Qiagen ( Courtaboeuf , France ) . DNase treatment was performed using 2 µg of extracted RNA , 1 µl of DNase I ( Amersham Biosciences , Orsay , France ) , and 0 . 5 µl of RNasin ( Promega , Madison , WI ) in a total volume of 20 µl in the manufacturer's buffer . cDNA were obtained by incubating RNA with 1 mM dNTP ( Eurobio , Les Ulis , France ) , 1 . 5 µl of hexamers as primers , 20 units of RNasin ( Promega , Madison , WI ) , and 300 units of Moloney murine leukaemia virus reverse transcriptase RNase H minus ( Promega , Madison , WI ) in a total volume of 50 µl of the manufacturer's buffer for 1 h at 42°C and 10 min at 70°C . Real-time PCR was performed using the SYBR Green kit ( Stratagene Brilliant II ) and analyzed using the MxPro software ( Stratagene , La Jolla , CA ) . The following primers were used . Proximal promoter of IL-8 ( ppIL-8 ) : sense 5′-AGTGTGATGACTCAGGTTTGCCCT-3; anti-sense 5′-AAGCTTGTGTGCTCTGCTGTCTCT-3′; proximal promoter of GAPDH ( ppGAPDH ) : sense 5′ TCCCATCACCATCTTCCAGG-3′ , antisens 5′-CATCGCCCCACTTGATTTTG-3′ . Proximal promoter of KC ( ppKC ) : sense 5′-GGAGCTCTGGAGTTTCGAGCATAA-3; anti-sense 5′-AGTCTGGAGTGCTGGAACTGGTTA-3′; KC-7 kb: sense 5′-TTCAAGCAGCCTCTCCCAGATCAA-3′ , antisense 5′-CATTTGCCAGTCCTTTGTGGCTGA-3′ . hCD44ex2: sense 5′-TGCCGCTTTGCAGGTGTATT-3′ , antisense 5′-GGCAAGGTGCTATTGAAAGCCT-3′ , mCD44ex2: sense 5′-TTTGAATGTAACCTGCCGCTACGC-3′ , antisense 5′-AGGTACTGTTGAAAGCCTGGCAGA-3′ . The PCR conditions were: 95°C 10 min , and 45 cycles at 94°C 15″ 60°C 30″ 72°C 20″ , followed by the dissociation curve program for the analysis of the PCR products . Formaldehyde-fixed human Beas-2B and mouse MLE-15 cells were extracted to remove unfixed components . The average size of sonicated chromatin was approximately 500 base pairs , allowing analysis of the PCR products ( 150 base pairs; data not shown ) . Sonicated chromatin corresponding to 5 . 106 cells was immunoprecipitated with 8 µg of irrelevant IgG or specific antibodies as indicated in corresponding figures . After extensive washing and reversed crosslinking , nucleic acids were used for quantitative real-time PCR , done with SYBR Green kits as detailed above . Cell viability was checked by the trypan blue dye exclusion test . Cell lysis was controlled measuring the release of lactate dehydrogenase ( LDH ) activity using a commercial kit from Boehringer ( Mannheim , Germany ) . Data are expressed as the means±S . E . M . and statistical analyses were performed using an unpaired Student's t test or an ANOVA test . Statistical significances are indicated as * P<0 . 05; ** p<0 . 01; *** p<0 . 001 .
|
Bacillus anthracis , the etiological agent of anthrax , can infect mammals either accidentally or as a potential consequence of a terrorism threat . Pulmonary infection is a life-threatening form of the disease , causing a near 100% mortality rate in the absence of appropriate therapy . Thus , it is important to understand the mechanisms of host defense against B . anthracis . We examined the effects of various B . anthracis strains on lung inflammation in a mouse model of pulmonary anthrax and on human lung epithelial cells , the first barrier of lung against invading pathogens . We showed that a B . anthracis strain expressing lethal toxin inhibits inflammation . In contrast , a strain in which this toxin has been inactivated induces lung inflammation . We next examined the mechanisms involved in the inhibitory effect of lethal toxin . We showed that B . anthracis injects lethal toxin into epithelial cells , blocks the molecules associated on the chromosome , and thus represses production of mediators involved in inflammation . As the latter is a key process in host defense , its alteration by lethal toxin predisposes the host to infection by B . anthracis . This effect on the chromosomal machinery may represent an efficient strategy used by B . anthracis for invading the host .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"immunology/cellular",
"microbiology",
"and",
"pathogenesis",
"molecular",
"biology/histone",
"modification",
"cell",
"biology/leukocyte",
"signaling",
"and",
"gene",
"expression",
"immunology/innate",
"immunity",
"infectious",
"diseases/respiratory",
"infections",
"infectious",
"diseases/bacterial",
"infections"
] |
2009
|
Anthrax Lethal Toxin Impairs IL-8 Expression in Epithelial Cells through Inhibition of Histone H3 Modification
|
Trypanosoma cruzi , the etiologic agent of Chagas Disease , is a major vector borne health problem in Latin America and an emerging infectious disease in the United States . We tested the efficacy of a multi-component DNA-prime/DNA-boost vaccine ( TcVac1 ) against experimental T . cruzi infection in a canine model . Dogs were immunized with antigen-encoding plasmids and cytokine adjuvants , and two weeks after the last immunization , challenged with T . cruzi trypomastigotes . We measured antibody responses by ELISA and haemagglutination assay , parasitemia and infectivity to triatomines by xenodiagnosis , and performed electrocardiography and histology to assess myocardial damage and tissue pathology . Vaccination with TcVac1 elicited parasite-and antigen-specific IgM and IgG ( IgG2>IgG1 ) responses . Upon challenge infection , TcVac1-vaccinated dogs , as compared to non-vaccinated controls dogs , responded to T . cruzi with a rapid expansion of antibody response , moderately enhanced CD8+ T cell proliferation and IFN-γ production , and suppression of phagocytes’ activity evidenced by decreased myeloperoxidase and nitrite levels . Subsequently , vaccinated dogs controlled the acute parasitemia by day 37 pi ( 44 dpi in non-vaccinated dogs ) , and exhibited a moderate decline in infectivity to triatomines . TcVac1-immunized dogs did not control the myocardial parasite burden and electrocardiographic and histopatholgic cardiac alterations that are the hallmarks of acute Chagas disease . During the chronic stage , TcVac1-vaccinated dogs exhibited a moderate decline in cardiac alterations determined by EKG and anatomo-/histo-pathological analysis while chronically-infected/non-vaccinated dogs continued to exhibit severe EKG alterations . Overall , these results demonstrated that TcVac1 provided a partial resistance to T . cruzi infection and Chagas disease , and provide an impetus to improve the vaccination strategy against Chagas disease .
American trypanosomiasis ( Chagas disease ) is a disease of humans and caused by the protozoan Trypanosoma cruzi of the family trypanosomatidae [1] . Approximately 30–40% of the infected patients develop a chronic debilitating illness of the cardiac system , characterized by clinically irreversible and progressive tissue destruction , and myocardial hypertrophy , eventually leading to heart failure and the patient’s death [2] , [3] . Several investigators have shown the potential utility of T . cruzi surface antigens as vaccine candidates in murine experimental models [4] , [5] ( reviewed in [6] , [7] ) . We have shown the protective efficacy of amastigote surface proteins ASP-1 and ASP-2 , and trypomastigote surface antigen TSA-1 as DNA vaccines in mice [8] . Vaccination with ASP-2 provided maximal immunity to T . cruzi infection in mice that was further enhanced by co-delivery of cytokine adjuvants [8] . In recent studies , we have identified additional potential vaccine candidates by computational screening of T . cruzi sequence database [9] . Of these , TcG1-TcG8 were phylogenetically conserved in clinically important strains of T . cruzi and expressed in the infective and intracellular stages of the parasite [9] . When delivered as a DNA vaccine in mice , TcG1 , TcG2 and TcG4 elicited a significant trypanolytic antibody response and Th1 cytokine ( IFN-γ ) response , a property associated with immune control of T . cruzi [10] . These novel vaccine candidates , thus , increased the pool of protective vaccine candidates against T . cruzi . In this study , we proceeded to examine the prophylactic and transmission-blocking efficacy of the multi-component vaccine constituted of TcG1 , TcG2 and TcG4 in dogs . We chose dogs for our studies because dogs provide an excellent model for studying the human disease [11]–[13] . Experimentally and naturally infected young dogs ( 2–3 months ) elicit reproducible and comparable acute infection associated with increase in blood parasitemia , IgG and IgM antibodies [14] , [15] and T cell response [16] . The presence of myocarditis with a moderate or small number of parasitized cells and extensive and frequent focal necrosis turns the disease in dogs similar to acute Chagas disease in humans [17] , [18] . A few infected dogs ( 10–20%; 5% humans ) develop severe acute myocarditis and may die of cardiac arrest . More than 80% of dogs recover from acute parasitemia as parasites become undetectable in the blood , and remnant mild myocardial changes with scattered microscopic foci of fibrosis and lymphocytic infiltration [19] present a picture similar to that of human infections [20] , [21] . At 12–18 months post-infection ( pi ) , ∼50% of infected dogs exhibit symptomatic chronic cardiac disease ( 20% develop severe myocarditis ) , associated with progressive cardiomegaly; arrhythmia , including RBBB with left anterior hemiblock [22] , [23]; diffused myocarditis with focal and interstitial fibrosis , and self-perpetuating myofibril destruction - a picture reminiscent of the chronic form of Chagas disease observed in humans . Thus , testing the vaccine efficacy in dogs would provide a strong basis for developing a human vaccine against T . cruzi and Chagas disease . Further , dogs are an important reservoir host for domestic transmission of T . cruzi . The prevalence rate of T . cruzi infection in dogs may reach up to 84% in endemic areas ( e . g . rural Argentina , Chiapas , Mexico ) , determined by serological procedures and xenodiagnosis [24] , [25] . Dogs are also the most frequent blood meal source for the domestic triatomines , i . e . , T . barberi and T . pallidipennis in Mexico [15] and T . infestans in Argentina [25] , [26] . Likewise , a high prevalence of seropositive dogs and infected triatomines is routinely noted in rural and urban development in the southern US [27] , [28] , and suggested to maintain T . cruzi transmission in the human habitat . Triatomines are several times more likely to take their blood meal from dogs than from humans [26] . The ratio of dog blood meals to human blood meals in the engorged guts of triatomines is estimated to be 2 . 3–2 . 6 times higher than the ratio of the number of dogs to the number of humans in a household [29] . Thus , the probability of infecting an insect in one blood meal from dogs is estimated to be 200-times higher as compared to the probability from adult humans [25] . These studies demonstrate that dogs are an important host blood source for domiciliary triatomines , and the risk of T . cruzi infection in humans is increased by the presence of infected dogs . Strategies that can limit T . cruzi infection in domestic reservoir host may , thus , prove to be effective in interrupting the parasite transmission to the vector , and consequently , to the human host . We immunized dogs with DNA-prime/DNA-boost vaccine ( TcVac1 ) . We examined the efficacy of TcVac1 in eliciting antigen-and parasite-specific antibody and T cell immunity , and determined if vaccination with TcVac1 modulated the host immune response towards protective type 1 upon T . cruzi infection . We also examined the efficacy of TcVac1 in controlling acute parasitemia , blocking the parasite transmission to triatomines , and preventing clinical severity of chronic disease .
Twelve mongrel dogs ( 6 males and 6 females , 3–4 months old ) were acquired locally and kept at the animal facility at the UAEM Research Center until they were included in the experiment , at eight months of age ( 8–12 kg body weight ) . Dogs were confirmed free of T . cruzi infection by microscopic examination of blood smears and serological evaluation of anti-T . cruzi antibodies using an indirect haemagglutination assay ( IHA ) and enzyme-linked immunosorbent assay ( ELISA ) [15] , [18] . Before inclusion in experimental studies , dogs were treated with anti-helminthes and vaccines against regional infectious diseases ( Canine distemper , Parvovirus infection , Canine hepatitis , Leptospirosis , and Rabies ) . All dogs received water ad libitum , and commercial dog food fed twice a day according to their age and development requirements . Experimental protocols were conducted under the technical specifications for the production , care , and use of lab animals from the Norma Official Mexicana ( NOM-062-ZOO-1999 ) , and the Council for International Organizations of Medical Science [30] , [31] . The research protocols were approved by the Laboratory Animal Care Committee at the Universidad Nacional Autonoma de Mexico . TcVac1 vaccine was constituted of antigen-encoding plasmids ( pCDNA3 . TcG1 , pCDNA3 . TcG2 and pCDNA3 . TcG4 ) and IL-12- and GMCSF-expression plasmids , described previously [8] , [32] . The eukaryotic expression plasmids encoding dog cytokines ( IL-12 and GM-CSF ) were a kind gift from Dr . Peter Melby [33] . All recombinant plasmids were transformed into E . coli DH5-α competent cells , grown in L-broth containing 100 µg/ml ampicillin , and purified by anion exchange chromatography using the Qiagen maxi prep kit ( Qiagen , Chatsworth , CA ) according to the manufacturer’s specifications . Trypomastigotes of T . cruzi ( SylvioX10/4 ) were maintained and propagated by continuous in vitro passage in C2C12 cells . Dogs ( n = 6/group , 3 males and 3 females ) were intramuscularly immunized with TcVac1 ( 200 µg each plasmid DNA/dog ) , delivered four-times at 2-week intervals . Dogs vaccinated with empty vector ( pcDNA3 only ) were used as controls . Two-weeks after the last immunization , dogs were challenged with culture-derived T . cruzi SylvioX10/4 ( 3 . 5×103 trypomastigotes/kg body weight , i . p . ) . The selected dose of the parasites was sufficient to produce acute parasitemia within 1–2 weeks of inoculation , and symptomatic clinical disease within 6–8 weeks post-infection [18] . Dogs were observed daily for general physical condition , at weekly intervals for clinical condition , and at 2-week intervals for cardiac function , monitored by electrocardiography ( EKG ) . Sera samples were obtained before each immunization and at two-week intervals thereafter . After challenge infection , in addition to sera samples , blood samples for parasitemia diagnostics were collected beginning day 5 pi , on alternate days up to 50 dpi and at two-week intervals thereafter . We measured blood parasitemia using hemacytometer counts of 5 µl blood mixed with equal volume of ACK red blood cell lysis buffer . Xenodiagnostic analysis was performed as described [25] , [34] , [35] . Briefly , stage 4 naive triatomine ( T . pallidipenis ) nymphs ( 6 per dog ) were fed on vaccinated and control dogs on day 30 and day 60 pi . Fecal samples were collected from triatomines at day 60 after feeding , and analyzed by light microscopy to detect epimastigote and metacyclic trypomastigotes . At least 10 microscopic fields were analyzed for each fecal sample , and triatomines were considered T . cruzi positive when >1 parasites were detected . The cDNAs for TcG1 , TcG2 and TcG4 were cloned in pET-22b plasmid ( Novagen ) such that the encoded proteins would be expressed in-frame with a C-terminal His-tag . All cloned sequences were confirmed by restriction digestion and sequencing at the Recombinant DNA Core Facility at UTMB . For the purification of recombinant proteins , plasmids were transformed in BL21 ( DE3 ) pLysS competent cells , and recombinant proteins purified using the polyhistidine fusion peptide-metal chelation chromatography system ( Novagen ) . Blood samples were obtained by venopuncture of the cephalic vein , and immediately processed to separate sera , using standard methods [15] , [18] . Sera samples ( 1∶50–1∶100 dilution ) were analyzed for IgM and IgG by using the Chagas diagnostic kits for ELISA ( Laboratorio-Lemos SRL , Buenos Aires , Argentina ) . The horseradish peroxidase ( HRP ) -labeled anti-human-IgG in ELISA kit was replaced with HRP-conjugated goat-anti-dog IgM- or IgG-specific secondary antibody ( Bethyl Laboratories ) [15] , [18] . In some experiments , instead of T . cruzi lysate , plates were coated with recombinant antigens ( TcG1 , TcG2 or TcG4 , 10-µg protein/ml ) to capture the antigen-specific antibodies . Sera samples from chronically infected dogs with confirmed T . cruzi infection and from healthy domestic dogs were used as positive and negative controls , respectively ( cut off value: ELISA , mean OD450nm from negative dogs ±2 SD; IHA , positive titer at ≥1∶16 serum dilution ) . To identify the antibody sub-types ( IgG1 and IgG2 ) , plates were coated with T . cruzi antigen , and , then sequentially incubated at room temperature with sera samples ( 1∶50 dilution ) for 2 h , biotin-conjugated goat anti-dog Ig subtypes ( IgG1 and IgG2 ) for 2 h , and streptavidin-horseradish peroxidase conjugate for 30 min . All antibodies and conjugates were from Bethyl Laboratories , and used at a 1∶3000 dilution in PBST-0 . 5% NFDM ( 100-µl/well ) . Color was developed with 100-µl/well Sure Blue TMB substrate ( Kirkegaard & Perry Labs ) , reaction was stopped with 2N sulfuric acid , and antibody response was monitored at 450 nm using a SpectraMax M5 microplate reader . Splenic level of CD4+ and CD8+ cell population in immunized/challenged dogs was determined by flow cytometry . Briefly , splenocytes were suspended in PBS ( 1×106 cells/100 µl ) and incubated for 30 min with FITC-conjugated anti-CD4 and PE-conjugated anti-CD8 antibodies ( 1∶50 dilution , from ABD Serotec ) . Following incubation , cells were fixed with 2% paraformaldehyde , washed and re-suspended in 500 µl PBS , and analyzed on a FACScan apparatus ( BD Biosciences ) . Cells stained with PE- and FITC- conjugated rat IgGs ( isotype matched ) were used as negative controls . Flow data were analyzed by Cell Quest software ( BD Biosciences ) . Cytokine levels in sera of vaccinated dogs were measured by sandwich ELISA . Briefly , 96 well plates were coated overnight with anti-IFN-γ or anti-IL-10 antibodies ( 500-ng/ml in PBS ) , washed with PBS/0 . 05% Tween-20 ( PBST ) , and incubated for 2 h with 1% BSA . Plates were then sequentially incubated at room temperature with sera samples ( 50-µl/well ) for 2 h , biotinylated anti-dog IFN-γ antibody ( 0 . 5-µg/ml ) or anti-dog IL-10 antibody ( 2-µg/ml ) for 2 h , and streptavidin conjugated horse radish peroxidase ( 1∶3000 dilution ) for 45 min . All antibodies were from R&D systems . Colorimetric reaction was performed as above . Cytokine concentrations were calculated using a standard curve derived using recombinant IFN-γ or IL-10 ( 1–4000 pg/ml ) . MPO activity was determined by a dianisidine-H2O2 method [36] , modified for 96-well plates [37] . Briefly , plasma samples ( 10-µg protein ) were added in triplicate to 0 . 53 mM o-dianisidine dihydrochloride ( Sigma ) and 0 . 15 mM H2O2 in 50 mM potassium phosphate buffer ( pH 6 . 0 ) . After incubation for 5 min at room temperature , the reaction was stopped with 30% sodium azide and the change in absorbance was measured at 460 nm ( ε = 11 , 300 M−1 cm−1 ) . Results were expressed as units of MPO/mg protein , whereby one unit of MPO was defined as the amount of enzyme degrading one n mol H2O2 per min at 25°C . The nitrite/nitrate content , indicative of inducible nitrite oxide synthase ( iNOS ) activity , was monitored by the Greiss reagent assay , as described [37] . In 96-well plates , reduced plasma samples ( 10 µg protein ) were mixed with 100 µl Greiss reagent , consisting of 1% sulfanilamide in 5% phosphoric acid and 0 . 1% N- ( 1-napthyl ) ethylenediamine dihydrochloride ( 1∶1 , v/v ) , and incubated for 10 min . The change in absorbance was monitored at 545 nm ( standard curve , 0–200 n mol sodium nitrite ) . Changes in cardiac rhythm and conduction in all dogs was monitored before inclusion in the study , and after challenge infection , at 2-week intervals up to 8-weeks and at monthly intervals thereafter . We used electrocardiograph ( Stylus , EK-8 , USA ) setting at 120 V , 60 Hertz , 20 amps , and 25 Watt in all experiments . Six leads of the electrocardiogram were considered at 25 mm/sec at 1-mV , standardized to 1 cm for the present study . Necropsy was performed the day animals died due to infection or after humanitarian sacrifice at day 60 ( acute phase ) and day 365 ( chronic phase ) post challenge infection . Dogs were sedated with xylazine ( 1–3 mg/kg body weight ) and then euthanized according to the Mexican Norma Official Mexicana [30] , [31] , using protocols approved by the Laboratory Animal Care Committee at the Universidad Nacional Autonoma de Mexico . A macroscopic and microscopic analysis of affected organs was performed . Postmortem studies were conducted using standard protocols with emphasis on macroscopic findings related to Chagas disease in heart tissue [15] . For histological analysis , tissue samples were fixed in 10% buffered formalin for 24 h , dehydrated in absolute ethanol , cleared in xylene , and embedded in paraffin . Tissue sections ( 5-µm thick ) were stained with hematoxylin-eosin , and evaluated by light microscopy at 100× and 400× [15] , [18] . Tissues were scored 0 to 4 in blind studies , according to the extent of inflammation and tissue damage from normal to total wall involvement [38] . Data are expressed as means ± SD , and derived from duplicate experiments ( n≥6 animals/group/experiment ) with at least duplicate observations per sample . Results were analyzed for significant differences using ANOVA procedures and Student’s t-tests . The level of significance was accepted at *p<0 . 05 ( vaccinated versus non-vaccinated ) .
The development of an antibody response induced by TcVac1 was determined by an ELISA . All dogs were seronegative before vaccination was initiated . The T . cruzi-specific IgM and IgG antibody response was detectable in sera ( 1∶50 dilution ) of vaccinated dogs after the first immunization , and moderately increased upon delivery of booster vaccine doses ( Fig . 1A&B ) . The level of antigen-specific antibody response was detected in the order of TcG1>TcG2>TcG4 , and was additive in nature ( Fig . 1E ) . The vaccine-induced antibody response was predominantly of the Th1 type with IgG2/IgG1 ratios >1 ( Fig . 1C&D ) . Control dogs immunized with plasmid vector alone exhibited no parasite- and antigen-specific antibody response ( Fig . 1 ) . After challenge infection with T . cruzi , sera samples were analyzed at 2-week intervals ( 1∶100 dilution ) ( Fig . 2 ) . Non-vaccinated/infected dogs exhibited a slight increase in parasite-specific IgM levels ( Fig . 2A ) . All dogs , irrespective of vaccination status , responded to T . cruzi infection by a gradual increase in anti-parasite IgG levels ( Fig . 2B ) . The TcVac1-immunized dogs exhibited a faster increase in IgG antibody response to T . cruzi infection ( Fig . 2B , p<0 . 05 ) as compared to that detected in non-vaccinated/infected dogs . Likewise , the vaccine-induced dominance of IgG2 antibodies ( compared to IgG1 subtype ) was significantly expanded after infection ( Fig . 2C&D ) . Together , the results presented in Fig . 1 and Fig . 2 suggested that vaccination of dogs with TcVac1 skewed the antibody response towards Th1 type that was further expanded upon challenge infection with T . cruzi . Next , we measured vaccine’s efficacy in activation of phagocytic ( neutrophils and macrophages ) response to T . cruzi by evaluating the plasma level of MPO activity and nitrite contents ( Fig . 3 ) . Vaccinated and non-vaccinated dogs exhibited no detectable level of MPO activity before challenge infection ( Fig . 3A ) . After exposure to T . cruzi , all dogs responded by a rise in MPO activity . During day 15–30 pi , non-vaccinated/infected and vaccinated/infected dogs exhibited a 2-fold and 25% increase in MPO activity in response to T . cruzi infection ( Fig . 3A ) . After day 30 pi , all dogs exhibited a similar decline in circulatory MPO activity . Likewise , the nitrite levels , indicative of iNOS activation and NO production , were increased by 2-fold in non-vaccinated/infected dogs at 30 dpi , while vaccinated/infected dogs exhibited ∼22% increase in plasma nitrite contents ( Fig . 3B ) . These data suggested that immunization with TcVac1 suppressed the T . cruzi-mediated activation of phagocytes evidenced by decreased plasma levels of MPO and nitrite in vaccinated/acutely-infected dogs as compared to non-vaccinated controls . A predominance of CD8+ T cells and type 1 cytokines ( IFN-γ ) is shown to be essential for control of T . cruzi infection [32] . All dogs , irrespective of vaccination regimen , responded to T . cruzi infection by a strong increase in parasite-specific lymphocyte activation ( Fig . 4 ) . The vaccinated/infected dogs exhibited a moderately stronger CD8+T cell response as compared to the non-vaccinated/infected dogs that was maintained during acute infection and chronic disease phase ( Fig . 4A ) . The circulatory cytokine levels ( IFN-γ and IL-10 ) were below detection limit before and after immunization with TcVac1 . The sera level of IL-10 remained undetectable after challenge infection with T . cruzi in all dogs . In comparison , all dogs responded to infection by a rapid increase in circulatory IFN-γ level that was significantly higher in vaccinated/infected dogs as compared to that noted in non-vaccinated/infected dogs ( Fig . 4B ) . These results indicated that TcVac1-immunized dogs were moderately better than non-vaccinated dogs in responding to T . cruzi infection by elicitation of higher level of type 1 biased CD8+T cell response . Detectable parasitemia that peaked during day 30–35 pi was noted in all dogs ( Fig . 5A ) . Dogs vaccinated with TcVac1 exhibited an early rise in parasitemia that was controlled by day 37 pi . In comparison , non-vaccinated/infected dogs exhibited a slight delay in peak parasitemia; however , blood parasitemia persisted beyond day 37 pi . No signs of clinical illness were apparent in vaccinated/infected and non-vaccinated/infected dogs during the physical exam , yet 33% of the TcVac1-vaccinated/infected dogs succumbed during 40–42 dpi ( Fig . 5B ) . Xenodiagnostic studies were performed to determine if dog’s infectivity to triatomines is altered by vaccination . Triatomines were fed on dogs during acute phase ( 30 dpi ) and after control of acute parasitemia ( 60 dpi ) , and feces were analyzed 30 days post-feeding for the detection of parasites by microscopy . In agreement with the peak parasitemia , all triatomines fed at day 30 pi on vaccinated and non-vaccinated dogs became T . cruzi positive . Of the 36 triatomines fed on each group of dogs at day 60 pi , 47% ( 17 out of 36 ) insects fed on TcVac1-vaccinated/infected dogs and 30% ( 11 out of 36 ) insects fed on non-vaccinated/infected dogs died during the incubation period . Of those surviving , we detected T . cruzi in feces of 52 . 63% ( 10/19 ) and 84 . 6% ( 21/25 ) of the insects fed on vaccinated/infected and non-vaccinate/infected dogs , respectively . These results indicated that TcVac1 was not effective in preventing infection or early rise in acute parasitemia , and was moderately effective in reducing the time-course of parasitemia and dogs’ infectivity to triatomines after day 37 post-infection . Normal electrocardiographic readings were noted in all dogs included in the study , before and after immunization . After challenge infection , vaccinated/infected and non-vaccinated/infected dogs exhibited no cardiac alterations up to 30 dpi . By day 60 pi , 67% of non-vaccinated dogs displayed electrocardiographic alterations , including reduced P-R interval , reduced R wave voltage , axis rotated to the right , S-T segment line elevation from the isoelectric line ( >0 . 2 mV ) , long QT segment , J wave elevation , and sinus tachycardia that were diagnostic of myocarditis , pericarditis , and high degree of myocardiocyte necrosis . Two vaccinated/infected dogs died by day 42 pi due to high electrical conductance problems and arrhythmia . Of remaining , 50% of the vaccinated/infected dogs exhibited at 60 dpi no electrocardiographic alterations , and other 50% exhibited a moderate level of EKG abnormalities including low voltage complex , a positive deviation of S-T with elevation of J-wave , left axel rotation , and tachycardia that were diagnostic of ventricular dilation , myocarditis , and arrhythmia . At one-year post-challenge infection , electrocardiographic analysis revealed a spectrum of cardiac dysfunction in chronically infected dogs . Among the unvaccinated/chronically-infected dogs , 66% exhibited major electrical conduction problems ( right axel rotation and LBBB ) , and 33% exhibited lateral re-polarization problems . Among the vaccinated/chronically-infected dogs , 33% exhibited electrical conduction problems ( right axel rotation and LBBB ) similar to that noted in unvaccinated/infected dogs , 33% showed minor axel rotation problems , and 33% dogs exhibited normal EKG . On a scale of 0 ( normal ) to 10 ( severe EKG alterations ) , 66% of non-vaccinated/chronically infected dogs were graded as 10 and 33% as normal ( zero EKG alterations ) . In comparison , 33% of vaccinated/chronically infected dogs were graded normal ( 0 ) , 33% moderate ( score: 5 ) , and 33% with severe electrical conduction problems ( score: 10 ) at one year post-infection . Next , we evaluated the pathology of the heart in dogs . Anatomo-pathological analysis of the heart , performed at day 60 pi , showed dilated cardiomyopathy ( bi-ventricular dilation ) and focal and diffused myocarditis in vaccinated dogs as well as in dogs injected with vector alone ( Fig . 6 ) . Irrespective of vaccination status , some animals exhibited whitish zones and rounded edges in the spleen , dilation of esophagus , and pinkish ampoules at the cecum . At one-year post-challenge infection , all dogs had round shaped hearts . Sixty six percent of non-vaccinated/chronically infected dogs exhibited severe right ventricle dilation . In comparison , 66% of vaccinated/chronically-infected dogs exhibited moderate level of right ventricle dilation . Epicardial hemorrhages were seen in 66% dogs from the control group and 33% dogs in the TcVac1 group . Histopathology studies on day 60 pi demonstrated some differences between two groups ( Fig . 7 ) . In epicardium of dogs vaccinated with TcVac1 , non-suppurative moderate to severe myocarditis with focal or zonal mononuclear and polymorphonuclear inflammatory infiltrate associated with presence of amastigotes nests and severe active necrosis was generally noted . In non-vaccinated/infected dogs , histopathological findings were similar to that noted in vaccinated dogs with the exception that inflammatory infiltrate tended to be mainly constituted of mononuclear rather than polymorphonuclear cell type ( Fig . 7 ) . In the myocardium , vaccinated and non-vaccinated dogs exhibited multiple coagulative necrosis foci with mononuclear infiltration and some necrotic areas with polymorphonuclear and neutrophil infiltration . The diffused multi-focal and zonal mononuclear inflammatory infiltrate and hemorrhagic areas in myocardium appeared to be larger in vaccinated/infected dogs ( Fig . 7C , C2 ) as compared to non-vaccinated/infected dogs ( Fig . 7B , B2 ) . Vaccinated/infected dogs also showed abundant cellular detritus in myocardium . Mononuclear infiltration was moderate in ventricles and septum of all dogs . Folding of myocardial fibers in right ventricle was observed in both vaccinated/acutely-infected and non-vaccinated/acutely-infected dogs . Mural multifocal endocarditis with mononuclear infiltration was also noted in dogs from the two groups . Abundant amastigote nests ( range 18–21 per microscopic field ( mf ) ) were found in each region ( right and left ventricles , and septum ) of the heart of TcVac1-vaccinated/acutely-infected dogs . Non-vaccinated/infected dogs exhibited , in general , lesser number of parasite foci in equivalent studied areas . Statistical analysis showed that overall , the number of myocardial necrotic foci , lymphocyte infiltration foci and number of amastigote nests were more abundant in TcVac1-vaccinated dogs than was observed in non-vaccinated/infected dogs at 60 dpi ( p<0 . 05 ) ( Fig . 7 ) . At one-year post-challenge infection , severe myocardial inflammation persisted in 66% of the non-vaccinated/infected dogs while remaining 33% exhibited slight inflammatory infiltrate in the heart . In comparison , 66% of vaccinated/chronically infected dogs exhibited moderate level of myocardial inflammatory infiltrate . Slight to moderate presence of connective tissue was apparent in all chronically infected dogs; however , it was more evident in TcVac1-vaccinated/chronic dogs . Folding of myocardial fibers and vacuolization of Purkinje fibers was observed in 33% of vaccinated/chronically infected dogs .
The objective of the present study was to test the efficacy of a multi-component DNA vaccine ( TcVac1 ) in dogs . The antigenic candidates included in TcVac1 were identified by computational analysis of T . cruzi sequence database and selected because they were conserved among several clinically relevant T . cruzi strains , expressed in infective and intracellular stages of T . cruzi [9] , and recognized by the antibody and T cell response in infected mice [32] and humans ( unpublished data ) . When delivered as a DNA vaccine in mice , TcG1 , TcG2 and TcG4 elicited trypanolytic antibody response and Th1 cytokines ( e . g . IFN-γ ) [9] that resulted in significant protection from acute infection and chronic disease severity . We utilized IL-12 and GM-CSF expression plasmids as adjuvants as these cytokines induce type 1 B and T cell responses [33] , and shown to significantly enhance the protective immunity elicited by the vaccine candidates in mice [8] , [39] and dogs [33] . To the best of our knowledge , this is the first report testing the prophylactic and transmission-blocking efficacy of DNA vaccine against T . cruzi in dogs . Immunization of dogs with TcVac1 resulted in elicitation of antigen-specific and parasite-specific antibody response that was dominated by IgG2 subtype . The delivery of booster doses of vaccine resulted in no significant increase in antibody response that could be explained , at least partially , by the fact that DNA delivery system , used in this study , is known to drive a low level of antigen expression . Several investigators have reported that needle delivery of DNA vaccines in muscle induce low immune response in large animals and humans , even when 1000-fold higher doses of DNA than those proved to be effective in rodents were given [40] , [41] . Other DNA vaccine delivery systems such as gene gun ( biolistic gun ) [42] , [43] , adenovirus or vaccinia virus delivery vectors [44] , replicating attenuated strains of intracellular microorganisms , such as Salmonella [45] have shown promising results in eliciting antigen expression . Additionally , heterologous prime/boost approaches are noted to be more effective in eliciting stronger , long-term immunity against intracellular pathogens [46] , [47] , to be tested in future studies . Despite low vaccine-induced antibody response , vaccinated dogs , upon challenge infection with T . cruzi , exhibited an early expansion in antibody response . The Ig ( G+M ) response during acute phase was of higher magnitude in vaccinated/infected dogs than that observed in dogs injected with vector alone . The IgG response in vaccinated/infected dogs was primarily of the Th1 type with IgG2/IgG1 ratios being >1 , known to provide protection from acute infection in dogs [48] , [49] . The higher level and rapid expansion of antibody titers indicates that TcVac1 primed the B cell response that was expanded upon exposure to T . cruzi . Previously , we have shown that TcG1- , TcG2- and TcG4-specific antibodies , elicited in vaccinated mice , were lytic in nature , and efficiently killed trypomastigotes in a complement-dependent manner [9] . In this study , our observation of a shorter detectable parasitemic period of 37 days in vaccinated/infected dogs than that noted in control dogs ( 44 days ) indicate that antibody response primed by vaccination with TcVac1 was lytic in nature and contributed to a control of blood parasitemia . Yet , immunization with TcVac1 failed to prevent peak parasitemia . This was likely because other components of immune system , i . e . , innate response constituted by phagocytes and type 1 biased CD8+ T cell response , were not strongly primed by vaccine or expanded upon challenge infection in vaccinated dogs . It is well documented that phagocytes , through activation of NADPH oxidase , MPO , and iNOS activities and production of cytotoxic reactive oxygen and nitrogen species , play an important role in control of T . cruzi [37] , [50]-[52] . Numerous studies have also demonstrated that an efficient control of acute parasitemia requires concerted activities of Th1 helper cells , and cytotoxic CD8+ T lymphocytes ( CTLs ) ( reviewed in [53] , [54] ) . Vaccination with TcVac1 resulted in a suppression of phagocytes’ response to challenge infection as was evidenced by decreased activation of MPO and iNOS activity in vaccinated/infected dogs as compared to that noted in non-vaccinated/acutely infected dogs . Equally , immunization with TcVac1 resulted in a significant but only moderately better expansion of CD8+ T cells and IFN-γ levels upon challenge infection when compared to that noted in non-vaccinated/infected dogs . Consequently , it was not surprising to find no significant decline in infectivity of vaccinated dogs to triatomines during the acute period of infection . All triatomines , fed on vaccinated/infected or non-vaccinated/infected dogs when dogs were exhibiting peak parasitemia ( day 30 pi ) , and analyzed at day 60 post-feeding , were infected evidenced by fecal presence of T . cruzi . However , at 60 dpi , vaccinated dogs exhibited a better control of parasitemia and moderately reduced infectivity to triatomines ( 52 . 63% versus 84 . 6% infected ) . The mathematical modeling of transmission dynamics [55] and other studies using insecticide-treated dog collars [56] , [57] indicate that a decline in infectivity to <20% would be required to block vectorial transmission of T . cruzi to humans . Thus , we surmise that current formulation of TcVac1 , though provided a decline in dogs’ infectivity to triatomines after peak parasitemia , would not be effective in blocking the transmission cycle , and further improvement in vaccination strategy is required . Infection of dogs with SylvioX10/4 strain of T . cruzi produced reproducible acute phase and chronic pathology as we have previously reported [18] , causing sudden death in some of the infected dogs , and cardiomyopathic changes in most of the infected animals during acute stage . EKG alterations were found in more than half of the acutely infected dogs and ranged from electrical conduction problems , ventricular dilatations , pericarditis , myocarditis , high lateral necrosis , and arrhythmia . Most of these changes were validated by necropsy and histopathology findings , thus , confirming that this T . cruzi strain is highly pathogenic in dogs . Despite a similar or higher infiltration of inflammatory infiltrate in the heart , vaccinated dogs exhibited a significantly higher number of amastigote nests ( P<0 . 05 ) in cardiac tissue than was observed in non-vaccinated control dogs ( Fig . 7 ) . Others have reported a direct correlation between in vitro infectivity and blood parasitism kinetics with heart parasitism intensity during long-term infection of Beagle dogs [58] , [59] . Because of high inflammatory infiltrate and tissue parasite burden , two of the vaccinated dogs exhibited myocarditis and died suddenly due to arrhythmia . Chronically infected/vaccinated dogs were better equipped in controlling the disease symptoms . EKG findings demonstrated mild-to-moderate cardiac alterations in animals given TcVac1 vaccine while severe EKG alterations persisted in dogs injected with vector alone . These findings were supported by anatomo-pathological analysis performed at one-year post-challenge infection . Anatomo-pathological lesions and epicardial hemorrhages were fewer and moderate in TcVac1-vaccinated/chronically infected dogs as compared to non-vaccinated/chronic dogs that exhibited severe right ventricle dilation and extensive epicardial hemorrhages . These findings were observed despite no decline in inflammatory infiltrate in the heart in chronically infected dogs . These data indicate that TcVac1-induced immunity was at least partially effective in controlling the clinical progression of cardiac disease severity in chagasic dogs . Summarizing , in this study , we tested a multi-component DNA vaccine against T . cruzi infection in dogs . Our data showed that TcVac1 geared a modest parasite- and antigen-specific type 1 antibody and CD8+ T cell response that was effective in providing an early control of acute parasitemia and moderately decreased the infectivity of dogs to triatomines . However , tissue parasite burden was not controlled in vaccinated dogs , likely due to suppression of phagocytic cell response , evidenced by decreased myeloperoxidase and nitrite ( iNOS ) levels in immunized dogs . Despite this , vaccinated dogs exhibited a moderate decline in cardiac alterations determined by EKG and anatomo-/histo-pathological analysis during chronic stage of disease development . Overall , our data demonstrated that TcVac1-elicited immunity provided a partial protection from chronic Chagas disease and provided an impetus to further improve the vaccination strategy against Chagas disease .
|
Immunization of dogs with DNA-prime/DNA-boost vaccine ( TcVac1 ) enhanced the Trypanosoma cruzi-specific type 1 antibody and CD8+ T cell responses that resulted in an early control of acute parasitemia and a moderate decline in pathological symptoms during chronic phase . Further improvement of vaccine-induced immunity would be required to achieve clinical and epidemiological benefits and prevent transmission of parasites from vaccinated/infected dogs to triatomines .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"cardiovascular",
"disorders/myopathies"
] |
2011
|
Testing the Efficacy of a Multi-Component DNA-Prime/DNA-Boost Vaccine
against Trypanosoma cruzi Infection in Dogs
|
In the last two decades , chikungunya virus ( CHIKV ) has rapidly expanded to several geographical areas , causing frequent outbreaks in sub-Saharan Africa , South East Asia , South America , and Europe . Therefore , the disease remains heavily neglected in Mozambique , and no recent study has been conducted . Between January and September 2013 , acute febrile patients with no other evident cause of fever and attending a health center in a suburban area of Maputo city , Mozambique , were consecutively invited to participate . Paired acute and convalescent serum samples were requested from each participant . Convalescent samples were initially screened for anti-CHIKV IgG using a commercial indirect immunofluorescence test , and if positive , the corresponding acute sample was screened using the same test . Four hundred patients were enrolled . The median age of study participants was 26 years ( IQR: 21–33 years ) and 57 . 5% ( 224/391 ) were female . Paired blood samples were obtained from 209 patients , of which 26 . 4% ( 55/208 ) were presented anti-CHIKV IgG antibodies in the convalescent sample . Seroconversion or a four-fold titer rise was confirmed in 9 ( 4 . 3% ) patients . The results of this study strongly suggest that CHIKV is circulating in southern Mozambique . We recommend that CHIKV should be considered in the differential diagnosis of acute febrile illness in Mozambique and that systematic surveillance for CHIKV should be implemented .
Chikungunya virus ( CHIKV ) is an arthropod borne virus ( arbovirus ) transmitted by Aedes mosquitoes and belonging to the Togaviridae family and alphavirus genus . Clinical presentation of CHIKV disease ranges from a self-limiting and undifferentiated febrile illness accompanied by exanthema , myalgia and headache to severe and debilitating polyarthritis and encephalitis . In a few cases death may occur [1–3] . CHIKV was described for the first time in 1952 during an outbreak in small villages in southern Tanzania close to the border with Mozambique . Cases were also reported in a few towns close to the border between Mozambique and Tanzania [2 , 4 , 5] . In recent years , the global health importance of CHIKV has increased significantly since the virus is a leading emerging vector borne infections worldwide [6 , 7] . Recently , several outbreaks have been reported in sub-Saharan Africa and South East Asia , including in Europe [1 , 2 , 7–10] . Outbreaks of chikungunya in temperate countries such as Italy represent a paradigm shift of mosquito-transmitted diseases [6 , 11] . The recent emergence of this virus in South America [12–15] , made CHIKV probably the second most widespread arbovirus following dengue viruses . The main reasons for the resurgence of CHIKV worldwide include global warming , intense commercial trading , deforestation , and changes in the ecology and geographical distribution of Aedes mosquitoes [6 , 7] . Although the initial discovery of CHIKV is geographically linked to Mozambique[5] , the disease has been heavily neglected locally for the followings reasons: i ) non-specific clinical presentation , ii ) lack of local diagnostic capacity for CHIKV confirmation and iii ) lack of epidemiological data on the chikungunya burden . The scarce available information of CHIKV in Mozambique is more than 40 years old [16] . Due to the lack of recent serological and epidemiological data , Mozambique has been repeatedly excluded from the list of affected countries in reports describing the global distribution of CHIKV [1 , 17 , 18] , and consequently , the country is frequently considered free of this virus . Evidence of CHIKV in the neighboring countries is also very scarce , and most of available data are old [19–22] . Mozambique is located on the south-eastern coast of Africa , with more than 2 , 500 Km of coast and represents a strategic hub for the region . The concern that CHIKV would currently constitute an important cause of acute febrile illness in Mozambique has recently increased since recent entomological research conducted and published by our group demonstrated an abundance of Aedes aegypti , the vector for CHIKV in several geographical areas in Mozambique , including Maputo city in southern Mozambique where no previous reports of CHIKV exist [23] . Furthermore , Aedes was involved in the 2014 outbreak of dengue in two provinces in northern Mozambique [24] and a recent case of concomitant Plasmodium falciparum and CHIKV infections in an adult patient living in northern Mozambique was laboratory confirmed in a private clinic in South Africa [25] . In addition , the largest CHIKV outbreak reported in history occurred during 2005–2007 in several islands situated in the Indian Ocean and Mozambique Channel , geographically close to Mozambique [26 , 27] and recently , several cases of CHIKV have been reported in Tanzania[28 , 29] and Angola[30] , respectively . CHIKV may also have an impact on efforts to control malaria and other vector borne acute febrile illnesses transmitted by Aedes , such as dengue , in countries endemic for these diseases , including Mozambique , as the clinical presentation of CHIKV overlaps those of other febrile conditions . In these settings , CHIKV may circulate unsuspected , leading to over-diagnosis of malaria and overuse of antimalarial drugs[31] . Although the available evidence strongly suggests that CHIKV circulates in Mozambique , the information on the epidemiology of the disease is dated as well as insufficient , and no study has as yet been conducted in southern Mozambique . This study was therefore conducted with the aim to provide recent information on the frequency of CHIKV in acute febrile patients in southern Mozambique .
This study was conducted at the Mavalane Health Center , a primary health care facility located in a large suburban area in Maputo City in southern Mozambique ( see Fig 1 ) . The study area is characterized by poverty , poor sanitation , and an extremely high population density . The prevalence of malaria is high in this setting . The main sources of income are from the informal sector and small business . The rainy season extends from November through March . Eligibility criteria included presence of acute febrile syndrome ( axillary temperature > 37 . 5°C ) , age > 5 years and ability to provide a written informed consent . Exclusion criteria . Pregnant women , individuals with psychiatric disease , individuals with a readily identifiable focus of infection , such as otitis media , sinusitis , purulent pharyngitis , cellulites , urinary tract infection , dental abscess , septic arthritis , pneumonia or pelvic inflammatory disease were excluded from participation in this study All eligible patients attending the Mavalane Health Center between January and September 2013 were consecutively invited to participate and enrolled . At the enrollment visit , a blood sample ( acute sample ) was requested from each patient and all of them were requested to return to the health facility after three weeks for collection of a convalescent sample . A few days before the scheduled date for the convalescent visit , the investigator made phone call to remind each participant of their appointment . Patients with malaria positive test results were not excluded . The WHO guidelines [32] for a laboratory confirmed CHIKV infection include the isolation of virus as well as the demonstration of virus specific IgM antibodies or a four-fold titer rise of IgG in samples collected at least three weeks apart . Patients with anti-CHIKV IgG in both acute and convalescent samples , but lacking a four-fold increase in titer were classified as having previous exposure to CHIKV . The absence of anti-CHIKV IgG was defined as no CHIKV infection . This study was approved by the National Bioethical Committee in Mozambique ( Ref: IRB00002657 ) prior to initiation . Written and informed consent was requested from each patient before enrollment . From each participant who gave the written consent to participate , clinical , demographic and epidemiological data was collected using a questionnaire . The collected information included age , gender , residence and clinical presentation . Venous blood was collected from all acute febrile patients who consented to participate in the study . Whole blood was collected aseptically; 5ml into a K3EDTA tube and 5ml into a Serum Separation Tube ( both from BD Vaccutainer , USA ) . Specimens were delivered to the laboratory within four hours of collection for centrifugation , separation and storage at -70°C . Sample size to estimate seroprevalence of Chikungunya ( IgG ) in febrile patients was calculated based on the following assumptions , i ) expected seroprevalence ( frequency of IgG anti-CHIKV ) of 15% using as the basis for comparison , the seroprevalence reported in recent studies conducted in Tanzania ii ) precision rate of 3 . 5% . With these assumptions , the final size of the sample was 400 febrile patients . Data was analyzed using the statistics package STATA 9 . 0 ( College Station , Texas: StataCorp , USA , 2005 ) . The Kruskal Wallis test was used to compare the study groups regarding numerical variables . Associations between categorical variables were determined using the Pearson Chi-square test . Logistic regression analysis was employed to determine the variables independently associated with CHIKV antibodies , controlling for confounders . The analysis was built using the backwards stepwise method and Log Likelihood Ratio Test . For this purpose , all variables with a P-value less than 0 . 1 on univariate analysis were included in the initial multivariate model . A p value < 0 . 05 was considered of statistical significance in the final model .
Four hundred patients were enrolled between January and August 2013 . Of these , 209 ( 52 . 3% ) returned for the follow-up visit , although efforts were undertaken by the investigators to recall each participant by phone call to return for their convalescent follow up ( see Fig 2 ) . The reason why 47 . 7% of patients did not return for their convalescent visits are unknown , but information provided by clinicians mention that after recovery patients rarely return to the health facility for convalescent follow up . The median age of study participants was 26 years ( IQR: 21–33 years ) and 57 . 5% ( 224/391 ) were female . The average time between onset of symptoms and collection of the acute sample was 24h . In regard to convalescent sample , the average number of days from disease onset was 25 days . Convalescent samples from the 209 patients who returned for the follow-up visit were tested for anti-CHIKV IgG using IIFA , 55 ( 26% ) of which were positive , 153 ( 74% ) tested negative and one sample was not tested due to insufficient volume . Subsequently , the corresponding acute samples ( n = 55 ) from the patients who tested positive in their convalescent sample were tested for anti-CHIKV IgG . Seroconversion or a four-fold titer rise representing the definition criteria for acute CHIKV infection , occurred in nine ( 4 . 3% ) patients . Anti-CHIKV IgM testing was performed in these 9 serum samples and all were negative . Virus isolation attempts from six CHIKV IgG negative samples were unsuccessful . Fig 2 demonstrates that anti-CHIKV IgG measurement was also performed in acute samples from patients who did not return for their follow-up visit and the positivity rate was similar to that observed for convalescent samples ( 24 . 4% , 38/156 for acute sample of patients who did not return for the follow-up visit versus 26 . 4% , 55/208 for convalescent samples from patients who returned for the follow- up visit ) . Results of malaria blood smears were available for 328 patients , of which 26 were smear positive , yielding an overall positivity rate for malaria of 7 . 9% . Based on IIFA for anti-CHIKV IgG , patients who returned for the follow-up visit were stratified into three main groups , i . e . , i ) acute infection ( 9/209 , 4 . 3% ) , ii ) previous exposure ( 44/209 , 21 . 2% ) and iii ) negative CHIKV infection ( 153/209 , 73 . 2% ) . Febrile patients with acute infection , previous exposure and negative CHIKV infection were similar in terms of age , gender and clinical presentation as shown in Table 1 . A few participants had a history of recent international travel , and the destination was South Africa for all of them . Arthralgia was de most common symptom in the patients with CHIKV infection , but in the other two groups , headache was the more common symptom . No patients presented any form of hemorrhage .
In Mozambique and other malaria endemic countries , the lack of epidemiologic information concerning the etiology of acute febrile illness results in over-diagnosis and over-treatment of malaria and increased mortality related to wrong therapeutic intervention[31] . Although available evidence from other sub-Saharan countries close to Mozambique[1] , strongly suggests that CHIKV would be among the main causes of acute febrile illness , this virus is heavily neglected in Mozambique and no data has been published reporting CHIKV in Mozambique over the last 40 years . This study was therefore conducted with the aim to determine the burden of CHIKV in acute febrile patients in Mozambique and represents the first sero-epidemiological investigation of CHIKV in southern Mozambique . Methodologically , this study represents an improvement over the few available studies conducted more than 40 years ago[33] , since , for the first time , paired acute and convalescent samples were collected from each participant and tested for CHIKV antibodies . Patients with acute febrile illness were invited to participate in this study and their samples were serologically screened for CHIKV IgG antibodies . We used the WHO criteria to define recent infection based on measurement of CHIKV IgG antibodies in acute and convalescent sample , as previously mentioned . CHIKV IgG seroconversion and titer rises were observed in 4 . 3% ( 9/208 ) of patients . This finding is similar to that of a recent study conducted in Tanzania , which found a frequency of acute CHIKV infection of 4 . 3% among febrile patients , but is slightly lower of than reported in another study conducted in northern Tanzania , which found a frequency of acute infection of 7 . 9%[34] . Altogether , these findings reinforce our previous hypothesis that CHIKV is prevalent in Mozambique[28] . Our result is not surprising since a recent paper published by our group demonstrated that Aedes mosquitoes is abundant in Maputo city . In our study , we demonstrated that Aedes was present in 66 . 4% ( 83/125 ) of mosquito reservoirs that we observed[23] . Furthermore , CHIKV has been involved in outbreaks in Southern Africa [1 , 28] and the largest outbreak reported in history occurred in several islands situated in the Indian Ocean and Mozambique Channel , close to Mozambique [26 , 27] . In addition , a recent case of co-infection by malaria and CHIKV in an adult patient living in Pemba , situated in the northern Mozambique , was recently confirmed at a private laboratory in South Africa[25] . We found a frequency of previous exposure to CHIKV in 24 . 6% ( 55/208 ) of patients who returned for a convalescent visit ( see Fig 2 ) . This demonstrates that exposure to CHIKV is common and strongly suggests that CHIKV would be an important cause of acute fever , causing unsuspected sporadic cases or outbreaks and consequently may have been overlooked and mis-diagnosed as malaria . We hypothesized that this not only hampers the national efforts to control malaria , but also contributes to the increased risk of mortality related to the administration of incorrect treatment . Although this study was conducted in a small geographical area in southern Mozambique , we can speculate that CHIKV is also circulating in other geographical areas of the country , since a recent entomologic investigation demonstrated that the density of Aedes mosquitos in two provinces situated in northern Mozambique was much higher than in Maputo[23] and that previous reports demonstrated that CHIKV caused outbreaks or sporadic cases in northern Mozambique[5 , 33] . Moreover , recent papers reported circulation of CHIKV in three different regions in Tanzania , a neighboring country in the northern Mozambique[28 , 29 , 34] . Data from our study show that the clinical presentation of patients with acute CHIKV infection was similar of that from patients with previous exposure or with no CHIKV infection . These findings are similar to those from previous studies in the Sub-Saharan region of Africa[27] . In order to understand the burden of malaria versus CHIKV , patients suspected of malaria were not excluded from this study and our results demonstrated that P . falciparum was confirmed in 7 . 9% ( 26/302 ) of febrile patients . This highlights the concern that in several settings in Mozambique and other sub-Saharan countries , malaria is no longer the main cause of febrile disease , in a context where the burden of other emerging diseases is increasing [35] . The serious lack of epidemiological data regarding the etiology of acute fever in Mozambique and other sub-Saharan countries represents a serious challenge for the proper management of these patients and is considered the main cause of over-diagnosis and over-treatment of malaria[31] . Further studies should be conducted in order to better understand the epidemiology of acute fever and improve the algorithms for management of acute febrile illness , not only in Mozambique , but in other sub-Saharan countries . Acute samples from patients with acute CHIKV infection were negative for anti-IgM CHIKV . This was not a surprising finding as acute samples were in average collected 1 day after onset of symptoms , and thus likely before sero-conversion of IgM . Virus isolation assay was performed on six acute samples from patients with CHIKVIgG seroconversions , but all were negative . This was also not surprising , as serum samples were poorly managed during transportation and storage , which may have resulted in viral RNA degradation . We acknowledge the limitations of our study , such as the rate of loss of participants for the follow up visit . Unfortunately , it is well known that acute febrile patients rarely return for convalescent visits , since most febrile diseases are of short duration , and after recovery the patients return to their routine activities with no interest in follow up visits . Serological cross-reactivity between arboviruses has been considered a limitation in several serological assays . In this regard , ten samples reactive for IgG anti-CHIKV were also tested for Sindbis virus IgG antibodies using an in-house ELISA , and all were non-reactive . In addition , the manufacturer claims that the specificity of EUROIMMUNE immunofluorescence assay is very high ( 100% ) . In conclusion , our data demonstrate that CHIKV infections are more frequent than thought , and thus we recommend that: i ) chikungunya should be considered as a differential diagnosis of acute febrile illness to reduce malaria over-diagnosis and over- treatment with antimalarial drugs , ii ) sentinel surveillance systems for CHIKV should be implemented and expanded to other regions in the country and iii ) further entomological studies should be conducted to better understand the distribution of the CHIKV vector in Mozambique .
|
Chikungunya virus ( CHIKV ) is an emerging arbovirus that remains heavily neglected in Mozambique , and no recent study has been conducted . Between January and September 2013 , four hundred acute febrile patients with no other evident cause of fever and attending a health center in a suburban area of Maputo city , Mozambique , were consecutively invited to participate . Paired acute and convalescent serum samples were drawn from each participant . Convalescent samples were initially screened for anti-CHIKV IgG , and if positive the corresponding acute sample was screened using the same test . Of the 209 patients from which paired samples was obtained , 26 . 4% ( 55/208 ) presented anti-CHIKV IgG antibodies in the convalescent sample . Seroconversion or a four-fold titer rise was confirmed in 9 ( 4 . 3% ) patients . Overall our findings demonstrate that CHIKV is circulating in southern Mozambique and suggest that CHIKV should be considered in the differential diagnosis of acute febrile illness .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Serological Evidence of Chikungunya Virus among Acute Febrile Patients in Southern Mozambique
|
Candida auris is an emerging , multi-drug resistant , health care-associated fungal pathogen . Its predominant prevalence in hospitals and nursing homes indicates its ability to adhere to and colonize the skin , or persist in an environment outside the host—a trait unique from other Candida species . Besides being associated globally with life-threatening disseminated infections , C . auris also poses significant clinical challenges due to its ability to adhere to polymeric surfaces and form highly drug-resistant biofilms . Here , we performed bioinformatic studies to identify the presence of adhesin proteins in C . auris , with sequence as well as 3-D structural homologies to the major adhesin/invasin of C . albicans , Als3 . Anti-Als3p antibodies generated by vaccinating mice with NDV-3A ( a vaccine based on the N-terminus of Als3 protein formulated with alum ) recognized C . auris in vitro , blocked its ability to form biofilms and enhanced macrophage-mediated killing of the fungus . Furthermore , NDV-3A vaccination induced significant levels of C . auris cross-reactive humoral and cellular immune responses , and protected immunosuppressed mice from lethal C . auris disseminated infection , compared to the control alum-vaccinated mice . The mechanism of protection is attributed to anti-Als3p antibodies and CD4+ T helper cells activating tissue macrophages . Finally , NDV-3A potentiated the protective efficacy of the antifungal drug micafungin , against C . auris candidemia . Identification of Als3-like adhesins in C . auris makes it a target for immunotherapeutic strategies using NDV-3A , a vaccine with known efficacy against other Candida species and safety as well as efficacy in clinical trials . Considering that C . auris can be resistant to almost all classes of antifungal drugs , such an approach has profound clinical relevance .
The fungus Candida auris was first detected in 2009 from an ear canal infection in Japan [1] . However , the earliest known strain of C . auris dates back to 1996 isolated in a retrospective analysis of previously misdiagnosed samples from Korea [2] . Since then , C . auris has been reported in more than 20 countries , with a significant number of cases detected in the Unites States [3] . Patients can remain colonized with C . auris for long periods and the yeast can persist on surfaces in healthcare environments , which results in spread of the organism between patients in healthcare facilities [4 , 5] . Clinical isolates of C . auris have been recovered from a variety of specimen types , including normally sterile body fluids , wounds , mucocutaneous surfaces , and skin [4] . However , bloodstream infection remains the most commonly observed clinical manifestation of C . auris , with alarming in-hospital global crude mortality rates of 30 to 60% [6 , 7] . Further and of high importance , some isolates of C . auris exhibit multidrug resistance with elevated MICs to all three major antifungal classes , including azoles , echinocandins , and polyenes , resulting in limited treatment options [8] . The ability of C . auris to persist and survive in an environment outside the host is unique from most other pathogenic Candida species . This characteristic of survival in a hostile environment perhaps means that C . auris has virulence determinants that help it adapt , adhere , and persist in those settings . Recent reports reveal that C . auris can adhere to polymeric surfaces and form highly drug resistant biofilms [5 , 9 , 10] . In fact , similar to other Candida infections , presence of a central venous catheter has been identified as a risk factor for C . auris [4 , 11] . We embarked on a study to determine if C . auris possessed evolutionarily conserved adhesin protein homologs similar to those found in another phylogenetically-related human fungal pathogen C . albicans . Using bioinformatic and structural homology modeling , we discovered that C . auris contained protein homologs of the C . albicans Als3p . Als3p , a member of the agglutinin-like sequence ( Als ) family of proteins , is also known to be conserved in other non-albicans Candida species [12–14] , and has a multifunctional adhesin and invasin properties essential for host pathogenesis [15 , 16] . The N-terminus of C . albicans Als3p has been developed as a vaccine that induces protective antibody and cell-mediated immune responses [17 , 18] , and has been shown to be safe and efficacious in a clinical trial against recurrent vulvovaginal candidiasis [17 , 19–24] . We found that NDV-3A vaccination induced C . auris cross-reactive antibody and T cell responses . Moreover , the sera from NDV-3A vaccinated mice blocked virulence characteristics of C . auris and enhanced opsonophagocytic killing of this fungus . Furthermore , vaccination with NDV-3A antigen protected mice from lethal C . auris infection and improved the efficacy of sub-therapeutic doses of micafungin . The afforded protection appears to be mediated by humoral and cellular immunity that leads to fungal clearance by tissue macrophages . These findings identify NDV-3A as a promising vaccine for adjunctive treatment of life-threatening bloodstream infections caused by the multidrug resistant C . auris .
Bioinformatic and 3-D structural modeling strategies identified three conserved proteins in C . auris with homologies to C . albicans Als3p . In particular , these C . auris Als3p homologs , PIS50650 . 1 , PIS50263 . 1 and XP_018167572 . 2 , displayed ~30% identity and ~50% similarity ( with several regions of homology reaching up to 88% ) to C . albicans Als3p . Furthermore , sequence alignments revealed that all three proteins possessed domains characteristic of Als3p , such as an N-terminal secretory signal sequence , central amyloid forming and repeated Ser/Thr-rich sequences , as well as the presence of a C-terminus GPI anchor . ( Table 1 , S1 Dataset ) . Once identified , the above sequences were assembled to produce 3-D structural models for analysis in relation to Als3p ( Swiss-model figures shown in Fig 1 ) . Based on this modeling strategy , C . albicans Als3p shares striking similarity to the C . auris proteins , particularly in the N-terminus motif ( Fig 1 , S1 Dataset ) . Because bioinformatic analysis revealed considerable sequence as well as 3-D structural homology between C . albicans Als3p and homologous proteins of C . auris , and predicted cell surface localization of these homologs , we hypothesized that anti-Als3p antibodies should recognize C . auris in vitro . The anti-Als3p antibodies were generated by vaccination of mice with NDV-3A , a vaccine based on the N-terminus of C . albicans Als3p , which is known to induce high titers of serum anti-Als3p antibodies [18 , 20 , 24] . Sera from these alum-vaccinated mice were examined in two different binding assays in vitro . First , different C . auris clinical isolates ( CAU-01 , 03 , 05 , 07 and 09 ) and C . albicans cells were grown for 90 min under germ-tube inducing conditions since Als3p is expressed on C . albicans hyphae [25] , and then treated with anti-NDV-3A sera , followed by immunostaining with fluorescent labeled secondary IgG . Immunostaining of germinating C . albicans cells ( positive control ) , confirmed the specificity of Als3p to the C . albicans filaments . C . auris yeast cells do not differentiate into hyphae , yet anti-Als3p antibodies bound to the cell surface of the fungus , as depicted by a diffused or punctate green fluorescence of the yeast cells . Fungal cells treated with sera from alum-vaccinated mice failed to fluoresce ( Fig 2 panel A ) . The binding of anti-Als3p antibodies to five strains of C . auris obtained from different clades showed the universal presence of Als3p homologs in this multidrug resistant yeast ( See Table B in S1 Table for drug susceptibility ) . Fungal cells treated with sera from alum-vaccinated mice failed to fluoresce ( Fig 2 panel A ) . Next , the extent of anti-Als3p antibody binding to the fungal cells was quantified by flow cytometry . The anti-Als3p antibodies present in the NDV-3A-vaccinated sera bound significantly to C . auris isolates ( indicated by a shift in the peak representing mean fluorescence intensity ) , while the negative control did not . These results demonstrate the specificity of anti-Als3p antibodies to C . auris and further validate the modeling strategy which previously revealed the homology of C . albicans Als3p to Als-like proteins on C . auris ( Fig 2 panel B [upper histograms] ) . To determine which of the three Als3p homologous protein cross reacted to ant-Als3p antibodies , we produced gene deletion mutants of C . auris PIS50650 . 1 , PIS50263 . 1 and XP_018167572 . 2 using CRISPR ( S1 and S2 Figs ) . All three individual Als3p homolog deletion mutants of CAU-09 ( C . auris ) showed reduced binding to anti-Als3p IgG antibodies compared to wild type CAU-09 strain , indicating their role in cross-reactivity to NDV-3A vaccine ( Fig 2 panel B [lower histograms] ) . We have previously shown that NDV-3A vaccine mediated protection against C . albicans infection required both Als3p-specific antibodies and CD4+ Th1/Th17 immune responses , and anti-Als3p antibody titer threshold predicts protective efficacy [21] . Thus , we reasoned that robust antibody and T cells responses would be critical for protection against C . auris in vivo . Therefore , we examined the magnitude of anti-Als3p antibodies binding to C . auris antigens by using cell-based ELISA . First , we confirmed that NDV-3A-vaccinated sera harbored high levels of anti-Als3p antibodies , by using ELISA coated with recombinant Als3p N-terminus . As expected , NDV-3A-vaccinated mice sera contained high titers of anti-Als3p antibodies ( mean end-point titer of 12 , 800 ) , while alum vaccinated sera had none . Next , we developed a cell-based ELISA of C . albicans or C . auris to compare the magnitude of anti-Als3p antibodies to antigens of yeast cells . We allowed the binding of serially diluted mouse sera to C . albicans germ tube cells or C . auris yeast cells and detected the endpoint titer with HRP-conjugated anti-mouse IgG detection antibodies . We found that NDV-3A vaccinated mice had high titers of cross-reacting antibodies against C . albicans cells , with titers similar to the recombinant Als3p N-terminus-based ELISA ( 12 , 800 ) . Consistent with the confocal microscopy and flow cytometry results , and compared to sera obtained from alum-vaccinated mice , NDV-3A-vaccinated mice sera were found to have cross-reactive anti-Als3p antibodies to C . auris cells albeit with 4-fold less than C . albicans ( mean end-point titer of 3 , 200 ) ( Fig 3 panel A ) . We also examined whether NDV-3A vaccination induced memory CD4+ T cells that are also cross-reactive to C . auris . Splenocytes from NDV-3A- or alum-vaccinated mice were stimulated with either recombinant Als3p-N-terminus or heat-killed C . auris for 5 days , followed by treatment with PMA/ionomycin in the presence of monesin/brefeldin A protein transport inhibitor . These cells were stained with fluorescent-labelled antibodies against surface markers and intracellular cytokines . Flow cytometry analysis of these stained cells revealed a significant presence of high frequency of Als3p-specific and C . auris cross-reactive CD4+ T cells , which were comprised of Th1 ( IFN-γ + ) , Th2 ( IL-4+ ) and Th17 ( IL-17+ ) cells ( Fig 3 panel B , S3 , S4 , S5 and S6 Figs ) . The C . auris cross-reactive T cells were not detected in alum control mice . Further , the total C . auris or Als3p-specific CD4+ T cell responses were similar in magnitude and were slightly biased towards Th2 type . Finally , the C . auris -specific Th17 cell response was higher compared to Als3p-specific responses ( Fig 3 panel B , S3 , S4 , S5 and S6 Figs ) . Overall , these results revealed that , NDV-3A induced robust C . auris cross-reactive antibodies as well as CD4+ T cell immune responses . Attachment of microorganisms to an abiotic surface is the first step in formation of a drug-resistant biofilm . Biofilm formation has been most studied in C . albicans and Als3p is essential for establishment of the early stages of biofilm growth [26] . Recently , we have reported that blocking Als3p by anti-Als3p antibodies abrogate C . albicans biofilm formation [20] . C . auris also has the capacity to form drug-resistant biofilms [9] . Thereby , we evaluated if anti-Als3p antibodies from NDV-3A-vaccinated sera could similarly inhibit biofilm formation in C . auris . Sera from NDV-3A-vaccinated mice significantly inhibited biofilm formation compared to control anti-alum sera ( p = 0 . 001 ) ( Fig 4 panel A ) . Of note , sera were used at 1:10 ( sera to media ) ratio , and heat-treated prior to their contact with fungal cells , to rule out the role of complement in the inhibitory activity ( Fig 4 panel A ) . We previously demonstrated that anti-Als3p antibodies act as an opsonin to enhance phagocyte-mediated killing of C . albicans [20] . Thus , we posited that Als3p antibodies would enhance sensitivity of C . auris to macrophage killing . To test this hypothesis , heat-inactivated mice sera ( at 10% concentration ) from NDV-3A- or alum-vaccinated mice were incubated with C . auris to allow binding of anti-Als3p antibodies to the fungal cell surface , and then added to mice macrophage cell-line ( 1:1 ) . As a control , C . auris yeast cells were subjected to macrophages without any sera . Both NDV-3A- and alum-vaccinated mice sera did not have any negative impact on growth of C . auris ( S7 Fig ) . However , sera from NDV-3A-vaccinated mice significantly enhanced macrophage-mediated killing of C . auris compared to sera from alum vaccinated mice ( 40% vs 2% killing respectively , normalized to C . auris killing by macrophage-only condition , p = 0 . 028 ) . Thus , anti-Als3p antibodies generated by the NDV-3A vaccine enhanced the opsonophagocytic killing of C . auris ( Fig 4 panel B ) . Blood stream infections are predominant manifestations of C . auris , accompanied by significant mortality in susceptible patients . Since , anti-Als3p antibodies generated by NDV-3A vaccination bound to C . auris , blocked virulence and rendered it susceptible to macrophages , we reckoned that NDV-3A vaccination would have the potential to protect against C . auris disseminated candidiasis . We developed a neutropenic mouse infection model by screening different clinical isolates of C . auris ( CAU-01 , 03 , 05 , 07 , 09 ) for their lethality in mice . An intravenous 5x107 CFU/mouse dose of C . auris CAU-09 showed 100% mortality in neutropenic mice and was used for vaccine efficacy testing ( immunocompetent mice are resistant to C . auris bloodstream infection ) . To test the vaccine-mediated protective efficacy , mice were either vaccinated with NDV-3A or alum , three times ( primary + two booster doses ) , and then infected with C . auris via the tail vein ( Fig 5 panel A ) . Mice vaccinated with alum displayed 100% mortality by day 7 , while NDV-3A-vaccinated mice showed 40% overall survival . Surviving mice appeared healthy at 21 days when the experiment was terminated ( Fig 5 panel B ) . Furthermore , while both sets of vaccinated mice displayed weight loss post infection compared to uninfected mice , NDV-3A-vaccinated mice had lesser extent of weight loss versus alum-vaccinated mice ( Fig 5 panel C ) . In replicate studies to the survival experiments , mice were sacrificed 4 days post infection , and their target organs ( kidneys and brain ) were collected for tissue fungal burden . The NDV-3A- vaccinated mice had a 10-fold lower ( ~1 . 0 log ) fungal burden in both kidneys and brains compared to alum control mice ( p = 0 . 0006 ) ( Fig 5 panel D ) . Histopathological examination of tissues collected from mice sacrificed on day 4 post infection , confirmed inhibition of disseminated fungal infection in kidneys of NDV-3A-vaccinated mice , while alum-treated mice had multiple abscesses containing C . auris cells throughout the organ ( Fig 6 panel A ) . Compared to kidneys , the brain had an overall 100-fold reduced fungal burden ( Fig 5 panel D ) , which was also apparent in the histopathological images ( Fig 6 ) . While overall only a few cells could be visualized in the brain , alum-vaccinated mice appeared to have more fungi compared to brains of NDV-3A-vaccinated mice . Interestingly , in both cases , C . auris cells were localized in the blood capillaries , rather than the brain tissue itself ( Fig 6 panel B ) . To dissect the role of NDV-3A vaccine-mediated protection , we conducted serum adoptive transfer and immune cell depletion experiments . Briefly , pooled anti-NDV-3A ( mean titer = 25600 , S8 Fig ) or alum serum was administered i . p . to naïve neutropenic Balb/c mice starting on day -1 and weekly thereafter , relative to i . v . infection with C . auris . The mice that received anti-NDV-3A serum had significantly enhanced median survival time and 21-day % survival compared to the mice that received anti-alum serum ( 6 days and 33% survival for anti-NDV-3A vs . 4 days and 0% for anti-alum serum , p = 0 . 0006 ) ( Fig 7 panel A ) . Since anti-NDV-3A antibodies enhanced the opsonophagocytic killing of the C . auris by murine macrophages ex vivo , we hypothesized that these antibodies help in controlling C . auris infection through macrophage-mediated killing in the neutropenic mouse model . To examine this possibility , we vaccinated mice as previously described and depleted their macrophages by injecting clodronate-liposome . Mice receiving PBS-liposome served as placebo control . Macrophage depletion in primary lymphoid organs was verified in a separate set of mice by staining of spleen and inguinal lymph node cells with macrophage-specific phenotypic markers ( CD11b+ F4/80+ ) . By day 4 after the second injection of clodronate-liposome , the macrophage population decreased by ~97% in spleens taken from neutropenic mice administered clodronate-lipsome when compared to PBS-liposome treat neutropenic mice , while lymph nodes maintained a very low frequency of macrophages in both depletion and control depletion groups ( S9 Fig ) . Interestingly , macrophage depletion completely reversed the survival benefit afforded by the NDV-3A vaccine ( from 37% survival to 0% survival , and from 11 days median survival to 5 median survival days , p value = 0 . 01 ) . This clearly indicates that loss of macrophages in neutropenic mice severely compromised the NDV-3A vaccine-mediated protection against systemic C . auris infection ( Fig 7 panel B ) . Next , to examine the role of CD4+ helper T cells , we vaccinated mice with NDV-3A or alum as previously described and depleted CD4+ T cells by injecting anti-mouse CD4 ( GK1 . 5 IgG2b ) antibodies through i . p . route on day -3 and 0 relative to infection . The mice were made neutropenic on day -2 and infected with C . auris on day 0 followed by survival analysis for 3-weeks post-infection . The CD4+ T helper cell depletion was verified on day 4 after the second injection of anti-mouse CD4 antibodies in primary lymphoid organs as described previously [27] . Mice treated with anti-CD4 antibodies had a 99% reduction in the frequency of CD4+ T cells compared to isotype control antibody treated mice ( S10 Fig ) . Further , CD4+ T cell depletion in NDV-3A vaccinated mice had marked reduction in median survival time against systemic C . auris challenge ( 12 vs 5 . 5 median survival days for CD4+ replete and CD4+ deplete arms [p = 0 . 04 by Mantel Cox comparisons] and 38% vs 13% survival , [p = 0 . 18 by Log Rank test] ) . Interestingly , the protection afforded by the NDV-3A vaccine in absence of CD4+ was still higher than survival of mice vaccinated with alum with or without depletion of CD4+ cells ( 13% vs 0% survival , and 6 vs 4 median survival days for NDV-3A vaccinated and CD4+ deplete vs . alum vaccinated mice , p value = 0 . 0028 ) ( Fig 7 panel C ) . Collectively , these adoptive and targeted depletion studies show both antibodies ( humoral ) and CD4 cells are required in providing protection against systemic C . auris infection in this neutropenic mouse model , likely via enhancing fungal-killing by tissue macrophages . In the clinical setting , any therapeutic vaccine approach is likely to be used in combination with standard antifungal therapy . Given that NDV-3A vaccination provided significant protection against C . auris candidemia , we questioned if vaccination could potentiate the activity of antifungal drugs , in vivo . To investigate this possibility , we vaccinated mice with NDV-3A or alum as above , made them neutropenic and then infected them with C . auris . Twenty-four hours post infection , one group of NDV-3A-vaccinated or alum control mice were treated with sub-inhibitory concentrations of micafungin ( 0 . 5 mg/kg body since CAU-9 is micafungin sensitive [MIC = 0 . 125 μg/ml] ) daily for 1 week . As expected , NDV-3A protected 40% of mice from succumbing to infection , while micafungin provided only 20% protection . Interestingly , when used in combination , NDV-3A and micafungin significantly enhanced the overall survival of mice to >70% ( p = 0 . 04 , compared to NDV-3A-vaccination alone; and p = 0 . 001 , compared to micafungin alone ) . These results show that NDV-3A vaccine acts additively with active antifungals in protecting mice from C . auris candidemia ( Fig 8 ) .
The simultaneous global emergence of the pathogenic yeast C . auris , its ability to persist in healthcare-settings , combined with its potential to resist all classes of antifungal drugs , has resulted in a renewed focus of the Centers for Disease Control ( CDC ) , health-care professionals and scientists alike , on multidrug resistant fungal organisms [3 , 8] . To add to the problem , C . auris reportedly possesses the capacity to adhere to and develop inherently drug-resistant biofilms on abiotic surfaces [10 , 28] , thus bolstering its ability to persevere in hospital environments , and cause outbreaks . Disseminated infections due to C . auris in susceptible patients ( e . g . advanced age , presence of central venous catheter , surgery , prolonged hospitalization ) , are associated with unacceptably high mortality rates between 30–60% [6 , 7] . It is unlikely that successful treatment will be achieved with antifungal drugs alone , even with proper antibiotic stewardship . Identification of alternative strategies to prevent and treat infections caused by this fungus are needed . Improved therapy is achievable by either discovering novel molecular targets in its genome , or identifying those that may be homologous to other therapeutically-targeted proteins from phylogenetically related fungi . We applied computational molecular modeling and bioinformatic strategies to demonstrate that C . auris possesses homologs of C . albicans major adhesin/invasin protein , Als3p ( Fig 1 , Table 1 and S1 Dataset ) . This was an intriguing finding , since Als3p is a hyphae specific protein and C . auris does not form filaments . The presence of Als3p-homologs in C . auris adds to the list of other genes or gene products with homologies to those associated with virulence and drug resistance in C . albicans . For example , homologs of several C . albicans efflux genes belonging to the major facilitator superfamily ( MFS ) , MDR and the ATP-binding cassette ( ABC ) transporter families have been identified , suggesting that efflux is likely a potential resistance mechanism mediating multidrug resistance in C . auris [29 , 30] . In fact , homologous genes of C . albicans virulence proteins such as Serine/Threonine Enzyme-related proteins , mannosyl transferases , a number of secreted aspartyl proteases , as well as kinases involved in virulence and antifungal stress response such as Hog1 protein kinase , 2-component histidine kinase etc . , have been identified in the C . auris genome [31] . Our discovery of Als3p-like proteins in C . auris clearly indicates that despite its extremely high genomic divergence from C . albicans , core gene families involved in adhesion/invasion , acquiring drug resistance and other virulence-related genes are conserved in C . auris; perhaps also a reason for its persistent nature and its success as a MDR organism . The Als3p homologs in C . auris are targets for vaccine strategies involving the NDV-3A vaccine , which was developed based on the N-terminus region of C . albicans Als3 , and formulated with alum [18] . Our group has previously reported on the efficacy of NDV-3 ( a His-tagged N-terminus of recombinant Als3p formulated with alum ) in preventing disseminated candidiasis in mice infected with C . albicans [17] . Vaccination with Als3p coupled with complete/incomplete Freund’s adjuvant also protects mice from hematogenously disseminated candidiasis due to other Candida species including C . glabrata , C . tropicalis , and C . parapsilosis ( S11 Fig ) . NDV-3A ( N-terminus of recombinant Als3p without his-tag and formulated with alum ) was also shown to be safe when given to healthy volunteers [18] . In a Phase 1b/2a exploratory clinical trial , NDV-3A was safe and immunogenic in women who suffer from recurrent vulvovaginal candidiasis and protected them from symptoms of infection [19 , 20] . Most recently we have described that anti-Als3p antibodies generated from NDV-3A-vaccinated patients ( and not alum-treated ) had the potential to affect properties of adherence , filamentation and biofilm formation in C . albicans [20] . Thus , we questioned if anti-Als3p antibodies could recognize C . auris , interfere in functions characteristically associated with Als3p , and prove clinically valuable as vaccine-based strategies against C . auris . Using different and complementary approaches ( immunofluorescence , flow cytometry , and cell-based ELISA ) , we validated that NDV-3A vaccination-generated high titer anti-Als3p antibodies that recognized different strains of C . auris ( Figs 2 and 3 panel A ) . Importantly , the endpoint IgG antibody titer determined by regular Als3p antigen ELISA and C . albicans cell-based ELISA was the same . This finding clearly demonstrates that the two assays are comparable and cell-based ELISA can correctly estimate the antigen-specific antibody titer in serum . Based on these results , we conclude that although C . auris cross-reactive IgG antibodies were 4-fold lower than the Als3p antigen specific IgG antibodies , these titers were still sufficiently high ( Fig 3 panel A ) . The protective efficacy of NDV-3A vaccine against C . albicans infection has been correlated with high antibody titers and Th1/Th17 cell mediated immune responses [17 , 21 , 23] . An effective immunity against a pathogen is usually comprised of both antibodies and memory T cell immune responses . Therefore , we examined the splenocytes of NDV-3A-vaccinated mice and detected high levels of C . auris cross-reactive CD4+ T cells responses in NDV-3A-vaccinated mice , which was similar to Als3p-specific T cells in magnitude . Further , these CD4+ T cells responses ( both C . auris and Als3p-specific ) were marginally shifted towards a Th2-type , probably because alum is known to favor Th2 cell induction [32 , 33] . Further , the NDV-3A vaccination also induced high frequency of Als3p-specific Th17 cells that cross-reacted to heat-killed C . auris . These results are in line with our previous reports showing that NDV3 and NDV-3A vaccines induce strong Als3p-specific Th17 immune responses [18 , 19 , 21–23] . Together , these results highlight that NDV-3A vaccine induced robust C . auris cross-reactive antibody and T cell immune responses ( Fig 3 ) . The contribution of anti-Als3p antibodies towards abrogation of virulence potential of C . auris was illustrated by its ability to block C . auris biofilm formation , in vitro ( Fig 4 panel A ) . The CAU-09 strain mutant lacking Als3p homolog gene showed reduced binding of anti-Als3p antibodies compared to wild-type CAU-09 strain ( Fig 2 panel B ) . This finding provides evidence that anti-Als3p antibodies obtained from mice vaccinated with NDV-3A , which prevent biofilm formation , indeed bind to the Als3p homologs on C . auris surface . Additionally , opsonization of C . auris with anti-Als3p antibody containing sera from NDV-3A- vaccinated mice resulted in efficient immune recognition of the pathogen by mice macrophages , leading to increased opsonophagocytic killing ( Fig 4 panel B ) , a feature that was noticed with the activity of anti-Als3p antibodies and C . albicans [20 , 34] . This finding led to the evaluation of NDV-3A for protection against C . auris infections , in vivo . Indeed , the results from in vitro assays translated successfully in vivo , where vaccination by NDV-3A protected some neutropenic mice from disseminated candidiasis caused by C . auris ( Figs 5 and 6 ) . These preclinical findings suggest that the NDV-3A vaccine may be useful against C . auris even in the setting of immunosuppression . Furthermore , although the use of cyclophosphamide in mice results in leukopenia , there is evidence that its effect on tissue macrophages is less severe . In fact , deficiency in alveolar macrophages due to continuous administration of cyclophosphamide was more of a chronic and gradual reduction that never reaches complete deficiency [35] . This finding is in agreement with a study showing that cyclophosphamide has a lesser effect on mouse tissue macrophages , which are usually derived from an embryonic , and not from hematopoietic , origin [36] . Indeed , our mouse macrophage-depletion studies showed that immunosuppressed ( cyclophosphamide/cortisone acetate ) mice treated with liposome had a considerable number of macrophages in splenocytes ( S9 Fig panel B ) . Thus , the residual tissue macrophages in the cyclophosphamide/cortisone acetate could contribute to the protection seen in vivo given the fact that anti-Als3p antibodies enhance OPK activity of macrophages in vitro ( Fig 4 panel B ) . The relative role of humoral and cellular immunity in NDV-3A vaccine-mediated protection against C . auris infection was demonstrated by anti-NDV-3A serum adoptive transfer and T cell depletion , respectively ( Fig 7 ) . Specifically , our data support a model by which anti-Als3p antibodies are opsonophagocytic and these antibodies and CD4+ cells activate tissue macrophages to enhance fungal clearance . The model is supported by results clearly showing that adoptive transfer experiments with anti-Als3p antibodies can afford protection against C . auris infection , but are critically dependent on macrophages for efficient protection ( Fig 7 panel B ) . Similarly , depletion of CD4+ compromised the protection afforded by the NDV-3A vaccine ( Fig 7 panel C ) . In this respect , CD4+ T helper cells play a critical role in protecting against fungal infection , and we previously have shown that Th1 and Th17 cells are required for NDV-3A vaccine-mediated protection against C . albicans [23] . Aluminum salts together with monophosphoryl lipid A ( MPL ) are known to prime long-lived memory CD8 T cells [37] . However , NDV-3A vaccine is alum formulated and doesn’t contain any MPL . Thus , it is not expected that NDVA-3A will differentiate CD8+ cytotoxic T-cells . However , the role of CD8+ cell cannot be ruled out especially with the reduced , but not complete reversal of , protection elicited by NDV-3A when CD4+ cells were depleted . In contrast , the complete reversal of protection in macrophage depleted mice argues that the main mechanism of action is reliant on macrophage-mediated opsonophagocytosis clearing of the fungus in infected tissues . Of great interest and clinical relevance is the additive effect in protecting mice from C . auris infection when the vaccine was given with micafungin ( Fig 8 ) . This result provides a strong rationale for the continued investigation of the combined use of active NDV-3A vaccination with antifungal drugs in multidrug-resistant lethal C . auris infections and gives hope to improved treatment outcomes . In summary , the unique property of C . auris to persist outside human body in the harsh environmental settings of the hospital , its rapid evolution of drug resistance , and its propensity to infect immunosuppressed hospitalized individuals , increasingly threaten global and personal health . The identification of Als3p homologs in C . auris opens an avenue for novel immunotherapeutic approaches utilizing either the currently available NDV-3A vaccine , or perhaps an anti-Als3p antibody mediated passive vaccination strategy in the near future . These immunotherapeutic approaches could enhance successful treatment of infections caused by such fungal “superbugs” , thereby reducing morbidity and mortality . Given the safe profile of NDV-3A in humans , future studies will focus on testing this vaccine in patients at high risk of acquiring C . auris infection ( e . g . colonized patients ) and/or those who already suffer from the infection ( e . g . therapeutic vaccine combined with antifungal agents ) .
All procedures involving mice were approved by IACUC of Los Angeles Biomedical Research Institute ( protocol 11672 ) , according to the NIH guidelines for animal housing and care . Moribund mice according to detailed and well characterized criteria were euthanized by pentobarbital overdose , followed by cervical dislocation . The C . auris strains ( CAU-01 , East Asian clade , ear; CAU-03 , African clade , blood; CAU-05 , South American clade , blood; CAU-07 , South Asian clade , blood; and CAU-09 , South Asian clade , bronchoalveolar lavage [BAL] ) were obtained from Dr . Shawn Lockhart at Centers for Disease Control and Prevention ( CDC , Atlanta ) . C . glabrata 31028 , C . parapsilosis 22019 , and C . tropicalis 4243 are all clinical bloodstream isolates obtained from the microbiology laboratory at Harbor-University of California at Los Angeles [UCLA] Medical Center ) , while C . krusei 91–1159 was obtained from the University of Texas Health Science Center at San Antonio . For routine culturing , C . albicans ( SC5314 ) and C . auris were grown in Yeast Extract Peptone Dextrose ( YPD ) broth overnight at 30°C with shaking at 200 rpm . Cells were washed with 1x phosphate buffered saline without Ca++/Mg++ ( PBS , Gibco by Life Technologies ) three times prior to counting blastopores with a hemocytometer . For germinating C . albicans , 5x106 yeast cells/ml in RPMI-1640 media ( supplemented with L-Glutamine ) were allowed to form germ tubes at 37°C for 75 minutes with shaking at 200 rpm . For staining , cell-based ELISA , and splenocytes stimulation , the C . auris strains and Als3p homolog mutants were grown in above germination conditions . For biofilm formation assay , the cells were incubated at 37°C in yeast nitrogen base ( YNB ) medium for 24 hours . For heat killing of yeast cells , 5 x 106 cells /mL of PBS were incubated at 65°C for 45 minutes . Yeast cell death was confirmed by plating the heat-subjected cells on fresh plate and incubating the cells at 30°C for several days . For fixation , the yeast cells were incubated with 4% paraformaldehyde solution ( in PBS ) at 4°C for 1 hour . The C . albicans N-terminal Agglutinin-Like Sequence protein ( Als3p , Gene Bank ID: AOW31402 . 1 ) amino acid sequence was aligned with the C . auris using protein BLAST and CLUSTAL-W ( NCBI ) . The predicted Als3p homologs in C . auris proteome were further screened for homology among each other using CLUSTAL-W and Als3p homologs having >95% homology with each other considered as same protein . The number of Als3p amino acids showing similarities with C . auris proteins was represented as percent positive . The functional domains were identified by using different online bioinformatics tools for GPI-anchor ( http://gpi . unibe . ch/ ) , amyloid sequence ( http://protein . bio . unipd . it/pasta2/ ) , and Ser/Thr rich sequence ( https://www . ebi . ac . uk/Tools/pfa/radar/ ) prediction . The 3-D protein structure models were built using amino acid sequences and the templates available in the Swiss-model database [38–40] ( https://swissmodel . expasy . org/ ) . Briefly , the templates were searched in SWISS-MODEL template library ( SMTL ) using BLAST and HHBlits . The target sequence was searched with BLAST against the primary amino acid sequence contained in the SMTL . The target-template alignment was performed to build the model by using ProMod3 , and coordinates that were conserved between the target and the template were copied from the template to the model . Insertions and deletions were re-modelled using a fragment library , and the side chains were rebuilt . Finally , the geometry of the resulting model is regularized by using a force field . In case loop modelling with ProMod3 fails , an alternative model is built with PROMOD-II . The models showing high accuracy values were finalized for similarity comparisons . To knock-out the Als3p homologs , we generated plasmid constructs containing the nourseothricin resistance gene cassette ( pNAT ) [41] . For each target Als3p homolog gene , we selected 700–900 bp of the promoter and terminator regions for PCR amplification , as follows: PIS50650 . 1 gene: upstream fragment ( 39887–40857 ) and downstream fragment ( 45859–46591 ) , PIS50263 . 1 gene: upstream fragment ( 923300–924246 ) and downstream fragment ( 926539–927376 ) , XP-018167572 . 2: upstream fragment ( 849450–8504433 ) and downstream fragment ( 852693–853581 ) , ( see Table A [I-IV] in S1 Table for primer sequence ) . We amplified the target gene segments from C . auris ( CAU-09 ) genomic DNA and cloned into plasmid pNAT using restriction enzymes ( KpnI/ApaI for the upstream and NotI/SacII for the downstream fragments ) or by Gibson method ( S1 Fig ) . Further , to release the gene deletion cassette , the construct was digested with KpnI/SacII restriction enzymes . The competent C . auris ( CAU-09 ) cells were transformed by using gene deletion cassette , CRISPR-Cas9 protein and guide RNA as described previously [42] . Transformants were verified by PCR and RT-PCR using specific primers ( Table A in S1 Table ) to confirm the deletion of the target gene [43] ( S1 and S2 Figs ) . We used 4–6 week old outbred ICR ( CD-1 ) mice in this study . The NDV-3A vaccine was formulated by mixing 300 μg of C . albicans recombinant Als3p with 200 μg alum adjuvant per dose . The recombinant Als3p was produced in Saccharomyces cerevisiae , and was a Gift from NovaDigm therapeutics . For immunization , the NDV-3A vaccine or alum was injected into mice subcutaneously ( s . c . ) at day 0 , 21 and 35 . Mice were bled 14 days after final immunization and serum was isolated for antibody titer determination . For efficacy studies against Candida species other than C . auris , mice were immunized by subcutaneous injection of recombinant N-terminus of Als3p-N ( 20 μg ) mixed with complete Freund’s adjuvant ( CFA; Sigma-Aldrich , St . Louis , MO ) at day 0 , followed by a booster dose in incomplete Freund’s adjuvant ( IFA; Sigma-Aldrich ) at day 21 . Control mice were immunized with CFA followed by IFA alone . Fourteen days following the boost , immunized mice were infected via the tail vein with the incoula mentioned in S11 Fig . The yeast or germ tube cells ( 2x106 cell ) from wild type C . auris strains ( CAU-01 , CAU-03 , CAU-05 , CAU-07 and CAU-09 ) , CAU-09 Als3p homolog mutants ( PIS50650 . 1-/- , PIS50263 . 1-/- , XP-018167572 . 2-/- ) or C . albicans were fixed with 4% paraformaldehyde at 4°C for 1 hour . After blocking the cells with 3% bovine albumin serum solution ( in 1x PBS ) , these cells were added to 96-well plate and centrifuged to pellet the cells . The pellet was resuspended in 100 μl of anti-NDV-3A or alum serum diluted at 1:500 in 1x PBS and incubated for 1 hour at room temperature . The cells were washed three times with 1x PBS prior to adding 100 μl of Alexa Fluor 488 labelled anti-mouse IgG detection antibodies ( 1:100 dilution in PBS ) . After 1 hour of incubation at room temperature , the cells were resuspended in 300 μl of PBS and analyzed using confocal microscopy or flow cytometry . For confocal microscopy , 20 μl of stained cells were added to the glass slides and covered with cover slips . The images were taken at 40x resolution using Laser Scanning Confocal microscopy ( Leica Confocal Microsystem ) . For flow cytometry , the stained cell suspension was transferred to flow tubes and 20 , 000 events were acquired using LSR II flow cytometer ( BD Biosciences ) . Flow cytometry data were analyzed using FlowJo software ( Version 10 ) . Ninety six-well plates were coated with 5 μg/ml of Als3p in bicarbonate/carbonate coating buffer ( pH 9 . 6 ) overnight at 4°C . The next day , the plates were washed three times with 1x wash buffer ( PBS containing 0 . 05% tween-20 ) and blocked with 3% BSA solution for 2 hours at room temperature . After washing three times , diluted serum samples were added to the plates in duplicates and incubated for two hours . After incubation , the plates were washed three times and 1:1000 diluted anti-mouse IgG antibodies ( Jackson , Cat#115-035-164 ) labelled with peroxidase were added and incubated for 1 hour at room temperature . Finally , the plates were washed five times with washing buffer , TMB ( 3 , 3′ , 5 , 5′-Tetramethylbenzidine ) substrate ( Invitrogen , Cat#00-4201-56 ) was added . Color development was allowed for 5–10 minutes and absorbance was measured at 450 nm after the reaction was stopped with 1 N sulfuric acid ( Sigma , Cat#339741 ) . To determine the C . auris cross-reactive antibodies in NDV-3A vaccinated mice , we used a cell-based Enzyme-Linked Immunosorbent Assay ( ELISA ) . As above , germinated C . albicans ( SC5314 ) or C . auris ( CAU-09 ) were counted and fixed in 4% cold paraformaldehyde ( Sigma-Aldrich , Cat#158127 ) for 30 min . The fixed cells were washed with 1x PBS . The U-bottom 96-well plates were blocked overnight with 1x blocking solution ( 3% BSA in 1x PBS , Thermo Fischer ) , and next day 2-fold serial dilutions of 100 μl serum samples ( diluted in 1x blocking solution ) per well were added in duplicates . Further , a total of 107 fixed cells/well/100 μl of C . auris and germinated C . albicans were added to these serum sample-containing wells . The plates were incubated for 2 hours at room temperature . After incubation , the cells in the 96-well plate were washed three times with 1x PBS and secondary HRP-conjugated anti-mouse IgG antibodies were added as per the manufacturer instructions . After 1 hour incubation at room temperature , the cells in 96-well plate were washed with 1x PBS . Further , 100 μl/well TMB substrate was added to each well and incubated until the color developed ( normally 5–20 min ) . The color reaction was stopped by adding 50 μl 1 N H2SO4 per well and plates were centrifuged at maximum speed to pellet the cells . One hundred microliter colored substrate supernatant from each well were transferred to fresh flat bottom 96-well plates and OD was measured at 450 nm . The endpoint titers in NDV-3A vaccinated mice were determined by plotting mean OD450 vs serial serum dilution , and noting the highest dilution with significantly higher OD450 compared to alum vaccinated mice group . The splenocytes were harvested from the NDV-3A or alum immunized mice by homogenizing individual spleens in 100 μm cell strainer . The RBCs were lysed by 1x RBC lysis buffer ( Santa Cruz Biotech , Dallas , Cat# SC-296258 ) , and filtered through the 100 μm sterile filters . The cells were resuspended in 1x RPMI supplemented with 10% FBS , counted and plated at 1x106 splenocytes/ 100 μl/ well in a U-bottom 96-well plate . The splenocytes were kept unstimulated or stimulated with 100 μl/well of 10 μg/ml recombinant Als3p , or 3x106 cells/well heat-killed C . auris for 5 days at 37°C . On day 5 , 100 μl of the culture supernatant was replaced with fresh RPMI media containing 50 ng/ml PMA ( phorbol 12-myristate 13-acetate , Sigma , Cat# P8139 ) , and 500 ng/ml ionomycin ( Sigma , Cat#I3909 ) . After 1 hour , protein transport inhibitor cocktail ( eBioscience , Cat# 00-4980-03 ) was added to each well at final 1x concentration , and plates were incubated for another 3 hours . Next , the cells were stained with extracellular CD3 APC-Cy7 ( BD Biosciences , Cat#557596 ) and CD4 PerCP Cy5 . 5 ( BD Biosciences , Cat#550954 ) antibodies ( 0 . 25 μg/sample ) for 30 minutes at room temperature . Subsequently , the cells were washed , fixed and permeabilized with Cytofix/Cytoperm solution ( Invitrogen , Cat#00-5123-43 ) and then stained intracellularly with IFN-γ APC ( BD Pharmingen , Cat#554413 ) , IL-4 FITC ( eBioscience , Cat#11-7042-82 ) and IL-17 PE ( R&D Systems , Cat#IC7211P ) antibodies ( 0 . 25 μg/sample ) for 30 minutes at room temperature . The stained cells were acquired in BD LSR-II flow cytometer ( BD Biosciences ) , and at least 30 , 000 CD3+ T cells were recorded . The data was analyzed using FlowJo ( Version 10 ) software . Biofilms were developed in 96-well polystyrene microtiter plates as previously described , with slight modifications [44] . Briefly , 95 μL of C . auris cells ( 2 × 105 cells/ml in YNB medium ) was added to the wells containing 5 μL of test or control mouse serum ( 5% serum vol/vol ) , and incubated at 37°C . Control wells had no serum . After 24 hours , wells were washed twice with PBS , and the extent of biofilm formation was quantified by XTT assay ( 490 nm ) [44] . The opsonophagocytic killing assay was based on a modification of a previously used method [45] . C . auris yeast cells ( CAU-09 ) were added into 96-well microtiter plates . Murine RAW 264 . 7 macrophage cells ( American Type Culture Collection [ATCC# TIB-71] , Rockville , MD ) were cultured at 37°C in 5% CO2 in RPMI-1640 ( Irvine Scientific , Santa Ana , CA ) with 10% fetal bovine serum ( FBS ) , 1% penicillin , streptomycin , and glutamine ( Gemini BioProducts ) , and 50 mM β-mercaptoethanol ( Sigma-Aldrich , St . Louis , MO ) . RAW 274 . 7 cells were activated by exposure to 1 ng/ml LPS ( Sigma-Aldrich ) for 24 hours . Activated RAW 264 . 7 macrophages were harvested after scraping with BD Falcon cell scrapers ( Fischer Scientific ) and added to the microtiter wells at a 1: 1 ratio of macrophages to C . auris . After 2 hours of incubation with gentle shaking , aliquots from the wells were quantitatively plated in YPD agar plates . The percent killing of C . auris was calculated using the following formula: 1- [CFUs from tubes with ( mouse serum + C . auris + macrophages ) /average CFU in tubes with ( C . auris + macrophage ) ] . CD-1 mice vaccinated with NDV-3A vaccine were infected with C . auris CAU-09 to evaluate the efficacy of the vaccine . Briefly , twelve days after the final boost , mice were made neutropenic by 200 mg/kg cyclophosphamide delivered intraperitoneally ( i . p . ) and 250 mg/kg cortisone acetate ( s . c . ) administered on days -2 and +3 , relative to infection . To prevent bacterial superinfection in the immunosuppressed mice , we added enrofloxacin ( at 50 μg/ml ) to the drinking water . Mice were infected through tail vein injection with 5x107 CFU/mouse . For combination with antifungals , alum- or NDV3-A-vaccinated and infected mice were treated with a minimal protective dose of 0 . 5 mg/kg/day of the clinically used micafungin by i . p . administration . Treatment started after 24 hours of infection and continued until day +7 . Mice were monitored for their survival for 21 days after the infection . For fungal burden determination , mice were vaccinated , made neutropenic and then infected as above and then euthanized on day 4 post infection to collect kidneys and brain . The organs from each mouse were weighed , homogenized and quantitatively cultured by 10-fold serial dilutions on YPD plates . Plates were incubated on 37°C for 48 hours prior to enumerating colony forming units ( CFUs ) /gram of tissue . Finally , histopathological examination of kidneys or brain from mice sacrificed on Day 4 post infection , were fixed in 10% zinc-buffered formalin , embedded in paraffin , sectioned , and stained with Pacific Acid Schiff ( PAS ) stain . Balb/c mice were vaccinated with NDV-3A or alum on day 0 , and boosted on days 21 and 35 . Vaccinated mice were made neutropenic using 200 mg/kg body weight of cyclophosphamide administered subcutaneously and 250 mg/kg body weight of cortisone acetate given by intraperitoneal injection on day 47 . Mice were infected intravenously with C . auris on day 49 as above . To prevent bacterial superinfection in the immunosuppressed mice , 5 mg ceftazidime /mouse was administered daily through subcutaneous route for up to 14 days post infection , and after that 50 μg/ml enrofloxacin was added to drinking water . All NDV-3A and alum vaccinated mice were grouped as depletion and control depletion arms . For CD4+ T cell depletion experiment , 200 μg/mouse dose of rat anti-mouse CD4 IgG2b ( clone GK1 . 5 , BioXcell , Cat #BE0003-1 ) ) or rat IgG2b isotype antibodies ( Clone: LTF-2 , BioXcell , Cat #BE0090 ) were administered intraperitoneally on day 11 and 14 post-second boost ( i . e . days -3 and 0 , relative to infection ) [27 , 46] . For macrophage depletion , 1 mg clodronate-liposome or PBS-liposome particles ( Liposoma B . V . , Cat # CP-005-005 ) were injected intraperitoneally on day 11 and 14 post second boost , and subsequently twice a week ( i . e . days -3 and 0 , 4 , 8 , 12 and 16 relative to infection ) [47 , 48] . Mice were infected intravenously with 5x107 C . auris ( CAU-09 ) and monitored for their survival for three weeks . Three additional mice in each depletion and control depletion arm were taken to verify the CD4 and macrophage depletion . Briefly , 4 days after administering the second dose of depletion drug , mice were euthanized , and spleen and lymph nodes were harvested and homogenized to make single cell suspension . For CD4+ cell depletion verification , splenocytes and lymph node cells were stained with anti-CD3 APC ( BD Pharmigen , Cat #BDB565643 ) and anti-CD4 Alexa Fluor 700 antibodies ( Biolegend , Cat #100536 ) . Macrophage depletion was verified by staining anti-CD11b PerCP Cy5 . 5 ( eBioscience , Cat #5015759 ) and anti-F4/80 APC ( eBioscience , Cat #501128925 ) antibodies . Stained cells were acquired in BD LSR II flow cytometer , and data were analyzed in FlowJo software . Differences in survival studies were analyzed by the non-parametric Log Rank test for overall survival and with Mantel-Cox comparisons for median survival times . All other comparisons were conducted with the non-parametric Mann Whitney test . P values <0 . 05 was considered significant .
|
Candida auris has emerged as a major health concern to hospitalized patients and nursing home subjects . C . auris strains display multidrug resistance to current antifungal therapy and cause lethal infections . We have determined that C . auris harbors homologs of C . albicans Als cell surface proteins . The C . albicans NDV-3A vaccine , harboring the N-terminus of Als3p formulated with alum , generates cross-reactive antibodies against C . auris clinical isolates and protects neutropenic mice from hematogenously disseminated C . auris infection . Importantly , the NDV-3A vaccine displays an additive protective effect in neutropenic mice when combined with micafungin . Due to its proven safety and efficacy in humans against C . albicans infection , our studies support the expedited testing of the NDV-3A vaccine against C . auris in future clinical trials .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"blood",
"cells",
"flow",
"cytometry",
"graduates",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"immune",
"cells",
"pathogens",
"immunology",
"microbiology",
"alumni",
"vaccines",
"animal",
"models",
"fungi",
"model",
"organisms",
"experimental",
"organism",
"systems",
"educational",
"status",
"infectious",
"disease",
"control",
"antibodies",
"immunologic",
"techniques",
"fungal",
"pathogens",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"mycology",
"white",
"blood",
"cells",
"spectrum",
"analysis",
"techniques",
"animal",
"studies",
"proteins",
"medical",
"microbiology",
"animal",
"cells",
"microbial",
"pathogens",
"immunoassays",
"mouse",
"models",
"spectrophotometry",
"yeast",
"people",
"and",
"places",
"biochemistry",
"cytophotometry",
"candida",
"macrophages",
"eukaryota",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"population",
"groupings",
"cellular",
"types",
"organisms",
"candida",
"albicans"
] |
2019
|
The NDV-3A vaccine protects mice from multidrug resistant Candida auris infection
|
Hair cells of the mammalian cochlea are specialized for the dynamic coding of sound stimuli . The transduction of sound waves into electrical signals depends upon mechanosensitive hair bundles that project from the cell's apical surface . Each stereocilium within a hair bundle is composed of uniformly polarized and tightly packed actin filaments . Several stereociliary proteins have been shown to be associated with hair bundle development and function and are known to cause deafness in mice and humans when mutated . The growth of the stereociliar actin core is dynamically regulated at the actin filament barbed ends in the stereociliary tip . We show that Eps8 , a protein with actin binding , bundling , and barbed-end capping activities in other systems , is a novel component of the hair bundle . Eps8 is localized predominantly at the tip of the stereocilia and is essential for their normal elongation and function . Moreover , we have found that Eps8 knockout mice are profoundly deaf and that IHCs , but not OHCs , fail to mature into fully functional sensory receptors . We propose that Eps8 directly regulates stereocilia growth in hair cells and also plays a crucial role in the physiological maturation of mammalian cochlear IHCs . Together , our results indicate that Eps8 is critical in coordinating the development and functionality of mammalian auditory hair cells .
The mechanoelectrical transduction of sound information is made possible by sensory hair cells ( inner and outer hair cells ) in the cochlea [1] . The initial step in the sound transduction cascade is performed by mechanically gated ion channels positioned near the tips of hair cell stereocilia . Stereocilia are microvilli-like structures that protrude from the apical surface of hair cells , with a core composed of tightly packed actin filaments [1] , [2] . Their lengths are scaled precisely to form bundles of stereocilia ( hair bundle ) with a staircase-like architecture [3] , [4] . Each hair bundle is composed of two or more rows of stereocilia that are coupled to one another by extracellular links of several types [2] . The staircase mainly develops postnatally when stereociliary elongation stops initially in the shortest rows at around postnatal day 5 ( P5 ) and the tallest row at about P15 in mice [3] . The height of stereocilia within a row is similar not only within a single hair bundle but also in the bundles of closely adjacent hair cells , indicating that the polymerization and depolymerization of their F-actin core is tightly regulated [4] . Several genes encoding for stereociliary proteins , including whirlin [5] , [6] , espin [7] , [8] , and the unconventional myosins VIIa [9] and XVa [10] , have been shown to cause deafness when mutated [2] . Although these proteins are important for the correct regulation of hair bundle length and development , they are unlikely to control actin polymerization directly [2] . Recently , it has been shown that the novel stereociliary protein twinfilin 2 , an actin filament barbed-end capping protein located only at the tips of the short and middle rows of stereocilia in IHCs , is able to control actin dynamics in developing and mature hair bundles by restricting their excessive elongation [3] . However , the nature of the protein ( s ) regulating actin dynamics in the tallest stereocilia remains unknown . Epidermal growth factor receptor pathway substrate 8 ( Eps8 [11] , [12] ) is an evolutionarily conserved signal transducer endowed with multiple functions in the control of actin dynamics and in the integration of these events with other receptor-activated signaling functions . Depending on its association with other signal transducers , Eps8 can regulate the activation of the Rac GTPase , a master regulator of actin remodeling [13]–[15] , and integrate cellular signaling and membrane EGF receptor internalization [16] . Eps8 can also act directly on actin by binding to it and exerting both actin bundling and actin barbed-end capping activity [14] , [17] , [18] . At the functional level , Eps8 has been implicated in axonal filopodia growth [19] , in modulating the activity of the NMDA receptor [20] , and in regulating the length of intestinal microvilli , which were found to be shorter in its absence [17] , [21] . Given the structural similarity between microvilli and stereocilia , we hypothesized that Eps8 may also be important for hair cell stereocilia growth and more generally to hair-cell function . To test these hypotheses we undertook a structural and an in-depth physiological investigation of Eps8 knockout mice . We report that hair cells from these mice indeed exhibit shorter stereocilia . Moreover , Eps8 knockout mice are deaf and the normal physiological development of IHCs is prevented . Thus our results identify a novel critical regulator of one of the most distinctive functional refinements of the mammalian auditory system .
The localization of Eps8 in cochlear hair cells was determined by performing immunofluorescence and post embedding immunogold labeling on immature and adult wild-type control mice . Eps8 expression was detected at the tip of the stereocilia of cochlear hair cells ( Figure 1A–D and Figure S1A–C ) . The specificity of the antibody was verified on hair cells from Eps8 knockout mice ( Figure S1D–F ) . The expression pattern of Eps8 was also confirmed by immunogold labeling ( Figure 1E , example for an adult IHCs ) , with a pattern remarkably reminiscent of the localization of Eps8 at the tip of intestinal microvilli [17] . Punctate expression of Eps8 , similar to that observed in the stereocilia ( Figure 1A–E ) , was also present in the IHC cytoplasm ( Figure 1F , G ) and was not observed in the absence of the primary antibody for Eps8 ( unpublished data ) . This pattern of Eps8 expression was further confirmed by immunogold experiments where gold particles were observed in the cytoplasm of IHCs ( density of 3 particles/µm2 measured from 10 IHCs ) at higher levels than over the tissue-free resin of the section . These findings indicate that there is likely to be some expression of Eps8 in the cell cytoplasm . Scanning ( SEM: Figure 2A–C for IHCs and Figure 2D–F for OHCs ) and transmission electron microscopy ( TEM: Figure 3 for IHCs ) were used to investigate the hair bundle morphology in Eps8 knockout mice . IHC stereocilia from Eps8 knockout mice were shorter than those of wild-type animals ( Figures 2 and 3 ) . Hair bundles of IHCs from Eps8 knockouts also had additional rows of stereocilia ( five or six instead of the typical three to four rows: Figure 2A , B ) , resembling those of immature hair cells [22] . OHCs from knockout mice also had shorter and additional rows of stereocilia compared to control cells ( Figure 2D , E ) . Despite the defects associated with the absence of Eps8 , the staircase-like architecture of hair bundles [4] was preserved , although shallower ( Figures 2 and 3 ) , and tip links were present in both immature and adult knockout hair cells ( Figure 2C and F for an adult IHC and OHC , respectively ) . We measured the lengths of individual stereocilia in IHCs from TEM images . We found that in knockout IHCs the first three rows of stereocilia were significantly shorter when compared to controls ( Figure 3 ) . The height of tall stereocilia ( Figure 3B ) was 4 . 1±0 . 2 µm ( n = 9 ) in control compared with 1 . 4±0 . 1 µm ( n = 8 ) in knockout cells ( p<0 . 0001 ) with an overall reduction of 65% ( Figure 3C ) . The height of the intermediate ( control: 1 . 5±0 . 1 µm , n = 8; knockout: 1 . 3±0 . 1 µm , n = 9 , p<0 . 01 ) and first shortest row ( Short 1 , control: 1 . 2±0 . 1 µm , n = 8; knockout: 1 . 0±0 . 1 µm , n = 7 , p<0 . 05 ) stereocilia were also found to be reduced by 13% and 15% , respectively , in knockout cells . The second shortest row ( Short 2 ) stereociliary height did not change significantly between controls and knockout cells ( control: 0 . 9±0 . 1 µm , n = 5; knockout: 0 . 8±0 . 1 µm , n = 7 ) . The physiological consequences of the absence of Eps8 in hair cells were investigated by testing hearing function in Eps8 knockout adult mice ( P30–P57 ) using auditory brainstem responses ( ABRs ) and electrocochleography . Eps8 knockouts were profoundly deaf since ABRs , which reflect the activity of the afferent auditory pathway and IHCs , could only be elicited in response to unphysiologically high sound stimulus levels ( pure tone threshold in knockout mice was 110–120 dB SPL: Figure 4 ) . ABR thresholds for broadband click ( Figure 4A ) , noise pulse stimuli ( unpublished data ) , and frequency-specific pure tone stimulation ( Figure 4B ) in Eps8 knockout mice were significantly higher than those in control mice ( p<0 . 001 for click and noise pulse and pure tone ) . In order to investigate cochlear function in more detail we used electrocochleography ( CAP , SP , and CM ) and DPOAEs . Compound action potentials ( CAPs ) , which represent the firing activity of auditory afferent fibres , were greatly reduced in Eps8 knockout mice compared to those in control animals ( Figures S2 ) . The thresholds for eliciting summating potentials ( SPs: reflecting the summation of IHC depolarization ) were also significantly higher in Eps8 knockout mice ( Figures S3 ) . The endocochlear potential ( EP ) , which reflects the electrical driving force for mechanoelectrical transduction , was found to be similar between Eps8 knockout ( 81 . 3±6 . 5 mV ) and control mice ( 85 . 1±5 . 3 in 10–11-wk-old mice ) , indicating that the driving force for generating the IHC receptor potential is likely to be normal . The activity of OHCs was investigated by measuring distortion product otoacoustic emissions ( DPOAEs ) and cochlear microphonics ( CM ) . In knockout animals both DPOAEs and CM thresholds were increased to stimulus levels close to the technical detection limit ( Figure S4 ) . The reduced physiological responses in knockout mice could be related to hair cell abnormalities such as in the mechanoelectrical transduction apparatus ( stereocilia and transducer channels ) and/or altered biophysical properties of their basolateral membrane . Mechanoelectrical transducer currents were recorded from hair cells , the hair bundles of which were stimulated with a piezo-driven fluid-jet . Transducer currents from apical-coil hair cells ( P6–P9 ) in Eps8 control and knockout mice were elicited by alternating inhibitory and excitatory bundle displacements using 50 Hz sinusoidal force stimuli [23] . When large/saturating excitatory stimuli were applied to the bundle , a large transducer current could be recorded in all control and knockout OHCs ( Figure 5A–C ) and IHCs ( Figure 5D–F ) , suggesting that the tip links observed with SEM ( Figure 2 ) were functional . It is worth noting that the hair bundles of OHCs and IHCs were stimulated from different sides ( OHCs: Figure 5A , top panel; IHCs: Figure 5D , top panel ) . This different approach , which was due to technical limitations of simultaneously positioning the recording patch electrode and the fluid jet with respect to the tissue , caused negative pressure at the tip of the jet to produce inhibitory responses in OHCs ( Figure 5A , B ) while being excitatory in IHCs ( Figure 5D , E ) . Upon moving the bundles in the excitatory direction and at negative membrane potentials , an inward transducer current could be elicited . Any resting current flowing through open transducer channels in the absence of mechanical stimulation was reduced when bundles were moved in the inhibitory direction ( i . e . away from the taller stereocilia ) , as was evident for control hair cells ( Figure 5A and D , arrows at the negative voltage step ) and knockout OHCs ( Figure 5B ) . In contrast , in most of the knockout IHCs large inhibitory bundle displacements elicited small inward currents ( Figure 5E: arrow ) instead of reducing the current available at rest ( Figure 5D ) . This anomaly was probably due to the disorganized hair bundles , where some stereocilia were oriented on the opposite side of the hair bundle compared to control cells ( Figure 2B: arrows ) . A similar behavior has also been described in myosin VIIa mutant mice [24] . OHCs , which appeared to have a less severe bundle disorganization ( Figure 2E ) , did not show abnormal directional sensitivity ( Figure 5B ) . By stepping the membrane potential from −122 mV to more depolarized values , the transducer current decreased in size at first and then reversed near 0 mV ( OHCs: controls , −5 . 4±0 . 7 mV , n = 3; knockouts −2 . 4±0 . 3 mV , n = 6; Figure 5C; IHCs: controls +1 . 2±1 . 8 mV , n = 4; knockouts −1 . 3±1 . 3 mV , n = 3 Figure 5F ) , in agreement with the non-selective permeability of MET channels to cations [25] . Note that the current became outward when excitatory bundle stimulation was applied during voltage steps positive to the reversal potential of the transducer current . At positive potentials , the larger resting transducer current , which was much more pronounced in OHCs ( Figure 5A: +93 mV , arrowhead ) than IHCs ( Figure 5D: +94 mV , arrowhead ) , is likely to be due to an increased open probability of the transducer channel resulting from a reduced driving force for Ca2+ influx [26] . The relationship between transducer current and membrane potential shows that the maximal size of the transducer current , measured in 1 . 3 mM extracellular Ca2+ , was on average 91% larger at all membrane potentials in knockout IHCs compared to that of control cells ( two-way ANOVA: p<0 . 0001 , Figure 5F ) . In OHCs the maximal size of the transducer current in Eps8 knockout mice was slightly but significantly larger than that of control cells ( overall about 18% larger , p<0 . 05 , Figure 5C ) . We have taken advantage of the less severe hair bundle disorganization in knockout OHCs to investigate whether the absence of Eps8 had any effect on the resting transducer current . In both control and Eps8 knockout OHCs , the resting current increased with membrane depolarization as previously shown in hair cells from wild-type CD-1 mice [23] . Although the resting transducer current was significantly different between control and knockout OHCs ( Figure 5G , −122 mV: p<0 . 01; +98 mV: p<0 . 0001 ) , its increase with depolarization was the same for both genotypes ( about 3 times ) , suggesting a similar Ca2+ sensitivity of the transducer apparatus . We tested whether this was also the case for IHCs by locally superfusing their hair bundle with a solution containing an endolymph-like concentration of Ca2+ ( 0 . 04 mM [27] ) . Lowering the extracellular Ca2+ concentration is known to increase both the maximum transducer current and its fraction activated at rest . Calcium is a permeant blocker of the transducer channel [28] , [29] , so the increased current amplitude in low Ca2+ is caused by the partial relief of this block . Moreover , extracellular Ca2+ causes adaptation and as such closes some transducer channels . Therefore , reducing Ca2+ influx into the transducer channel , by either depolarizing hair cells to near the Ca2+ equilibrium potential ( as shown in Figure 5A , D ) or lowering the extracellular concentration , causes an increased open probability of the channel [30] . In Eps8 mice , both phenomena were observed since decreasing the Ca2+ concentration from 1 . 3 mM to 0 . 04 mM increased the holding and maximal transducer current in both control and knockout IHCs ( Figure 5H and I ) . Increasing the extracellular Ca2+ from 1 . 3 mM to 5 mM had the opposite effect ( Figure 5H , I: tested in knockouts only ) . The overall change in the holding current and maximal transducer current in response to extracellular Ca2+ was significant in both control ( t test: p<0 . 01 and p<0 . 005 , respectively ) and knockout ( one-way ANOVA: p<0 . 005 and p<0 . 0001 ) IHCs . The maximal current , but not the holding current , was significantly larger ( P<0 . 0005 ) in knockout compared to control IHCs when the same Ca2+ concentration was used ( Figure 5I ) . The above results indicate that the biophysical properties of the transducer channel , including adaptation and the presence of a resting current , are not affected by Eps8 . The above experiments were performed on young animals ( P6–P9 ) since this age is the most reliable for recording accurate transduction currents from mouse hair cells [31] . We tested whether transduction was likely to be functional in adult hair cells by using the styryl dye FM1-43 ( Figure S5A , B ) . We did the same for immature cells as a comparison ( Figure S5C , D ) . FM1-43 is a permeant blocker of the hair cell transducer channel that has previously been used to assess the presence of the resting transducer current in hair cells [23] . The advantage of this method is that the possible presence of the resting transducer current can be determined without the need of interfering with the hair bundle , thus effectively eliminating any possible artifact resulting from the abnormal orientation of stereocilia in knockout mice . Bath application or local superfusion of FM1-43 resulted in the selective labeling of immature ( P7 ) and adult ( P15–P21 ) control and knockout hair cells ( Figure S5 ) . Together the above findings indicate that the biophysical properties of the transducer channel are not affected by the absence of Eps8 and that the abnormalities observed in knockout hair cells are the consequence of the disrupted hair bundle morphology , with IHCs being more affected than OHCs . The biophysical properties of IHCs from Eps8 knockout mice were investigated to determine whether the loss of auditory function was associated with abnormalities in the normal function or development of these cells . In pre-hearing animals ( <P12 in most rodents ) , the resting membrane potentials and size of K+ currents recorded from knockout IHCs was similar to that of controls ( Table 1 ) . Moreover , all immature IHCs investigated were able to generate spontaneous or evoked repetitive Ca2+ action potentials ( Figure 6A , B for control and knockout IHCs , respectively ) as previously shown in wild-type cells [32] . These results show that the absence of Eps8 did not interfere with the biophysical properties of pre-hearing cells . We then investigated whether IHCs were able to acquire the electrical properties characteristic of adult cells , including a rapidly activating large conductance Ca2+-activated K+ current ( IK , f [33] ) and a current carried by KCNQ4 channels ( IK , n ) with an unusually hyperpolarized activation range [34] . While both K+ currents were present in adult control IHCs ( Figure 6C ) , they were absent in Eps8 knockout cells ( Figure 6D; see also Table 1 ) , which instead retained an immature phenotype by expressing the inward rectifier K+ current IK1 [35] . The physiological consequence of failing to acquire these adult-type currents was that Eps8 knockout IHCs retained the ability of generating slow Ca2+ action potentials ( Figure 6F ) , similar to those recorded from embryonic and early postnatal IHCs [34] . They did not acquire the fast , small , and graded voltage responses ( Figure 6E ) as previously described in adult IHCs of normal CD-1 mice [33] . In knockout adult IHCs , the resting membrane potential ( Vm ) was significantly more depolarized and the cell membrane capacitance ( Cm ) significantly smaller than in control cells ( Table 1: p<0 . 005 for both measurements ) , and similar to those found in immature cells , further supporting the role of Eps8 in IHC maturation . We investigated the effect of the efferent neurotransmitter acetylcholine ( ACh ) on control and Eps8 knockout IHCs . The ACh-activated current , which is mediated by Ca2+ entering hair cells through α9α10-nAChRs and activating SK2 channels , is normally expressed in immature IHCs [36] , [37] or adult OHCs [38] but not in adult IHCs [36] , [37] . In agreement with the above findings , adult control IHCs did not respond to the extracellular application of 100 µM ACh ( Figure 6G ) . In contrast , all knockout adult IHCs showed a large ACh-activated current at around the holding potential of −84 mV ( Figure 6H ) , which further supports the requirement of Eps8 for their full physiological maturation . It is possible that Eps8 knockout IHCs could , to some extent , retain direct axosomatic olivocochlear efferent fibres of the auditory nerve that transiently modulate the electrical activity of pre-hearing IHCs before taking up their final position on adult OHCs [39] . Overall , the biophysical properties of the IHC membrane in Eps8 knockout mice suggest that Eps8 is required for IHC physiological maturation . We investigated immature and adult Eps8 knockout OHCs to determine whether the development of their basolateral membrane properties was affected as in IHCs . Immature OHCs from knockout mice exhibited biophysical characteristics similar to those measured in control cells ( Table 2 ) . However , in contrast to IHCs , the resting membrane potential and cell membrane capacitance of adult OHCs were found to be similar between control and knockout cells ( Table 2 ) . IK , n , the major current component in adult mouse OHCs and normally expressed from P8 onwards [40] , was seen in all adult OHCs investigated ( Figure 7A , B; Table 2 ) , and its size , measured in isolation as previously described [40] , did not differ significantly between control and knockout cells ( Table 2 ) . Moreover , adult knockout OHCs were sensitive to ACh ( Figure 7C , D ) , exhibited electromotile activity ( Figure 7E ) , and expressed the motor protein prestin [41] in their basolateral membrane ( Figure 7F ) . Despite the fact that IK , n and ACh-mediated responses in IHCs and OHCs are carried by the same channels , their normal developmental expression was only affected in IHCs . These findings indicate that although IHCs and OHCs share some similar biophysical properties , only the maturation of IHCs appeared to be affected by the absence of Eps8 . Since the K+ currents did not mature in knockout IHCs , we investigated whether the development of ICa and the induced exocytosis were altered in the absence of Eps8 . Exocytosis was estimated by measuring increases in cell membrane capacitance ( ΔCm ) following depolarizing voltage steps , which is generally interpreted as an indication of neurotransmitter release from presynaptic cells . The synaptic machinery of IHCs becomes more sensitive to Ca2+ as they mature , causing synaptic vesicles to be released linearly with increases in Ca2+ current [42]–[45] . We found that in the absence of Eps8 the developmental linearization of the exocytotic Ca2+ sensitivity in IHCs did not occur . In adult Eps8 knockout IHCs the maximal size of the Ca2+ current ( ICa ) was significantly larger ( p<0 . 0001 ) than that of control cells ( Figure 8A–C ) but similar to that measured in pre-hearing cells [42]–[45] . However , the corresponding ΔCm was similar between the two genotypes ( Figure 8B , C , lower panel ) . As a consequence the exocytotic Ca2+ dependence , defined as the variation in ΔCm as a function of ICa and displayed as a synaptic transfer function [42]–[45] , was significantly less linear in the knockout ( power of 3 . 4±0 . 6 , n = 5 ) than in control ( power of 1 . 2±0 . 1 , n = 5: Figure 8D ) adult IHCs and was instead comparable to that of immature cells [42] , [44] . Despite these physiological abnormalities , adult knockout IHCs appeared to show a normal distribution of both Ca2+ channels and synaptic ribbons ( Figure 8E ) . In order to account for the larger Ca2+ current in knockout IHCs , it is likely that either the Ca2+ channel density per spot is larger or single Ca2+ channel properties , such as open probability and regulation , have been affected .
Transduction of acoustic stimuli into electrical signals relies on hair bundles being oriented along the axis of mechano-sensitivity , which is critical for the optimal opening of transducer channels [1] . The maturation and maintenance of hair bundle height requires continuous turnover of the actin filaments that form the core of each stereocilium [4] , [46] . Several stereociliary proteins are known to contribute to the normal development and/or maintenance of hair bundle structure and function [2] , [47] , but none of these appear to exert direct control over actin polymerization in hair cell stereocilia . Recently it has been shown that twinfilin 2 , an F-actin barbed-end-capping and G-actin-sequestering protein that can inhibit actin polymerization , is expressed at the tips of the shorter stereocilia of IHCs and OHCs . Overexpression of twinfilin 2 in cultured IHCs resulted in a significant reduction of stereocilia length , suggesting that twinfilin 2 limits elongation of the shorter stereocilia in order to maintain the mature staircase architecture of cochlear hair bundles [3] . Gelsolin is another actin capping protein present at the tip of the shorter stereocilia on OHCs and is crucial for regulating their elongation [48] . Here we show that Eps8 is expressed at the tips of stereocilia in mammalian cochlear IHCs and OHCs , which is in agreement with recent findings in OHCs [49] . The absence of Eps8 results in abnormally short stereocilia , particularly those in the tallest row , with IHCs being substantially more affected than OHCs . Therefore , in contrast to twinfilin 2 and gelsolin , Eps8 appears to favor stereocilia elongation . Eps8 , like twinfilin 2 and gelsolin , is an actin capping protein able to inhibit the growth of actin filaments at their plus end [17] , [50] , but it also acts as a cross-linking or bundling protein and regulates microvillar morphogenesis in Caenorhabditis elegans [18] . Actin-bundling-proteins are required for forming protrusions such as filopodia and microvilli in eukaryotes [51] . Interestingly , Eps8 can switch between capping and bundling activity , with capping activity activated by Abi-1 [50] and bundling by IRSp53 [52] . Recent evidence has indicated that the capping activity of Eps8 is required for the formation of filopodia in hippocampal neurons [19] . Although Eps8 has been shown to regulate the length of intestinal microvilli in mice [21] , it is not known whether it controls this via its capping and/or bundling function . The identity of Eps8 interaction partners in hair cells is not known , although our data , together with that of a recently published study [49] , are consistent with similar activities for Eps8 in hair cell stereocilia growth and maintenance . Scanning electron microscopy of Eps8 knockout hair cells revealed the presence of tip links , which are thought to be a pre-requisite for mechano-electrical transduction [53] . We found that large transducer currents could be elicited following saturating hair-bundle stimulation in the absence of Eps8 , suggesting that the observed tip links are functional and that this novel stereociliary protein is not essential for mechano-electrical transduction in cochlear hair cells . The larger transducer current recorded in Eps8 knockout hair cells , with IHCs being more affected than OHCs , could be explained by the presence of more functional transducer channels per hair bundle . This is conceivable considering the presence of the additional rows of stereocilia , especially in knockout IHCs . If we assume that all additional stereocilia on knockout IHCs have functional transducer channels connected by tip links , then there would be twice the number of transducer channels per bundle , which would account for the larger transducer current . The presence of the resting transducer current and the similar Ca2+ sensitivity of the channel in control and knockout hair cells ( Figures 5 and S5 ) indicate that Eps8 is not involved in determining the biophysical properties of the transducer channel . This is not true for many other proteins associated with hair bundles and hearing loss [2] , [54] including myosin VIIa [24] , myosin XVa [55] , and harmonin-b [56] . IHCs from myosin XVa mutant mice have short hair bundles similar to those in Eps8 mice [55] , but they also have a series of additional defects in the transducer apparatus , including absence of tip links and loss of adaptation and Ca2+ sensitivity of the transducer current [55] . Nevertheless , the short hair bundles in Eps8 knockout hair cells could have a large impact in adult animals because the relationship between force applied and hair bundle displacement could compromise the ability to detect physiological sound pressures . This scenario is suggested by the CAP I/O relation ( Figure S2C ) and in particular by the absence of CM responses in knockout mice ( Figure S4 ) . The latter indicates that the much smaller hair bundles of OHCs ( Figure 2; see also [49] ) , which was the only abnormality observed in these cells , are unlikely to be coupled to the tectorial membrane and therefore would not be stimulated effectively in normal physiological conditions . Prior to the onset of hearing ( postnatal day 12 in most rodents ) immature hair cells of the mammalian cochlea follow a developmental program that includes the acquisition and/or elimination of different basolateral membrane proteins ( e . g . ion channels and synaptic molecules [57] ) in order to mature into fully functional sensory receptors . These proteins in adult IHCs are directly involved in shaping the receptor potential generated by the opening of the transducer channels and in triggering exocytosis at hair cell ribbon synapses [58] . Although the physiology of hair cells changes progressively through development ( Figure S6 ) , the most significant and abrupt transition occurs at the onset of hearing [57] . The prevailing hypothesis is that hair cell functional maturation is controlled by a developmental switch , which is thought to be influenced by spontaneous Ca2+ action potential activity [32] , [34] , [59] . We found that , although the basolateral properties of immature IHCs were indistinguishable between control and Eps8 knockout mice , in adult cells the absence of Eps8 was accompanied by morphological differences , such as a smaller cell size and apparent persistence of axo-somatic efferent connections . There was also a failure in the normal appearance of mature biophysical characteristics , namely K+ currents ( IK , f , IK , n ) [32] , [34] , linear exocytotic Ca2+ dependence [42]–[45] , and the down-regulation of immature-type channels such as those carrying IK1 , ICa , and ACh-responses ( Figure S6 [32] , [35] , [37] . The possibility that an absence of Eps8 only delays the normal physiological maturation of cochlear hair cells , as previously described in mice lacking thyroid hormone receptors [60] , is unlikely since the observed defects persisted in IHCs from nearly 2-mo-old Eps8 knockout mice . Calcium-dependent action potential activity in immature IHCs is required for the expression of the BK current IK , f [61] and the linear exocytotic Ca2+ dependence [62] in adult cells . It is also possible that IHC maturation depends on the presence of the resting transducer current in immature cells , especially from the second postnatal week when the endocochlear potential begins to increase [63] , since it would depolarize the hair cells and affect the action potential activity . However , the biophysical properties of pre-hearing Eps8 knockout IHCs , including action potential activity and resting mechano-electrical transducer current , appeared similar to those of control cells . This excludes the possibility that in Eps8 knockout mice the developmental switch between immature and adult IHCs is prevented by a functional defect in IHCs during immature stages . Despite the similarities between hair cells , the fact that OHCs appear unaffected in the absence of Eps8 suggests that their functional maturation is regulated somewhat differently to that of IHCs . The unconventional myosin VI , which when mutated causes hereditary deafness in mice ( Snell's waltzer ) and humans [54] , [64] , is a protein responsible for actin-based motility . In hair cells myosin VI is required for the normal developmental expression of adult-like ion channels/presynaptic proteins , most likely by affecting intracellular trafficking [65] , [66] . However , in contrast to Eps8 knockout mice , the hair-cell hair bundles in Snell's waltzer mice are profoundly disorganized [67] , which is likely to affect the biophysics of mechanoelectrical transduction ( i . e . resting transducer current and adaptation ) as also previously shown in myosin VIIa mutants [24] . A similar phenotype to that of Eps8 knockout mice has been observed in mutant mice lacking the transmembrane protein Tmc1 [68] , but in this case both IHCs and OHCs failed to mature . Tmc1 has been suggested to be involved in intracellular trafficking or , more generally , in the activation or modulation of intracellular signals associated with hair cell maturation [68] . Based on current experimental evidence , a similar effect on hair cell maturation to that of Tmc1 could also be postulated for Eps8 , whether localized in the stereocilia or in the cytoplasm ( Figure 1 ) . Currently , little is known about the specific role of Eps8 in actin remodeling in mammals . In mice , it has been shown that the absence of Eps8 causes the cytoskeleton to be more refractory to actin depolymerization ( i . e . a less dynamic cytoskeleton ) and a larger NMDA current in cerebellar granule cells [20] . Eps8 activity has also been described to regulate actin dynamics in response to extracellular ( e . g . BDNF [19] ) and intracellular ( e . g . IRSp53 [52] ) factors , suggesting that the absence of IHC maturation in knockout mice could originate from a more general/indirect role of Eps8 in regulating developmental signaling at around the onset of hearing . Mutant mice have proven to be a powerful means for identifying the molecular mechanisms responsible for the development and maintenance of normal hearing [2] , [47] , [64] . We found that in the absence of Eps8 mice are deaf and that , in the cochlea , the protein directly influences stereocilia growth and is required for IHC maturation . The exact mechanism by which Eps8 is able to control such a complex developmental program remains a major challenge for future studies .
Inner and outer hair cells ( IHCs: n = 82; OHCs , n = 58 ) from Eps8 mutant mice [20] were studied in acutely dissected organs of Corti from postnatal day 3 ( P3 ) to P51 , where the day of birth is P0 . Animals were killed by cervical dislocation in accordance with UK Home Office regulations . Cochleae were dissected as previously described [42]–[45] in normal extracellular solution ( in mM ) : 135 NaCl , 5 . 8 KCl , 1 . 3 CaCl2 , 0 . 9 MgCl2 , 0 . 7 NaH2PO4 , 5 . 6 D-glucose , 10 Hepes-NaOH , 2 sodium pyruvate , amino acids , and vitamins ( pH 7 . 5; osmolality ∼308 mmol kg−1 ) . Superfusion of hair cells with 100 µM ACh ( Sigma ) was performed with a multi-barreled pipette positioned close to the patched cells . Unless otherwise stated , all recordings were performed near body temperature ( 35–37°C ) and using 1 . 3 mM Ca2+ in the extracellular solution . All animals were genotyped as previously described [20] . Voltage and current recordings were obtained using the following intracellular solution ( in mM ) : 131 KCl , 3 MgCl2 , 1 EGTA-KOH , 5 Na2ATP , 5 Hepes-KOH , 10 sodium phosphocreatine ( pH 7 . 3 ) . The pipette intracellular solution for exocytosis measurements contained ( in mM ) : 106 Cs-glutamate , 20 CsCl , 3 MgCl2 , 1 EGTA-CsOH , 5 Na2ATP , 0 . 3 Na2GTP , 5 Hepes-CsOH , 10 Na2-phosphocreatine ( pH 7 . 3 ) ; that for mechano-electrical transduction contained ( in mM ) : 135 CsCl , 2 . 5 MgCl2 , 1 EGTA-CsOH , 2 . 5 Na2ATP , 10 sodium phosphocreatine , 5 Hepes-CsOH ( pH 7 . 3 ) . Patch pipettes were coated with surf wax ( Mr . Zogs SexWax , USA ) to minimize the fast patch pipette capacitance transient . Electrophysiological recordings were made using Optopatch ( Cairn Research Ltd , UK ) or Axopatch 200B ( Molecular Devices , USA ) amplifiers . Data acquisition was controlled by pClamp software using Digidata 1320A or 1440A boards ( Axon Instruments , CA , USA ) . Recordings were low-pass filtered at 2 . 5 kHz ( 8-pole Bessel ) or 2 . 0 kHz ( 4-pole Bessel ) , sampled at 5 kHz , and stored on computer for off-line analysis ( Origin: OriginLab , USA ) . Membrane potentials were corrected for the Rs ( IHCs: 3 . 5±0 . 3 MΩ , n = 74; OHCs: 3 . 4±0 . 3 MΩ , n = 54 ) and liquid junction potential ( Cl- and Glutamate-based intracellular solution: −4 mV and −11 mV , respectively ) . The overall average voltage-clamp time constant ( product of Rs and membrane capacitance Cm ) was 27±3 µs for IHCs ( n = 74 ) and 24±2 µs for OHCs ( n = 54 ) . Real-time changes in membrane capacitance ( ΔCm ) were measured using the Optopatch as previously described [42]–[45] . Briefly , a 4 kHz sine wave of 13 mV RMS was applied to IHCs from −81 mV and was interrupted for the duration of the voltage step . The capacitance signal from the Optopatch was amplified ( ×50 ) , filtered at 250 Hz , and sampled at 5 kHz . Capacitance changes were measured by averaging the Cm traces after the voltage step ( around 200 ms ) and subtracting the pre-pulse baseline . ΔCm was recorded while applying 30 mM TEA and 15 mM 4AP ( Fluka , UK ) and 60 µM linopirdine to reduce K+ currents . The relation between Ca2+ entry and exocytosis in IHCs ( Figure 8D ) , estimated using a synaptic transfer function [42]–[45] , was obtained by plotting ΔCm against the peak ICa for 50 ms voltage steps from −71 mV to that where the maximal ICa occurred from the I-V curves ( Figure 8B ) . Data were approximated using a power function: ( eqn . 1 ) , where N is the power . Mechano-electrical transduction was recorded by mechanically stimulating the hair bundles of immature apical-coil IHCs using a solution-filled pipette ( 8–10 µm tip diameter ) inserted into a piezoelectric disc-driven fluid jet as described before [23] . The pipette tip was positioned near to the bundles to elicit a maximal transducer current . Saturating mechanical stimuli were applied as 50 Hz sinusoids ( filtered at 0 . 25 kHz , 8-pole Bessel ) with ±40 V driving voltages . For these experiments we used KCl- or CsGlutamate-based intracellular solution . Voltage clamp protocols are referred to a holding potential of −84 mV or −82 mV . The effect of endolymph-like Ca2+ ( 20–40 µM: [27] ) was examined by using a solution containing low Ca2+ ( 40 µM Ca2+ , buffered with 4 mM HEDTA ) . During the experiments in which different extracellular Ca2+ concentrations were used ( 0 . 04 mM , 1 . 3 mM , or 5 mM ) , the hair bundles were topically superfused and the fluid jet pipette was also filled with the same solution . The presence of electromotile activity in OHCs was estimated by applying a depolarizing voltage step from the holding potential of −64 mV to +56 mV and cell length change recorded using a CCD camera , with a ×3 magnifier , attached to the microscope ( ×63 objective ) . The aim of this experiment was to assess whether Eps8 knockout OHCs retained their electromotile ability . OHC contraction was visually obvious and was typically measured as length change between the patch electrode ( positioned just below the cuticular plate ) and the nucleus region . Measurements were performed in Photoshop and were calibrated using a microscope grid ( 20 µm = 520 pixels ) . Our data ( control: 1 . 19±0 . 12 µm , n = 5; knockout: 1 . 20±0 . 22 µm , n = 6 ) are in agreement with previous recordings using a similar technique ( 13 nm/mV [69] , giving 1 . 56 µm contraction for a similar 120 mV range to that used in our study ) . Statistical comparisons of means were made by Student's two-tailed t test or , for multiple comparisons , analysis of variance , usually one-way ANOVA followed by the Tukey test . Mean values are quoted ± s . e . m . where p<0 . 05 indicates statistical significance . Stock solutions of 1 or 3 mM FM1-43 ( n- ( 3-triethylammoniumpropyl ) -4- ( 4- ( dibutylamino ) styryl ) pyridiniumdibromide: Molecular Probes or Invitrogen ) were prepared in water . FM1-43 dye labeling was studied using bath or topical application methods . After dissections the apical coils of control and knockout cochleae ( aged P7–P21 ) were immobilized at the bottom of a microscope chamber containing normal extracellular solution . Cochleae were then bathed or superfused with a solution containing 3 µM FM1-43 for 10–15 s and immediately washed several times with normal extracellular solution . The cochleae were then viewed with an upright microscope equipped with epifluorescence optics and FITC filters ( excitation 488 nm , emission 520 nm ) using a 63× water immersion objective . Images were captured within 10–20 min from the dye application using a CCD camera ( SPOT-JNR ) . Some experiments were also performed using a confocal microscope . A total number of 20 control and 15 knockout cochleae from 11 and 8 mice , respectively , were used . These experiments were performed at room temperature ( 22–25°C ) as previously described [23] . All in vivo measurements were performed on anesthetized adult mice ( 7 controls and 8 knockouts ) . Auditory brainstem responses ( ABRs ) were performed as previously described [70] . Briefly , to record auditory brainstem responses , subdermal silver-wire electrodes were inserted at the vertex ( reference ) and ventro-lateral to the measured ear ( active ) and the back of the animal ( ground ) . Responses were measured for click and noise burst stimuli and stimulus frequencies between 2 . 0 and 45 . 2 kHz in 2 steps per octave . Responses were amplified by 94 dB and band pass filtered between 0 . 2 and 5 kHz . Stimulus sound pressure levels were typically 20–100 dB SPL , presented in steps of 5 dB . Cochleae from immature ( P5–P7 ) and adult ( P45–P46 ) control mice ( NMRI , C57BL/6 , Eps8 ) and Eps8 knockout mice were used to prepare cryosections or whole-mount preparations for immunofluorescence microscopy and processed as previously described [65] . Briefly , cochleae prepared for cryosections were dissected and fixed for 2 h with 2% paraformaldehyde , decalcified , embedded in Tissue-Tek optimal cutting temperature compound , and cryosectioned at a thickness of 10 µm . Sections were embedded with Vectashield mounting medium with DAPI ( Vector Laboratories ) . For whole-mount preparations , the whole organs of Corti were dissected out in 1×PBS and mounted on slides . Antibodies to Eps8 ( mouse , monoclonal , BD Transduction Laboratories , diluted 1∶50 ) , otoferlin ( rabbit , diluted 1∶6000 [71] ) , CaV1 . 3 ( rabbit , Alomone Laboratories , diluted 1∶50 ) , prestin ( rabbit , diluted 1∶3000 [72] ) , and CtBP2/Ribeye ( mouse , BD Transduction Laboratories , diluted 1∶50 ) were used . Primary antibodies were detected with Cy3-conjugated ( Jackson ImmunoResearch Laboratories ) or Alexa Fluor 488–conjugated secondary antibodies ( Molecular Probes ) . Sections were viewed using an Olympus BX61 microscope equipped with motorized z axis , epifluorescence illumination , and differential interference contrast ( DIC ) . Images were acquired using a CCD camera and analyzed with cellSense Dimension software ( OSIS ) . To display Ca2+ channel and ribbon distribution , cochlear slices were imaged over a distance of several µm with the coverage of the IHC synaptic region in an image-stack along the z axis ( z stack ) followed by 3-dimensional deconvolution using cellSense Dimension module with the advanced maximum likelihood estimation algorithm ( ADVMLE , OSIS ) . Figure 7F shows composite images , which represent the maximum intensity projection over all layers of the z stack . Images were processed with Photoshop . The distribution of Eps8 and prestin in apical coil hair cells was determined in at least three animals of a given age and done at least in triplicate on each . For confocal microscopy , whole-mount cochleae were counterstained for F-actin with Alexa 568 ( 1∶400 dilution ) conjugated phalloidin ( 1∶1000 dilution ) or labeled with the IHC marker VGLUT3 ( rabbit , 1∶200 dilution: Synaptic System ) [65] , mounted in Vectashield , and viewed with a Zeiss 510 Meta confocal laser scanning microscope . For SEM cochleae were excised from wild-type , including CD-1 mice , and knockout Eps8 mice ( P2–P5 and P18 ) and a hole made in the apex . They were fixed by intralabyrinthine perfusion using a fine hypodermic needle through the round window with 2 . 5% glutaraldehyde in sodium cacodylate buffer containing 2 mM calcium chloride ( pH 7 . 4 ) and then immersed in this fixative for 2 h . They were stored in fixative diluted 1/10th in buffer and subsequently dissected by removing the bone from the apical coil to expose the organ of Corti and then immersed in 1% osmium tetroxide in the same cacodylate buffer for 1 h . For osmium impregnation , which avoids gold coating , cochleae were incubated in solutions of saturated aqueous thiocarbohydrazide ( 20 min ) alternating with 1% osmium tetroxide in buffer ( 2 h ) twice ( the OTOTO technique [73] ) . They were dehydrated through an ethanol series and critical point dried using CO2 as the transitional fluid , then mounted on specimen stubs using silver conducting paint ( Agar Scientific , Stansted , UK ) , and examined in a Hitachi S4500 field emission SEM operated at 5 kV accelerating voltage . Images were obtained from >10 control and four knockout mice . For TEM cochleae from adult Eps8 knockout and wild type mice ( P22 , P26 , and P35 ) were fixed as for SEM , the shell was partially removed from both sides of the spiral , and the cochleae postfixed by immersion for 1 h in 1% osmium tetroxide in the cacodylate buffer , dehydrated , and embedded in Spurr resin using a standard protocol [74] . The apical region was exposed by cutting longitudinally through the centre of the modiolus using an annular diamond blade on a Malvern Instruments ( Malvern , UK ) 2A microslicer . Semi-thin serial sections ( 250 nm ) were cut from the apical coil and mounted on copper hole grids coated with a formvar film . They were examined unstained in a JEOL 1230 TEM operated at 100 kV accelerating voltage . To measure the height of the stereocilia , digital images were acquired using a Megaview III ( SIS systems , Olympus Microscopes Ltd ) and stereocilia were selected for measuring provided the tips and more than 75% of the shaft were evident in the section . The majority of sections contained the whole length of stereocilia . Measurement was performed using the “arbitrary length” tool on the analySIS program and was made along the long axis of the stereocilium between the tip and entry into the cuticular plate . The thickness of the stereocilia rootlets were taken from their narrowest region . For immunogold , cochleae were excised from wild-type control mice , fixed as above using 4% paraformaldehyde in 0 . 1 M sodium phosphate buffer , decalcified using 5 mM EDTA containing 1% paraformaldehyde for 3 d , dehydrated , and embedded in LR-White ( London resin company ) as previously described [75] . After slicing in a midmodiolar plane as before , 100 nm ultrathin sections of the apical coil were cut , mounted onto 200 mesh thin bar nickel grids coated with QuickCoat glue ( Agar Scientific , Stansted ) , and allowed to dry . To immunolabel them , the grids were placed in sequence in drops of the following solutions placed on a parafilm sheet in a moist chamber ( at room temperature unless otherwise stated ) : 0 . 05 M Tris buffered saline ( TBS—0 . 05 M Tris and 0 . 9% sodium chloride , pH 7 . 4 ) for 5 min , 10% goat serum ( GS ) in TBS containing 0 . 1% Tween 20 for 30 min , 1% GS-TBS containing the Eps8 primary antibody ( BD Biosciences , Germany ) diluted 1∶50 overnight at 4°C , 10% GS-TBS for 15 min , goat-anti mouse IgG secondary antibody conjugated to 15 nm gold particles ( British Biocell , UK ) diluted 1∶20 for 2 h , TBS X3 , distilled H2O X2 , and finally saturated aqueous uranyl acetate for 20 min . After washing in distilled water and drying for at least 1 h , grids were examined in a JEOL 1230 TEM operated at 100 kV . For TEM experiments we analyzed between one and two cochleae for each age/genotype tested .
|
Mammalian hearing depends on mechanosensory inner and outer hair cells within the inner ear that convert sound vibrations into electrical signals . While inner hair cells are the primary sensory receptors , outer hair cells improve auditory sensitivity . Although we know that sensory cells of the auditory , visual and olfactory systems undergo a series of regulated physiological and morphological changes during development , relatively little is known about the molecular mechanisms that regulate the development of these cells . In this study , we find that the protein Eps8 , which binds to the key structural protein actin and regulates cell growth and neural development , is an essential component of auditory hair cell development and function . We show that mice lacking Eps8 are profoundly deaf and that their mechanically sensitive hair bundles do not fully grow . However , we also show that the bundles retain their ability to transduce mechanical stimuli . Further study revealed that Eps8 has additional functions in the physiological maturation of inner hair cells and in their ability to transmit electrical information to the brain . Combined , our results provide evidence for the complex physiological role of Eps8 in hair cells and the reason why its absence causes profound deafness .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"molecular",
"neuroscience",
"developmental",
"neuroscience",
"cellular",
"neuroscience",
"biology",
"sensory",
"systems",
"neuroscience"
] |
2011
|
Eps8 Regulates Hair Bundle Length and Functional Maturation of Mammalian Auditory Hair Cells
|
TDP-43 , an RNA-binding protein that is primarily nuclear and important in splicing and RNA metabolism , is mislocalized from the nucleus to the cytoplasm of neural cells in amyotrophic lateral sclerosis ( ALS ) , and contributes to disease . We sought to investigate whether TDP-43 is mislocalized in infections with the acute neuronal GDVII strain and the persistent demyelinating DA strain of Theiler’s virus murine encephalomyelitis virus ( TMEV ) , a member of the Cardiovirus genus of Picornaviridae because: i ) L protein of both strains is known to disrupt nucleocytoplasmic transport , including transport of polypyrimidine tract binding protein , an RNA-binding protein , ii ) motor neurons and oligodendrocytes are targeted in both TMEV infection and ALS . TDP-43 phosphorylation , cleavage , and cytoplasmic mislocalization to an aggresome were observed in wild type TMEV-infected cultured cells , with predicted splicing abnormalities . In contrast , cells infected with DA and GDVII strains that have L deletion had rare TDP-43 mislocalization and no aggresome formation . TDP-43 mislocalization was also present in neural cells of TMEV acutely-infected mice . Of note , TDP-43 was mislocalized six weeks after DA infection to the cytoplasm of oligodendrocytes and other glial cells in demyelinating lesions of spinal white matter . A recent study showed that TDP-43 knock down in oligodendrocytes in mice led to demyelination and death of this neural cell [1] , suggesting that TMEV infection mislocalization of TDP-43 and other RNA-binding proteins is predicted to disrupt key cellular processes and contribute to the pathogenesis of TMEV-induced diseases . Drugs that inhibit nuclear export may have a role in antiviral therapy .
Trans-activation response ( TAR ) DNA-binding protein of 43 kDa ( TDP-43 ) is an RNA-binding protein ( as well as DNA-binding protein ) primarily present in the nucleus and important in RNA processing , mRNA transport/stability , and mRNA translation [2–4] . A variety of cellular stresses normally triggers TDP-43 to transiently shuttle into the cytoplasm and assemble into stress granules ( SGs ) . Due to an abnormality of nucleocytoplasmic transport that is known to occur in amyotrophic lateral sclerosis ( ALS ) , TDP-43 accumulates in insoluble aggregates in the cytoplasm of glia and degenerating neurons in the central nervous system ( CNS ) [5–7] . The mislocalized TDP-43 is cleaved into C-terminal fragments ( CTFs ) , phosphorylated , and/or ubiquitinated [8–10] . The importance of TDP-43 in disease pathogenesis is evidenced by the fact that mutant TDP-43 is a rare cause of familial ALS and , like wild type ( wt ) TDP-43 , is mislocalized to the cytoplasm . TDP-43 proteinopathy has been described in a number of diseases in addition to ALS [11] . Since the leader ( L ) protein of Theiler’s murine encephalomyelitis virus ( TMEV ) , a member of the Cardiovirus genus of Picornaviridae , is known to disrupt nucleocytoplasmic transport [12 , 13] , we wondered whether TDP-43 proteinopathy occurs in infections with this pathogen; however , it is known that different RNA binding proteins and different protein compositions of the nuclear pore complex are present in different cell types [14] . TMEV includes strains of two subgroups with different disease phenotypes in mice [15] . GDVII strain and other members of the GDVII subgroup do not persist , but cause an acute fatal gray matter disease . In contrast , DA strain and other members of the TO subgroup induce a subclinical acute gray matter disease followed by an immune-mediated demyelinating myelitis with virus persistence in the CNS for the life of the mouse . DA-induced demyelinating disease serves as an experimental model of multiple sclerosis ( MS ) . Here we report that TMEV infection of cultured cells causes L-dependent mislocalization of TDP-43 , and L-independent cleavage and phosphorylation of TDP-43 along with splicing abnormalities . Mislocalization and phosphorylation of TDP-43 also occurs in neuronal cells following early TMEV infection of mice , and in oligodendroglia and other glial cells in demyelinated areas 6 weeks after DA virus infection . These results suggest that TDP-43 mislocalization occurs and presumably contributes to cellular dysfunction and death in TMEV infections . An important role for TDP-43 mislocalization in TMEV-induced demyelinating disease is suggested by recent findings that TDP-43 binds to mRNAs encoding myelin genes , and that a knockdown of TDP-43 in oligodendrocytes of mice leads to demyelination and the death of this neural cell [1] .
In control mock-infected BHK-21 cells , expression of TDP-43 was primarily restricted to the nucleus ( Fig 1A ) . Following infection with DA or GDVII virus , which was detected by positive staining for TMEV VP1 capsid protein , TDP-43 was depleted from the nucleus and aggregated in the cytoplasm ( Figs 1A , 1B and S1 ) . The location of TDP-43 was juxtanuclear in structures that resembled aggresomes ( see below ) , which have been previously observed in TMEV-infected cells [16 , 17] . In addition , phosphorylated TDP-43 ( pTDP-43 ) was present in the cytoplasm of TMEV-infected cells ( Figs 1C and S2 ) . We questioned whether other RNA-binding proteins were also mislocalized to the cytoplasm in TMEV-infected cells . For this reason , we investigated the localization in cells of i ) fused in sarcoma ( FUS ) , which like TDP-43 is a cause of familial ALS when mutated , and ii ) polypyrimidine tract binding protein ( PTB ) , which is known to be mislocalized in TMEV infections , where it plays a role in TMEV translation [18 , 19] . DA infection induced cytoplasmic mislocalization of both FUS and PTB1 , one of PTB isoforms , along with TDP-43 ( Fig 1D and 1E ) . Since TMEV L protein is known to disrupt nucleocytoplasmic trafficking , we investigated TDP-43 localization following infection with mutant TMEV that had an L deletion . As predicted , DAΔL and GDVIIΔL infection failed to induce mislocalization of TDP-43 in VP1-positive cells ( Fig 1A and 1B ) , demonstrating that TDP-43 mislocalization is indeed L-dependent . In order to further confirm the importance of TMEV L in TDP-43 mislocalization , we transfected eukaryotic expression constructs pDA L and pGDVII L into BHK-21 cells . Although both of these expression constructs caused cytoplasmic mislocalization of TDP-43 in the three cell lines that were tested ( Figs 1F and S3 ) , TDP-43 was present in small aggregates in the cytoplasm rather than the aggresome that had been detected in wild type ( wt ) TMEV-infected cells . The different effect of the TMEV L expression constructs was not a result of a different level of L protein expression when compared to TMEV L protein expression ( S4 Fig ) . In order to confirm the cytoplasmic mislocalization of TDP-43 in TMEV-infected cells , we separated the nucleus and cytoplasm of cultured cells infected with TMEV ( S5 Fig ) . The results confirmed the prominent TDP-43 mislocalization in infected cells . Some TDP-43 is present in the cytoplasm of mock and TMEVΔL-infected cells presumably due to the normal shuttling of this protein from the nucleus . As noted above , the juxtanuclear location of TDP-43 seen following TMEV infection had a morphology typical of an aggresome . Vimentin surrounded these juxtanuclear structures ( Fig 2A ) , as is true in the case of aggresomes [20] . TMEV infections of L929 cells also induced a juxtanuclear aggresome that contained PTB1 ( Fig 2B ) . In contrast , TDP-43 was diffusely present in the nucleus and cytoplasm of DA- and GDVII-infected HeLa cells ( Figs 2C and S6 ) , and not in an aggresome , perhaps related to the poor growth of TMEV in these cells [21] . Aggresomes result from a remodeling of intracellular membranes to generate sites of virus replication [20] . Fig 3A shows that VP1 and double-stranded RNA ( ds-RNA ) , produced during TMEV replication , decorated the margins of aggresomes in TMEV-infected BHK-21 cells; an orthogonal view demonstrates that there is only very partial colocalization of VP1 and TDP-43 ( S7 Fig ) . DA L was present within the aggresome’s vimentin cage , while DA L* , a non-structural protein that inhibits RNase L , was in the cytoplasm , but outside the aggresome ( Fig 3B ) . Although ds-RNA was detected in DAΔL virus-infected BHK-21 cells , it tended to be present in small aggregates throughout the cytoplasm ( Fig 3C ) . VP1 generally had a similar localization to that found with dsRNA in DAΔL virus-infected cells , however , at times it was diffusely distributed in the cytoplasm , presumably related to increasing virion production over time ( see later ) . In order to assess the importance of aggresomes in TMEV infection , we made use of nocodazole , a microtubule inhibitor that interferes with aggresome formation . BHK-21 cells were exposed to nocodazole ( 10 μM , 1hr ) , and then infected with DA virus . Compared to levels obtained with no nocodazole treatment , nocodazole led to a 10-fold reduction in virus genome at an MOI of 1 , and 100-fold reduction at an MOI of 0 . 25 ( Fig 3E ) . As expected , nocodazole treatment decreased the virus titer by more than 10-fold at 12 HPI ( Fig 3F ) . In contrast , the effect of nocodazole on the level of viral genome and infectivity was relatively small in HeLa cell ( S8 Fig ) . These findings suggest that the effect of nocodozole on TMEV replication is not related to this drug’s general disruption of the cytoskeleton , but a more specific effect on aggresomes . SGs are mainly composed of stalled translation preinitiation complexes , markers such as G3BP1 , eIF3A , and TIA1 , and RNA-binding proteins including TDP-43 . These structures are cytoplasmic non-membranous structures that appear in cells exposed to various stresses , including virus infections [22] . Certain viruses are known to induce SGs while others inhibit SG formation [23] . At times of stress or following treatment with a SG inducer , there is formation of SGs < 1 μm or 1–2 μm in size ( S9 Fig ) . Borghese and Michiels [24] previously reported that DA L inhibits SG formation in HeLa cells , a human cell line . We examined this issue in HeLa cells as well as two rodent cell lines . Uninfected control BHK-21 cells have homogeneous cytoplasmic immunostaining of SG markers G3BP1 , eIF3A and TIA1 ( Figs 4 and S9 ) . In DA- and GDVII-infected ( rodent ) BHK-21 and L929 cells , but not in infected HeLa cells , these markers are located in the aggresome of VP1-expressing cells and not in SGs ( Figs 4A–4C and S10 ) ; the lack of aggresome formation in TMEV-infected HeLa cells may be associated with the inefficient TMEV infection described in these cells [21] . At times , a VP1-expressing BHK-21 cell expressed these markers in what appeared to be typical SG structures as well as aggresomes , suggesting that the markers ( and RNA-binding proteins ) may transiently assemble in SGs , and then over time , when there is increasing virus production , relocalize in aggresomes ( Fig 4D ) . Other picornavirus infections are reported to also transiently induce SG formation , followed by an inhibition of SGs later in infection [23] . In the case of TMEVΔL virus-infected cells , typical SGs were induced that immunostained with G3BP1 , eIF3A and TIA1 ( Figs 4E , 4F and S10 ) , indicating that L interferes with SG formation . The SGs induced by TMEVΔL virus infections rarely colocalized with TDP-43 and PTB1 ( Fig 4G and 4H ) . To determine whether TMEV infection induces cleavage of TDP-43 , as in the case of ALS , we carried out Western blots on RIPA-soluble and insoluble ( but urea soluble ) fractions extracted from TMEV-infected BHK-21 cell lysates at 8 HPI . Following infection with both wt and TMEVΔL virus , ~35-kDa and ~25-kDa bands as well as the expected 43-kDa band of full-length TDP-43 were detected in the urea-soluble , but not RIPA-soluble fraction , of BHK-21 cell lysates ( Fig 5A ) . These findings suggest that L-independent cleavage of TDP-43 occurs in BHK-21 cells . Of note , there was no clear correlation between TDP-43 cleavage and TMEV infection , as monitored by VP1 immunodetection . TDP-43 is known to have an important role in alternative splicing , including cystic fibrosis transmembrane conductance regulator ( CFTR ) exon 9 skipping [25] . In order to assess splicing abnormalities in infected cells , we transfected L929 cells with a CFTR minigene construct . Compared to uninfected cells , DA or GDVII virus-infected cells had a decrease of the lower band , which corresponds to the exon 9 spliced product ( Fig 5B and 5E ) . These results provide evidence of impaired splicing regulatory activity in the infected cells , presumably because of abnormal TDP-43 localization and aggregation associated with TMEV infection . In the case of ALS and ALS/FTD , TDP-43 is depleted in nuclei of neural cells , and mislocalized and phosphorylated in inclusions in the cytoplasm ( Fig 6A and 6B ) . In order to determine whether the findings that we observed in cultured cells were also present in TMEV-induced disease , we carried out immunohistochemical staining of the CNS of mice 1 week following infection with GDVII virus , a time when mice are paralyzed and near moribund . Neurons in the CA1 region of the hippocampus had VP1 immunostaining ( Fig 6C ) with mislocalization of TDP-43 to the cytoplasm ( Fig 6D and 6E ) . pTDP-43 was present in the nucleus ( Fig 6G ) and cytoplasm ( Fig 6H ) , at times in a compact cytoplasmic inclusion body ( Fig 6I ) . Approximately 60% of VP1-positive cells in TMEV-infected mice had evidence of pTDP-43 ( S11 Fig ) . In contrast , TDP-43 maintained its normal nuclear localization in uninfected CA3 region neurons from the same TMEV-infected mouse ( Fig 6F ) and in normal uninfected mice ( S11 Fig ) . The spinal cord of infected mice showed perivascular mononuclear infiltrates ( Fig 6J ) with numerous VP1-positive anterior horn cells ( Fig 6K and 6L ) that had TDP-43 and PTB2 depleted from the nucleus ( Fig 6M and 6N ) . Immunofluorescent staining confirmed the presence and aggregation of TDP-43 in the cytoplasm of VP1-positive motor neurons ( Fig 6O ) . DA virus produces a biphasic disease in SJL mice with minimal or subclinical disease within the first two weeks post-infection , followed by progressive paralysis from an inflammatory demyelination that peaks at 6 weeks post-infection . In the acute phase of DA virus infection , VP1-positive neurons and axons were present in the CA2 region of the hippocampus ( Fig 7A and 7B ) ; however , the severity of infection and inflammation was mild compared to that seen in GDVII virus-infected mice . Some cells appeared to have cytoplasmic as well as nuclear staining of TDP-43 and PTB2 , a splicing isoform of PTB that is increased in neurons compared to other cell types ( Fig 7C and 7D ) . The infected regions generally had a decrease in TDP-43 staining , perhaps partly because many of the infected cells had pTDP-43 ( Fig 7E ) , which was not stained by the anti-TDP-43 antibody that was used . Six weeks after infection with DA virus , the ventral region of the thoracic spinal cord showed perivascular mononuclear cell infiltrates , ( Fig 7F ) , demyelination , and vacuolation ( Fig 7G ) . Activated microglia clustered within or around the demyelinated areas ( Fig 7H ) . In these demyelinated areas , TDP-43 was depleted from the nucleus and mislocalized to the cytoplasm of VP1-positive white matter glial cells ( Fig 7I and 7J ) , including oligodendrocytes ( Fig 7K ) .
TDP-43 is a ubiquitously expressed RNA-binding protein that predominantly resides in the nucleus , but shuttles across the nuclear membrane in association with mRNAs [26] . A hallmark of almost all cases of ALS is disruption of nucleocytoplasmic trafficking with cytoplasmic mislocalization , aggregation , cleavage , and phosphorylation of TDP-43 in neural cells [5 , 7 , 9] . TDP-43 mislocalization is thought to lead to abnormalities of splicing and RNA metabolism with subsequent neuronal dysfunction [4 , 27 , 28] . It is likely that the cytoplasmic mislocalization of other RNA-binding proteins also contributes to the abnormalities of splicing in ALS [29] . In the present study , we demonstrate that TMEV infection leads to cytoplasmic mislocalization of TDP-43 ( as well as FUS and PTB ) along with cleavage into products similar in size to those found in ALS [7] and TDP-43 phosphorylation . Importantly , TDP-43 mislocalization was also found in neural cells following acute infections of mice , and in oligodendrocytes and other glial cells in demyelinated regions 6 weeks after DA infection . As is true of many pathogens , picornaviruses disrupt nucleocytoplasmic trafficking during infection , leading to cellular dysfunction as well as the redistribution and hijacking of nuclear proteins into the cytoplasm for use during virus replication [30 , 31] . For example , in infections of cultured cells by coxsackievirus B3 ( CVB3 ) , a member of the Enterovirus genus of Picornaviridae , TDP-43 is mislocalized ( by viral protease 2A ) and cleaved ( by viral protease 3C ) [32] . In CVB3 infections , TDP-43 colocalized with SGs in the cytoplasm at 3 HPI , the longest time observed . Human immunodeficiency virus-positive neural cells have also been reported to have TDP-43 in cytoplasmic inclusions [33] . In ALS , TDP-43 is thought to shuttle into the cytoplasm initially into SGs , and then remain aggregated in the cytoplasm . In the case of TMEV-infected BHK-21 and L929 cells , we detected the normal markers for SGs ( G3BP1 , TIA1 and eIF3A ) in aggresomes rather than SGs . Of note , SGs were present following TMEVΔL infections , suggesting that L inhibits SG formation , as has been reported by others [24] . The aggresomes in TMEV-infected cells also contained TDP-43 , FUS , PTB1 , TMEV proteins ( VP1 , L ) , and dsRNA . Nocodazole , a microtubule inhibitor that interferes with aggresome formation , decreased viral replication , suggesting that TMEV uses the aggresome as a “viral factory , ” perhaps by concentrating proteins and genome in one region of the cell , as described for other virus infections [34]; however , nocodazole’s disruption of the cytoskeleton with a resultant disturbance of cell physiology may also have had a substantial indirect effect on viral replication . In contrast , TMEV infection of HeLa cells led to minimal cytoplasmic translocation of TDP-43 with no aggresome formation , perhaps a reflection of the reported inefficient infection of these cells [21]; the reasons for the lack of aggresome formation and inefficient infection remain unclear . Cytoplasmic TDP-43 aggregates in ALS have also been referred to as aggresomes [35–37] . In the latter case , the aggresome is thought to be a cytoprotective response that sequesters potentially toxic misfolded proteins and facilitates their clearance by autophagy [20 , 38] . Mislocalization and phosphorylation of TDP-43 occurred in TMEV-infected cultured cells as well as neuronal and glial cells of TMEV-infected mice . In DA virus-induced demyelinated regions , TDP-43 and other RNA-binding proteins were mislocalized in glial cells , including oligodendrocytes . The fact that TDP-43 was not present in the cytoplasm following infection with TMEVΔL virus , suggests that L interfered with nucleocytoplasmic transport . TDP-43 mislocalization in neural cells may also be influenced by inflammatory stimuli since tumor necrosis factor-α can lead to mislocalization of TDP-43 [39] . In addition , interferon-γ leads to hnRNP A1 mislocalization and accumulation into the cytoplasm [40] . The mislocalization of RNA-binding proteins in TMEV infections may disrupt cellular splicing and mRNA translation , thereby contributing to neuronal dysfunction and death in GDVII and DA early disease as well as oligodendrocyte dysfunction in the late demyelinating disease of DA-infected mice . Our previous study suggested the possibility that PTB mislocalization in TMEV-infected neurons played a role in neuronal dysfunction [41] . The deleterious effect of PTB2 mislocalization in neurons may be compounded by the FUS mislocalization that was also present , since the latter RNA binding protein is important in axonal transport [42] . Importantly , recent studies have demonstrated that: i ) TDP-43 binds to mRNAs of myelin proteins , ii ) knockdown of TDP-43 in oligodendrocytes of mice leads to demyelination and death of this neural cell [1] . TMEV L-dependent nucleocytoplasmic trafficking defect is likely to also interfere with other RNA binding proteins in addition to the three that were investigated as well as to disrupt the proper subcellular localization of a number of key transcription factors and proteins in oligodendrocytes and oligodendrocyte precursor cells that are needed for efficient myelination and remyelination [43–46] . Altered nucleocytoplasmic transport leading to mislocalization of RNA-binding proteins and other macromolecules with associated cellular dysfunction may underlie a number of disease states , both infectious as well as non-infectious . The importance of this mechanism of cell dysfunction highlights the potential relevance of antiviral drugs that target nucleocytoplasmic transport .
The study involving the analysis of human subjects was approved by The University of Chicago Institutional Review Board for Clinical Research . Informed written consent for an autopsy was obtained from an immediate member of the deceased’s family . Animal use was approved by The University of Chicago Institutional Animal Care and Use Committee ( IACUC ) under the Protocol Number 71772 . Animal work conducted at the University of Chicago complies with all applicable provisions of the Animal Welfare Act ( AWA ) and the Public Health Service ( PHS ) Policy on Humane Care and Use of Laboratory Animals . The PHS Policy incorporates the standards in the Guide for the Care and Use of Laboratory Animals and the U . S . Government Principles for the Utilization and Care of Vertebrate Animals Used in Testing , Research and Training and requires euthanasia be conducted according to the AVMA Guidelines for the Euthanasia of Animals . The University of Chicago Animal Care Program has an approved Assurance with the National Institute of Health ( NIH ) , is registered with the United States Department of Agriculture ( USDA ) and is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International ( AAALAC ) . DA and GDVII viruses were derived from a full-length infectious cDNA clone [15] . DAΔL virus has a deletion of amino acids 2 to 67 of L [47] . GDVIIΔL virus ( originally referred to as dl-L virus ) [48] has a deletion of amino acids 2 to 71 of L , and was previously received as a gift from M . K . Rundell . Infections were carried out in BHK-21 ( ATCC , CCL-10 ) , L929 ( ATCC , CRL-6364 ) or HeLa cells ( ATCC , CCL-2 ) , usually with a multiplicity of infection ( MOI ) of 10 . BHK-21 cells were used for plaque assays and the growth of virus stocks , as previously described [49] . For the study of TDP-43 cleavage , cells were treated with 1μM of the proteasome inhibitor MG-132 ( Cell Signaling Technology , Danvers , MA ) for 16hs prior to harvest . For induction of SGs , BHK-21 cells were treated for 45 min with 0 . 5mM sodium arsenite ( Sigma Aldrich , St Louis , MO ) . In investigations of the aggresome , nocodazole ( Sigma Aldrich ) , a microtubule inhibitor , was solubilized in DMSO and added at varying concentrations to the culture medium for 1h prior to infection . pDAL and pGDVIIL , which are eukaryotic expression constructs of DA L and GDVII L respectively , with myc/His epitope tags at the carboxyl terminus [47] , were transfected into BHK-21 cells using Lipofectamine 3000 ( Thermo Fisher Scientific , Waltham , MA ) . Cells on coverslips were harvested 8hs post infection ( HPI ) or 48hs after transfection , fixed in 4% paraformaldehyde for 5 min , and then permeabilized with phosphate buffered saline ( PBS ) with 0 . 1% Triton X-100 for 20 min at room temperature . The coverslips were then incubated overnight at 4°C with primary antibodies ( S1 Table ) . After rinsing , cells were incubated for 30 min with Alexa 594-conjugated goat anti-mouse IgG and Alexa 488-conjugated goat anti-rabbit IgG ( Invitrogen , Carlsbad , CA ) , and then counterstained with 4' , 6-diamidino-2-phenylindole ( DAPI ) . Images were captured using a confocal laser microscope system ( Leica TCS SP5 , Leica Microsystems , Wetzlar , Germany ) . A sequential multiple fluorescence scanning mode was used to avoid nonspecific overlap of signals . In some experiments , manual counting of infected cells was carried out in five different regions of the coverslips . Cells were lysed 8 HPI with a radioimmunoprecipitation assay ( RIPA ) buffer containing a protease inhibitor and phosphatase inhibitor cocktail ( Thermo Fisher Scientific , Waltham , MA ) . Lysates were centrifuged at 14 , 000 rpm for 30 min at 4°C , and supernatants collected as RIPA buffer-soluble proteins . The pellets were sonicated and centrifuged twice at 14 , 000 rpm for 30 min at 4°C to obtain RIPA buffer-insoluble pellets . Pellets were dissolved in urea buffer ( 8 M urea , 50 mM Tris-HCl , pH 8 . 5 ) and then sonicated again prior to electrophoresis . Ten μg of total protein quantified by a Pierce BCA Protein Assay Kit ( Thermo Fisher Scientific , Waltham , MA ) was subjected to electrophoresis on 10% SDS polyacrylamide gels , and then transferred to Amersham Hybond P 0 . 45 μm PVDF membrane ( GE Healthcare , Buckinghamshire , UK ) . The membrane was first blocked with 5% non-fat skim milk in Tris-buffered saline ( TBS ) containing 0 . 05% Tween-20 for 30 min at room temperature , and then incubated for 1h at room temperature with a rabbit antibody directed against C-terminal TDP-43 ( 1:1000 , Proteintech , Rosemont , IL ) or a mouse monoclonal antibody against TMEV VP1 ( 1:2000 ) , which was previously called GDVII mAb2 [50] , or a mouse monoclonal antibody against Lamin A/C ( 1:1000 , Cell Signaling Technology , Danvers , MA ) , or a mouse monoclonal antibody against β–actin ( 1:5000 , Sigma Aldrich , St Louis , MO ) . Following washing , the membrane was incubated with anti-rabbit or anti-mouse horseradish peroxidase–conjugated secondary antibodies ( GE Healthcare , Buckinghamshire , UK ) for 1h at room temperature . The signal was detected using SuperSignal West Dura Extended Duration Substrate ( Thermo Fisher Scientific , Waltham , MA ) , and analyzed using ChemiDoc MP Imaging System ( Bio-Rad Laboratories , Hercules , CA ) . DA RNA was extracted from BHK-21 and HeLa cell homogenates using an RNeasy Plus minikit ( Qiagen ) . A region between nt 1485 and 1684 was amplified using forward primer TACTATGGCACCTCTCCTCTTGGA and reverse primer CAGCCGCAAGAACTTTATCCGTTG with a Superscript III Platinum two-step qRT-PCR kit with SYBR green ( Invitrogen ) . A region between nt 182 and 721 of the murine β-actin gene , which was used for normalization and determination of the quality of total mRNA , was amplified using forward primer GTGGGCCGCTCTAGGCACCAA and reverse primer CTCTTTGATGTCACGCACGATTTC . qRT-PCR was conducted on a CFX96 Real-Time System ( Bio-Rad ) . The ΔΔCT method of relative quantitation was used to calculate fold change of DA with β-actin . In order to assess splicing in TMEV-infected cells , a cystic fibrosis transmembrane conductance regulator ( CFTR ) minigene construct designed to evaluate CFTR exon 9 splicing ( a gift from Virginia Lee’s lab and described in Buratti et al . [25] ) was transfected into L929 cells using Lipofectamine LTX reagent ( Invitrogen ) . Twenty-four hours later , the cells were infected separately with DA or GDVII viruses at an MOI of 10 . Total RNA was prepared from cells 12 h after infection of viruses , and RT-PCR was performed with 1 μg of total RNA and 1 μl of resulting cDNA . The relative exclusion of exon 9 was evaluated by primer extension from the flanking sequence of exon 9 using the following primers , as previously described [25]: TAGGATCCGGTCACCAGGAAGTTGGTTAAATCA; CAACTTCAAGCTCCTAAGCCACTGC . The PCR products were visualized on a 2% agarose gel . Relative amounts of different splice products were quantified and visualized using Image J . The experiments were repeated in triplicate . Immunohistochemical studies were performed on autopsied brain specimens of a patient with ALS . TMEV was inoculated intracerebrally in weanling SJL mice ( Jackson Laboratory , Bar Harbor , ME ) , and mice were sacrificed at 1 , 2 or 6 weeks post infection ( PI ) . At the time of sacrifice , mice were deeply anesthetized and perfused transcardially first with PBS , and then with 4% paraformaldehyde in 0 . 1 M phosphate buffer . CNS tissues were fixed in 10% buffered formalin and processed into paraffin sections ( 5 μm thick ) . Deparaffinized sections were hydrated in ethanol and then incubated with 0 . 3% hydrogen peroxide in absolute methanol for 30 min at room temperature to inhibit endogenous peroxidase . After rinsing with tap water , sections were washed twice using Tris–HCl with 0 . 1% Triton X-100 for 5 min , and then with Tris–HCl for 5 min . Sections were then incubated at 4°C overnight with primary antibody ( S2 Table ) diluted in 5% normal goat serum , 50 mM Tris-HCl ( pH 7 . 6 ) and 1% BSA . After rinsing , sections were subjected to labeling by an enhanced indirect immunoperoxidase method . The reaction product was developed using a solution of 3 , 3’-diaminobenzidine ( DAB ) . Sections were counterstained with hematoxylin . Double immunostaining was carried out with two enzyme systems , peroxidase and alkaline phosphatase , followed by staining with Vector Red ( Vector Laboratories ) . Paraffin sections were also used for immunofluorescent staining . Sections were deparaffinized in xylene , rehydrated through an ethanol gradient , and then incubated with primary antibody for 1h at room temperature . The following antibodies were used: mouse monoclonal anti-VP1 , rabbit anti-TDP-43 , and rabbit anti-2' , 3'-cyclic nucleotide-3'-phosphodiesterase ( CNPase ) ( S2 Table ) ; rabbit anti-TDP-43 was pre-conjugated with Zenon Alexa Fluor 647 rabbit IgG ( Invitrogen ) . After rinsing , sections were incubated for double immunofluorescence with Alexa 488-conjugated goat anti-rabbit IgG and Alexa 594-conjugated goat anti-mouse IgG ( Invitrogen ) and for triple immunofluorescent staining with Alexa 488-conjugated goat anti-rabbit IgG , Alexa 555-conjugated goat anti-mouse IgG ( Invitrogen ) , and DAPI . Images were captured using a confocal laser microscope system ( Leica TCS SP5 , Leica Microsystems , Wetzlar , Germany ) with sequential multiple fluorescence scanning mode to avoid non-specific overlap of colors . All photographs were captured under the same magnification , laser intensity , gain and offset values , and pinhole setting . Statistical analysis was performed by an unpaired t-test or one-way ANOVA with Tukey's multiple comparisons test using GraphPad Prism version 7 . 0a . A P-value of <0 . 05 was considered significant . The data are presented as the mean ± standard deviation ( S . D . ) .
|
TDP-43 is a widely expressed nuclear protein that shuttles between the nucleus and cytoplasm , and regulates many aspects of RNA processing , such as splicing , trafficking , stabilization , and miRNA production . In almost all cases of ALS , neuronal and glial TDP-43 is phosphorylated , cleaved , and mislocalized to the cytoplasm , where it aggregates into stress granules and insoluble inclusion bodies . Although the mechanisms involved in TDP-43 proteinopathy remain unclear , impaired nucleocytoplasmic trafficking is thought to play an important role . Here we investigated whether TDP-43 proteinopathy also occurs during TMEV infection since TMEV L protein is known to perturb nucleocytoplasmic transport . We found evidence of TDP-43 proteinopathy in both TMEV-infected cultured cells , with predicted splicing abnormalities , as well as in neural and glial cells of TMEV-infected mice . The findings suggest that TDP-43 may contribute to the pathogenesis of TMEV-induced diseases , including TMEV-induced immune-mediated demyelination .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"phosphorylation",
"immunofluorescence",
"staining",
"hela",
"cells",
"biological",
"cultures",
"neuroscience",
"cytoplasmic",
"staining",
"cell",
"cultures",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"animal",
"cells",
"proteins",
"cell",
"lines",
"cytoplasm",
"biochemistry",
"cellular",
"neuroscience",
"cell",
"staining",
"cell",
"biology",
"post-translational",
"modification",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cultured",
"tumor",
"cells",
"cytoplasmic",
"inclusions"
] |
2019
|
TDP-43 proteinopathy in Theiler’s murine encephalomyelitis virus infection
|
Phagocytosis is an ancient mechanism central to both tissue homeostasis and immune defense . Both the identity of the receptors that mediate bacterial phagocytosis and the nature of the interactions between phagocytosis and other defense mechanisms remain elusive . Here , we report that Croquemort ( Crq ) , a Drosophila member of the CD36 family of scavenger receptors , is required for microbial phagocytosis and efficient bacterial clearance . Flies mutant for crq are susceptible to environmental microbes during development and succumb to a variety of microbial infections as adults . Crq acts parallel to the Toll and Imd pathways to eliminate bacteria via phagocytosis . crq mutant flies exhibit enhanced and prolonged immune and cytokine induction accompanied by premature gut dysplasia and decreased lifespan . The chronic state of immune activation in crq mutant flies is further regulated by negative regulators of the Imd pathway . Altogether , our data demonstrate that Crq plays a key role in maintaining immune and organismal homeostasis .
Mounting appropriate immune responses against pathogens is critical for the survival of all animals . Mechanisms to both eliminate microbes and resolve infection by returning the immune system to basal activity are necessary to maintain an adequate and balanced immune response [1 , 2] . Alterations in these responses can lead to immune deficiency or auto-inflammation [3–5] . Yet , to date , how these mechanisms are coordinated upon infection remains unclear . Drosophila is a prime model to genetically dissect humoral and cellular innate immune responses to a variety of pathogens [6–8] . Humoral responses include the pro-phenoloxidase ( PO ) cascade , which leads to the generation of reactive oxygen species and melanization , and the rapid production of antimicrobial peptides ( AMPs ) regulated by the Toll and Imd pathways [7] . Upon recognition of microbial lysine ( Lys ) -type peptidoglycan ( PGN ) , damage-associated molecular patterns ( DAMPs ) , or exogenous protease activity , the Toll pathway promotes the nuclear translocation of the NF-κB-like transcription factor Dorsal-related Immune Factor ( Dif ) to induce AMP genes , such as Drosomycin [6 , 9] . In contrast , detection of bacterial meso-diaminopimelic acid ( DAP ) -type peptidoglycan activates the Imd pathway and leads to the nuclear translocation of the NF-κB-like transcription factor Relish ( Rel ) to induce transcription of AMP genes , such as Diptericin [10 , 11] . It has also been shown that proteases , such as Elastase and Mmp2 , can activate the Imd pathway through cleavage of the receptor PGRP-LC [12] . As in mammals , chronic activation of immune responses is deleterious to the fly , and negative regulators are required to maintain immune homeostasis [13–15] . For instance , amidase PGN recognition proteins ( PGRPs ) , such as PGRP-LB and PGRP-SC , negatively regulate the Imd pathway by enzymatically degrading PGN [14–16] . Phagocytosis and encapsulation are key cellular innate immune responses [7] . Phagocytosis allows for the uptake and digestion of microbes and apoptotic cells by phagocytes , including specialized immune cells called plasmatocytes [7 , 17] . Encapsulation results in the isolation and melanization of large materials , such as wasp eggs or damaged tissues , by dedicated immune cells named lamellocytes [18] . Both phagocytosis and humoral responses are required to fight infection . Indeed , decreasing the phagocytic ability of plasmatocytes by pre-injecting latex beads , which they take up , impairs fly survival upon infection with Gram-positive bacteria [19] . Similarly , inhibiting phagocytosis increases the susceptibility of Imd pathway-deficient flies to Escherichia coli ( E . coli ) infection , arguing that phagocytosis and the humoral response act in parallel [20] . Plasmatocytes were proposed to activate the production of AMPs by releasing immunostimulatory pathogen-associated molecular patterns ( PAMPs ) following phagocytosis [21] . They also express cytokines such as Unpaired 3 ( Upd3 ) , a ligand of the JAK-STAT pathway , which regulates immune-related genes [22] . Yet , ablation of the majority of plasmatocytes by targeted apoptosis has only a moderate effect on the fly’s ability to fight infection [23 , 24] . Therefore , the role of phagocytosis in the regulation of the humoral response and the resolution of infection remains unclear . Several plasmatocyte receptors promote the recognition and engulfment of bacteria [25] . The scavenger receptor dSR-CI and a transmembrane protein , Eater , bind to both Gram-negative and -positive bacteria [26 , 27] . The membrane receptor PGRP-LC binds to and engulfs Gram-negative but not Gram-positive bacteria , and its membrane localization is dependent on the nonaspanin TM9SF4 [28 , 29] . Draper ( Drpr ) promotes clearance and degradation of neuronal debris and apoptotic cells via phagosome maturation , as well as phagocytosis of Staphylococcus aureus ( S . aureus ) together with the integrin βv and PGRP-SC1 [30–32] . Nimrod C1 , which is related to Eater and Drpr , promotes phagocytosis of both S . aureus and E . coli by Drosophila S2 cells , and suppression of its expression in plasmatocytes inhibits phagocytosis of S . aureus [33] . Peste , a member of the CD36 family of scavenger receptors plays a role in the recognition and uptake of Mycobacterium by S2 cells [34] . Finally , croquemort ( crq ) , another CD36 family member , promotes apoptotic cell clearance by embryonic plasmatocytes [35] and phagosome maturation of neuronal debris by epithelial cells [36] . In mammalian immunity , CD36 promotes the uptake of oxidized low density lipoproteins ( oxLDLs ) [37 , 38] and also regulates the host inflammatory response [39 , 40] . In addition , it is required to fight Mycobacteria and S . aureus infections in mice [41] and to induce pro-inflammatory cytokines in response to Plasmodium falciparum infection [42] . Using two lethal deficiencies that delete crq ( as well as other genes ) , we previously proposed that Crq was specific to apoptotic cell clearance , as crq-deficient embryonic plasmatocytes retained some ability to engulf both E . coli and S . aureus in vivo [35] . However , Crq was subsequently implicated in phagocytosis of S . aureus by S2 cells , a heterogeneous cell line with phagocytic abilities derived from late embryonic stages [41] . Thus , we generated a knock-out of crq and further investigated its role in microbial phagocytosis and its relationship with the humoral response at larval and adult stages in vivo . Drosophila plasmatocytes derive from pro-hemocytes originating either in the procephalic mesoderm of the embryo , with some further expanding by self-renewal in larval hematopoietic pockets , or from a second hematopoietic organ , the larval lymph glands , that persist to adulthood , or finally from adult hematopoietic hubs [43–51] . Here , we show that Crq is a major marker of plasmatocytes that is not required for hematopoiesis . The survival to adulthood of crq knock-out ( crqko ) mutants allowed us to quantitatively demonstrate that crq is required for pupae to survive environmental microbe infections and for adults to resist infection against Gram-negative and Gram-positive bacteria and fungi . crqko flies tolerate infections as well as control flies , but are unable to efficiently eliminate microbes . Indeed , crqko plasmatocytes are poorly phagocytic and defective in phagosome maturation . Crq acts parallel to the Imd and Toll pathways in eliminating pathogens , and crqko flies display elevated and persistent Dpt and upd3 expression , demonstrating that mutating crq promotes a state of chronic immune activation . As a consequence , crqko flies die prematurely with early signs of gut dysplasia and premature intestinal stem cell hyperproliferation . Therefore , we propose a model wherein crq is central to immune and organismal homeostasis . Overall , our results shed new light on the links between phagocytes , commensal microbes , gut homeostasis , and host lifespan .
In Drosophila adults , plasmatocytes ( the phagocytic hemocyte lineage ) originate from both embryonic and larval hematopoiesis [52] . crq is expressed in embryonic and larval plasmatocytes , as well as in S2 cells [53] . To test whether crq is expressed in adult plasmatocytes , we performed dual staining with combinations of GFP or dsRed and Crq antibodies of hemocytes bled from previously characterized transgenic plasmatocyte-reporter lines: eater-nls::GFP , eater-dsRed , and Hml-Gal4>UAS-GFP ( Hemolectin-positive hemocytes ) [54 , 55] ( Fig 1A and 1B ) . We found that 83 . 3±4 . 4% of hemocytes of Hml-Gal4>UAS-GFP and eater-dsRed carrying flies were positive for both markers , while 16 . 7±4 . 4% were positive for eater-dsRed alone ( Fig 1C ) . Crq immunostaining of hemocytes bled from eater-nls::GFP flies revealed that 85 . 2±2 . 6% of them were Crq and eater-dsRed positive , while 14 . 8±2 . 6% were Crq-positive but did not express eater-dsRed ( Fig 1C ) . From this , we extrapolated that about 72 . 4% of circulating hemocytes are positive for all three markers , 12 . 8% are double positive for Crq and Eater , and 14 . 8% solely express Crq ( Fig 1C ) . Therefore , crq is expressed in all Eater and Hml-positive hemocytes and marks the majority , if not all , adult plasmatocytes . To study its role in vivo , we generated a knock-out allele of crq ( crqko ) by homologous recombination [36] . This mutant deletes the entire crq open reading frame ( S1A Fig ) , and thus abolishes its expression [36] . As previously reported [35] , crq was not required for embryonic hematopoiesis . As for crq deletion mutants , crqko embryonic plasmatocytes were less efficient at clearing apoptotic cells , having a phagocytic index of 1 . 6±0 . 2 versus 2 . 45±0 . 3 apoptotic cells/plasmatocyte for wild-type embryos ( p<0 . 05 , S1B Fig ) . Homozygous crqko flies were viable and appeared morphologically normal . To ask whether crq is required for hematopoiesis at later developmental stages , we recombined an eater-nls::GFP transgene ( i . e . , the broadest plasmatocyte reporter after Crq ) ( Fig 1C ) into the crqko mutants , bled larvae and adults , and semi-automatically scored their eater-nls::GFP positive plasmatocytes by microscopy ( S1C Fig and Fig 1D ) . As previously reported for wild-type [56 , 57] , adult crqko flies had about 5-fold less plasmatocytes than larvae , and their number of eater-nls::GFP-positive plasmatocytes at both larval and adult stages were similar to that of wild-type flies ( Fig 1D ) . Pro-hemocytes that differentiate into plasmatocytes can also differentiate into crystal cells , which are involved in melanization [58] . Furthermore , self-renewing plasmatocytes of the embryonic lineage can also differentiate into crystal cells by trans-differentiation [59 , 60] . Thus , we tested whether crqko flies have differentiated crystal cells by scoring the melanotic dots formed following heat-induced crystal cell lysis . We found no significant difference between crqko and wild-type larvae ( S1D Fig ) . Therefore , Crq is a major plasmatocyte marker that is not required for hematopoiesis or hemocyte differentiation . While crqko homozygous flies were viable to adulthood , we could not maintain a homozygous stock on conventional fly food . We found that 36±3 . 2% homozygous crqko larvae arose from crosses between crqko heterozygous flies over GFP-marked CyO balancer chromosome , indicating full viability of the homozygous larvae ( Fig 1E ) . However , only 18±1 . 7% of emerging adults were homozygous crqko flies , indicating that half of the crqko homozygous progeny died during pupariation . Because flies with decreased plasmatocyte counts undergo pupal death associated with the presence of otherwise innocuous environmental microbes [23] , we asked whether supplementing the food with antibiotics could rescue crqko lethality . With this treatment , we recovered 29±3 . 6% of crqko homozygous adults ( Fig 1E ) , indicating a partial rescue of pupal lethality ( homozygous vs balanced adults , p = 0 . 021 ) . These results suggest that crqko pupae are susceptible to environmental microbes . No adult progeny could be recovered from crqko homozygous crosses on conventional fly food , but crqko adults emerged in the presence of antibiotics that gave rise to a second adult progeny ( Fig 1E and 1F ) . Maintaining a homozygous viable stock with antibiotics , however , remained difficult . We next bleached homozygous crqko embryos and raised them on sterile food . Under these axenic conditions , we successfully cultured a homozygous crqko line ( Fig 1F ) . Therefore , environmental microbes represent a health constraint for crqko homozygous flies . The susceptibility of crqko pupae to environmental microbes suggested that crq is required to mount an appropriate immune response . We next asked whether crq was up-regulated in flies injected with the Gram-negative bacterium Pectinobacterium ( previously known as Erwinia ) carotovora 15 ( Ecc15 ) or the Gram-positive Enterococcus faecalis ( E . faecalis ) . As anticipated , there was no crq expression in unchallenged ( UC ) or infected crqko flies as detected by RT-qPCR ( Fig 2A ) . While crq was expressed in both UC pXH87-crq transgenic ( the parental transgenic strain used for the generation of crqko flies , hereafter referred to as PXH87 ) and Canton S ( Cs ) control flies , it was not up-regulated within the first 24hrs of infection with Ecc15 or E . faecalis ( Fig 2A ) . However , we cannot exclude the possibility that crq may be up-regulated in plasmatocytes specifically at these early time points after infection . Its expression was also not altered in mutant flies for the NF-κB-like transcription factor Relish ( RelE20 ) downstream of the Imd pathway , or in flies mutant for the Toll ligand spz ( spzrm7 ) , upstream of the Toll pathway during that time-frame [9] . Surprisingly , we did observe an increase in crq mRNA levels at 36 ( p = 0 . 0076 ) and 132 hrs ( p = 0 . 0213 ) post Ecc15 infection ( S2A Fig ) , but did not detect any upregulation of crq mRNA levels at 36 and 132 hours post E . faecalis infection ( S2B Fig ) ( p>0 . 05 ) . Altogether , our data show that crq does not appear to be induced by infection in whole adult extracts during the first 24 hours post infection with Ecc15 and E . faecalis , and its expression appears independent of the Toll and Imd pathways . However , at later time-points after infection crq can be upregulated in a pathogen-specific manner , as seen with Ecc15 here . To assess the susceptibility of crqko male ( Fig 2B–2G ) and female ( S2B–S2G Fig ) flies to a variety of pathogens , we monitored their survival to these infections over time . When challenged by septic injury with the Gram-negative bacterium Ecc15 , male ( Fig 2B ) and female ( S2C Fig ) crqko flies were more susceptible than Cs and PXH87 control flies to this infection ( p<0 . 0001 ) . crqko flies all died within 336 hrs post-infection ( hpi ) , while only 64±6 . 8% and 67±6 . 5% of PXH87 and Cs flies had died by that time-point . crqko flies were , however , less susceptible than RelE20 mutants ( p<0 . 0001 ) , which are defective in the production of AMPs downstream of Imd [61] . All RelE20 flies died within 72 hpi , while only 56±7 . 7% of crqko flies had succumbed by that same time-point ( Fig 2B ) . To verify that the susceptibility to Ecc15 infection was due to the crqko mutation and not to a background mutation , we infected trans-heterozygous flies for crqko and Df ( 2L ) BSC16 , which deletes crq , with Ecc15 ( S2D Fig ) . These flies were as susceptible to Ecc15 infection as the crqko homozygous flies; they all died within 288 hpi , indicating that the crq mutation is responsible for this phenotype ( S2D Fig ) . crqko flies also succumbed to infection with E . coli ( 39±8 . 1% survival at 336 hpi ) , a Gram-negative bacterium that does not kill Cs ( 97±2 . 5% survival ) or PXH87 ( 86±5 . 6% survival ) flies . However , crqko flies were less susceptible to E . coli infection than RelE20 flies , which all died within 312 hpi ( p<0 . 0001 ) ( Fig 2C and S2E Fig ) . Therefore , crqko flies are susceptible to various Gram-negative bacterial infections . Similarly , crqko flies were more susceptible to infection with the Gram-positive bacterium E . faecalis than controls ( p = 0 . 0006 ) ( Fig 2D and S2F Fig ) and died in 312 hpi . However , they were less susceptible than spzrm7 flies ( p<0 . 0001 ) , which are defective in the production of AMPs downstream of Toll and died within 72 hpi ( Fig 2D and S2F Fig ) . crqko flies also died with intermediate susceptibility between that of control and spzrm7 flies ( p<0 . 0001 for both ) after septic injury with the pathogenic yeast Candida albicans ( Fig 2E and S2G Fig ) . Similarly , crqko flies were significantly more susceptible to exposure to spores of the entomopathogenic fungus Beauveria bassiana than Cs and PXH87 flies ( p<0 . 0001 ) , but less susceptible than spzrm7 flies ( p<0 . 0001 ) ( Fig 2F and S2H Fig ) . Finally , crqko flies were more susceptible to S . aureus infection than spzrm7 flies ( p = 0 . 0073 ) ( Fig 2G and S2I Fig ) , and spzrm7 flies were only slightly more susceptible than Cs and PXH87 flies ( p = 0 . 0006 and p<0 . 0001 respectively ) . Therefore , crqko flies are susceptible to Gram-positive bacteria and fungal infections and strongly susceptible to infection with S . aureus , a bacterium specifically cleared by phagocytosis [19 , 62 , 63] . These results argue that crq is required to fight infection . To further confirm this , we drove the expression of a UAS-crq transgene under the control of a crq promoter-Gal4 driver in the crqko flies ( crqko; crq-Gal4>UAS-crq ) . These rescue flies were no longer susceptible to Ecc15 ( S3A Fig ) , E . faecalis ( S3B Fig ) , and B . bassiana ( S3C Fig ) infections ( non-significant ( ns ) compared to PXH87 , and p<0 . 0001 when compared to crqko flies ) ( S3A–S3C Fig ) . To assess the possible requirement of crq in hemocytes , we drove the expression of a UAS-crq transgene under the control of a hemocyte-specific serpent promoter-Gal4 driver in the crqko flies ( crqko; srp-Gal4>UAS-crq ) . These flies were significantly less susceptible to Ecc15 , E . coli , E . faecalis and C . albicans infections than crqko flies ( p<0 . 0001 , p<0 . 0001 , p = 0 . 0004 and p<0 . 0001 , respectively ) ( S3D–S3G Fig ) . We did not observe any significant differences between rescue experiments with the crq-Gal4 or srp-Gal4 drivers after infection with Ecc15 , E . coli , or E . faecalis ( p>0 . 05 ) . The hemocyte-specific rescue of crqko flies infected with C . albicans , however , was slightly less efficient than the rescue with the crq-Gal4 driver ( p = 0 . 0269 ) . Thus , crq appears to be required mostly in phagocytes to fight infection by both Gram negative and Gram positive bacteria , although it appears to also be required in other tissues to fight C . albicans infection . Multi-cellular organisms use two complementary strategies to fight infection: resistance , to eliminate microbes , and tolerance , to allow them to endure the infection and/or its deleterious effects [64 , 65] . Compared to controls , crqko flies die prematurely at around 552 hours even in the absence of infection ( S4A Fig ) , suggesting these flies could be generally unfit or susceptible to damage . To test their response to abiotic damage , we pricked crqko flies with sterile needles at two separate thoracic sites . These flies did not die any earlier than non-pricked crqko flies ( S4A Fig ) . Thus , despite their decreased lifespan , crqko flies are not susceptible to aseptic wounds . To date , few studies have quantified the tolerance of immune-deficient flies [66 , 67] . Tolerance can be measured as the dose response curve relating health to microbe load . This curve takes the shape of a sigmoid; life expectancy in unchallenged conditions is considered as vigor , and the slope of the response curve ( the portion of the health/load curve which is linear ) estimates the ability to tolerate infection ( S4B Fig ) [67] . crqko flies have shortened lifespan and therefore an altered vigor ( S4A Fig ) . We further aimed to estimate whether crqko flies show a decrease in tolerance by measuring the relationship ( statistical interaction ) between microbial load and the corresponding health of the host [64 , 67] . We used three approximations to relate health to microbe load of crqko flies and focused on the linear part for each regression . First , we estimated the regression between the LT50 ( time at which 50% of the flies are dead ) of Ecc15 or E . faecalis-infected flies and the number of bacteria injected ( measured as colony forming units or CFUs ) ( S4C and S4D Fig ) . We did not detect any significant LT50~Time interaction between PXH87 and crqko flies ( p = 0 . 21782 for E . faecalis , p = 0 . 55800 for Ecc15 ) ( S4C and S4D Fig ) . However , this measure of bacterial load does not take into account the growth of the pathogen within the host . We therefore also quantified the regression between LT50 and the number of bacteria in the flies at 24 hpi ( Fig 3A and 3B ) . We detected significant LT50~Time interaction between PXH87 and crqko flies ( p = 0 . 008486 for E . faecalis , p = 0 . 018965 for Ecc15 ) , with PXH87 flies having lower tolerance than crqko flies ( Fig 3A and 3B ) . Finally , to get another estimate of the health of the flies , we plotted the health/bacterial load curve using survival at 3 time-points post Ecc15 infection and their corresponding bacterial load ( S4E Fig ) . We did not detect any significant survival-time interaction between PXH87 and crqko flies ( p = 0 . 335111 ) . Thus , while crqko flies die prematurely in the absence of infection , they do not show any decreased tolerance to infection when compared to control flies . These data suggest that the increased susceptibility of crqko flies to infection is due to their inability to control bacterial growth . In order to test this hypothesis , we monitored bacterial load during the course of Ecc15 and E . faecalis infections . In PXH87 flies , Ecc15 is eliminated within the first 48hrs of infection to reach an apparent plateau of low number of CFUs that persist at 72hrs post-infection ( Fig 3C ) . crqko flies were less able to clear Ecc15 than controls with higher bacterial loads throughout the infection ( p<0 . 001 for 24 , 48 and 72hrs ) ( Fig 3C ) . In contrast , despite an initial decline of CFUs at 48hrs , E . faecalis grew within control flies at 96 and 168hrs ( Fig 3D ) . During the whole course of infection with E . faecalis , the bacterial loads were significantly lower in wild-type control flies than in the crqko flies ( p<0 . 001 at 48 , 96 and 168hrs ) ( Fig 3D ) . These data indicate that crq is required for efficient elimination of both Ecc15 and E . faecalis . crq is required for efficient phagocytosis of apoptotic cells ( also known as efferocytosis ) in vivo , and phagocytosis of S . aureus by S2 cells ( S1B Fig and [35 , 41] ) . In addition , rescue of crq expression in hemocytes improved survival to various infections ( S3D–S3G Fig ) , suggesting that crq could alter microbial phagocytosis . To test this hypothesis , we first compared the susceptibility of crqko flies to infection with that of mutants for two phagocytic receptors , Eater and Drpr [26 , 31] . crqko flies succumbed to Ecc15 infection significantly faster than eater-deficient ( p = 0 . 0002 ) and drprrec8Δ5 loss-of-function flies ( p<0 . 0001 ) . 90±3 . 58% of crqko flies died within 192 hpi , while only 60±6 . 77% of drpr and eater mutants died in that same time ( Fig 4A ) . However , the crqko flies were significantly less susceptible to Ecc15 than RelE20 flies ( p<0 . 0001 ) , which all died within 48 hpi ( Fig 4A ) . In contrast , crqko flies succumbed to E . faecalis infection at a similar pace to that of both eater-deficient and drpr rec8Δ5 flies with 80–90% of all strains dying within 240 hpi ( Fig 4B ) . However , all mutants were significantly less susceptible than spzrm7 flies , which died within 48 hrs of E . faecalis infection ( p<0 . 0001 ) ( Fig 4B ) . To examine the precise role of crq in phagocytosis , we compared the amount of bacteria engulfed within 45min of thoracic injections of dead , Alexa 480-labeled E . coli and S . aureus in Cs , PXH87 , and crqko flies as previously described [20 , 26] ( Fig 4C and 4D ) . The crqko flies engulfed both E . coli and S . aureus bacteria with on average 66% less efficiency than control flies ( Fig 4C–4F , respectively ) . This phenotype was completely rescued in crqko flies expressing a UAS-crq transgene under a crq-Gal4 driver ( crqko , crq-Gal4>UAS-crq ) , which appeared to engulf more efficiently than control PXH87 flies ( S5A Fig and S5B Fig ) . We speculate that this difference was due to the overexpression of crq in those flies . To further assess the phagocytosis phenotype , wild-type and crqko flies carrying the eater-nls::GFP plasmatocyte-reporter were injected with dead rhodamine-labeled E . coli , bled , and their plasmatocytes were analyzed by confocal microscopy for internalized bacteria ( Fig 4G ) . The rhodamine-fluorescence per eater-nls::GFP plasmatocyte was quantified at 45min , 3hrs , and 5hrs post-injection and normalized to that of WT plasmatocytes at 3hrs post-injection ( Fig 4H ) . In WT plasmatocytes , the relative rhodamine-fluorescence increased as early as 45min , peaked at 3hrs , and decreased after 5hrs , as bacteria were presumably digested in mature phagosomes ( Fig 4H ) . In contrast , crqko plasmatocytes accumulated about 2-fold fewer bacteria than controls at 45min and 3hrs post-injection , but accumulated 1 . 7-fold more bacteria by 5hrs post-injection . In addition , at 45min post-injection , most bacteria were internalized within wild-type plasmatocytes ( S5C Fig ) , whereas bacteria were often bound to the cell surface of crqko plasmatocytes without being internalized ( S5D Fig ) . Thus , crqko plasmatocytes can engulf bacteria but are less efficient at it than controls at early time-points; they also appear to accumulate internalized bacteria over time . These results are consistent with a role for crq in promoting efficient uptake of bacteria . Moreover , the observed accumulation of bacteria in crqko plasmatocytes at 5hrs post-injection suggested that crq could also be required for phagosome maturation and digestion of bacteria . To test this , we injected control , crqko , and rescue flies with pH-sensitive pHrodo E . coli and S . aureus . pHrodo bacteria fluoresce when engulfed into a fully mature , acidified phagosome [68] ( S5E and S5F Fig ) . After quantification , we observed about 50% less fluorescence in crqko when compared to controls at 1 , 3 , and 5hrs post-injection ( S5G and S5H Fig , p<0 . 5 when comparing PXH87 and crqko flies ) . This phenotype was again completely rescued in crqko , crq-Gal4>UAS-crq flies ( S5E and S5F Fig and S5G and S5H Fig , p>0 . 5 when comparing PXH87 and rescue flies ) . At the single cell level , crqko plasmatocytes had up to 63±5 . 66% and 55±7 . 46% less pHrodo E . coli than controls at 3 and 5hrs , respectively ( Fig 4I ) . Finally , to ask whether mutating crq resulted in persistence of pathogenic bacteria , we injected live GFP-labeled Ecc15 in control and crqko flies carrying the eater-dsred plasmatocyte reporter ( Fig 4J and S5I Fig ) . Control PXH87 plasmatocytes had little to no GFP signal at 4 days post-infection , indicating that most bacteria had been engulfed and digested ( Fig 4J and S5I Fig ) . In contrast , crqko plasmatocytes had a 6-fold higher GFP signal , demonstrating that live Ecc15 accumulate in crqko plasmatocytes ( Fig 4J and S5I Fig ) . Taken together , these results show that crq is required for efficient microbial phagocytosis by playing a role in bacterial uptake and phagosome maturation . Phagocytosis has been proposed as a key step to initiate AMP production [21] . To assess the effect of mutating crq on AMP production downstream of both the Imd and Toll pathways , we next quantified the expression of Diptericin ( Dpt ) and Drosomycin ( Drs ) -encoding genes by RT-qPCR after Ecc15 or E . faecalis infections ( Fig 5A and 5B ) . As previously reported , septic injury of control flies with Ecc15 induced Dpt expression , which peaked at 10 hpi and returned to near-basal levels within 48 hpi ( Fig 5A ) [7] . In crqko flies , Dpt expression was 2-fold higher than in control flies at 10hrs post-infection and failed to return to basal levels within 48hrs ( Fig 5A ) . In contrast , there was no significant difference in Drs induction between control and crqko flies after E . faecalis inoculation ( Fig 5B ) . Survival curves indicated that crqko flies were less susceptible to a non-pathogenic E . coli infection than RelE20 flies , while double mutants for crqko and RelE20 were statistically more susceptible than RelE20 or crqko mutants alone ( Fig 5C ) . The extreme sensitivity of RelE20 flies to infection with pathogenic bacteria prevented us from carrying out these experiments with Ecc15 . Instead , we inoculated the flies with 20 times fewer E . coli than previously used in Fig 2 . Similarly , crqko and spzrm7 double mutants were also statistically more susceptible to C . albicans infection than the spzrm7 or crqko mutants alone ( Fig 5D ) . Therefore , crq is not required for the induction of AMPs and acts in parallel to the Toll and Imd pathways . These results suggested that aberrant phagocytosis in crqko flies can result in enhanced and persistent Imd pathway activation . Multiple negative regulators of the Imd pathway help maintain immune homeostasis . For example , Peptidoglycan Recognition Proteins ( PGRPs ) with amidase activity , such as PGRP-LB , degrade immunostimulatory molecules [15] . Thus , we next assessed Dpt expression levels by RT-qPCR in single and double PGRP-LBΔ and crqko mutants upon Ecc15 infection . Single crqko and PGRP-LBΔ mutants expressed statistically higher levels of Dpt than Cs and PXH87 controls at 48hrs ( Fig 5E ) . The Dpt expression resolved back to basal levels within 72hrs post infection in control flies , but remained high in single PGRP-LBΔ or crqko mutants despite a steady decline in its expression ( Fig 5E ) . Moreover , double mutants for crqko and PGRP-LBΔ expressed Dpt at levels 5-fold higher than controls at 24hrs post-infection , and levels remained high at 48 and 72hrs ( Fig 5E ) . These results demonstrate the critical interplay between phagocytosis and negative regulators of the immune system to achieve proper resolution of AMP expression upon systemic infection . Plasmatocytes are also a major source of cytokine production upon systemic infection . Upd3 , the Drosophila analogue of IL-6 , can induce the JAK-STAT pathway , which regulates the systemic immune response and metabolic homeostasis in the fat body , as well as gut homeostasis [6 , 22 , 69 , 70] . Using RT-qPCR , we asked whether crq is required for upd3 expression upon Ecc15 infection . Control flies displayed a small and temporary induction of upd3 expression that resolved within 72hrs ( Fig 5F ) . In contrast , UC and Ecc15-challenged crqko flies showed a 1 . 5-fold stronger induction of upd3 expression , which further increased over 72hrs ( Fig 5F ) . Thus crq is not required to induce upd3 expression , but crq mutation results in enhanced and continuously increasing upd3 expression . Altogether , these results demonstrate that crq is required for bacterial clearance and mutation of crq alters the resolution of AMPs and Upd3 cytokine production . PGRP-LBΔ and RelE20 mutants all die prematurely , within about 696 hrs ( 29 days ) of age , when compared to wild-type ( p<0 . 0001 ) and PXH87 ( p<0 . 0001 ) control flies , which die after about 912 hrs ( 37 days ) on conventional food at 29°C [15] ( Fig 6A ) . crqko flies died on average within 552 hrs ( 23 days ) , considerably earlier than RelE20 and PGRP-LBΔ mutants ( p<0 . 0001 ) . Double crqko and PGRP-LBΔ or crqko and RelE20 mutants died within about 480 hrs ( 20 days ) and 408 hrs ( 17 days ) of age , respectively ( Fig 6A ) . Antibiotic treatment partially rescued these phenotypes , as the lifespan of crqko flies and the double mutants increased significantly ( p<0 . 0001 ) ( Fig 6B ) . To ask whether the premature aging of crqko flies might correlate with a loss of immune cells or their function , we estimated the number of plasmatocytes present in control and crqko flies using the eater-nlsGFP reporter ( Fig 6C ) . As previously reported [56 , 57] , the number of plasmatocytes was decreased by about 40% in 16-day-old control flies ( Fig 6C ) , while similarly aged crqko flies had lost 80% of their plasmatocytes ( Fig 6C ) . Treatment with antibiotics rescued this crqko phenotype but had no effect on the plasmatocyte counts of control flies . crqko flies also lost about 40% of their plasmatocytes at 4 days post-E . faecalis infection when compared to similarly challenged wild-type controls ( Fig 6D ) . This loss of crqko hemocytes may be a consequence of accumulation of undigested bacteria inside their phagosomes . Thus , crq is required for plasmatocytes to survive innocuous or pathogenic bacterial infection . Dpt expression in wild-type and PXH87 flies is relatively low and stable over the first 8 days of their lives and increases as flies age [71] ( Fig 6E ) . Strikingly , Dpt expression was 70-fold higher in 8-day-old crqko flies and nearly 1 , 100-fold higher in the double mutants for crqko and PGRP-LBΔ compared to controls ( Fig 6E ) . Thus , the Imd pathway is strongly up-regulated early on in the life of these mutant flies , even in the absence of infection . This points to a role for Crq in phagocytosis and in maintaining immune homeostasis . Likewise , upd3 expression steadily increased as PXH87 flies aged , and it was further enhanced by nearly 10-fold in 8- and 16-day-old crqko flies ( Fig 6F ) . Antibiotic treatment partially rescued the levels of Dpt expression in crqko flies ( S6A Fig ) , arguing that the hyper-activation of the Imd pathway in these flies results from their inability to control environmental microbes . To address this , we plated fly extracts on both LB ( on which most pathogens can grow ) and MRS ( on which most Drosophila microbiota can grow ) agar plates and quantified the resulting CFUs ( S6B Fig ) . In line with previous studies , the CFUs obtained from 2 week-old control flies were in the range of 2 , 000 per fly ( S6B Fig ) [72–74] . Significantly fewer CFUs were recovered from PGRP-LBΔ mutants , while both crqko and RelE20 extracts showed a 10-fold increase . Double mutants for crqko and RelE20 had 50-fold more CFUs than controls ( S6B Fig ) . Altogether , these results demonstrate that Crq and the Imd pathway act in parallel and are required for the management of environmental microbes . Elevated levels of Upd3 are associated with midgut hyperplasia in aging flies [72 , 75] . In addition , loss of gut barrier integrity leads to early death in a microbiota-dependent manner [76 , 77] . Because 8-day-old crqko flies expressed high levels of upd3 , we asked whether they also displayed premature gut hyperplasia by looking at the number of mitotic PH3-positive intestinal stem cells of their midgut . While PXH87 and crqko flies did not show any signs of midgut hyperplasia at day 7 , midguts of 16 day-old crqko flies had a 2-fold increase in PH3-positive cells compared to that of similarly aged controls ( p = 0 . 0109 ) ( Fig 6G ) . This phenotype was completely rescued in crqko; crq-Gal4>UAS-crq flies ( Fig 6G ) . The double mutants for crqko and PGRP-LBΔ or for crqko and RelE20 showed even higher levels of intestinal stem cell proliferation than controls ( p = 0 . 03 ) and did so more prematurely ( already in 7-day-old flies ) ( Fig 6G ) . The premature increase in midgut stem cell proliferation was partially dependent on Upd3 , as upd3;crqko double mutant flies had significantly less mitotic cells ( p = 0 . 04 ) and lived longer than crqko flies ( p<0 . 0001 ) ( Fig 6H and 6I ) . However , the lifespan of upd3;crqko double mutants flies was still shorter than that of PXH87 flies ( p<0 . 0001 ) , suggesting that additional mechanisms play a role in the shortened lifespan of crqko flies . We further asked whether crq is required in hemocytes to maintain intestinal homeostasis . Hemocyte-specific re-expression of crq led to a strong rescue of lifespan compared to crqko flies ( p<0 . 0001 ) but not to the levels of PXH87 flies ( p = 0 . 0462 ) and to a partial rescue of midgut hyperplasia in 16-day-old flies ( p = 0 . 003 for crqko vs rescue and p = 0 . 0123 for rescue vs PXH87 ) ( Fig 6H and 6I ) . Altogether , these results indicate that flies lacking crq display chronically elevated expression of upd3 that triggers early midgut hyperplasia and promotes premature death .
Our study shows that Crq is required for the engulfment of microbes by plasmatocytes and their clearance , and that the mild immune deficiency due to crq mutation is associated with increased susceptibility to infection , defects in immune homeostasis , gut hyperplasia , and decreased lifespan ( S7 Fig ) . We have also re-confirmed a role for crq in apoptotic cell clearance , although the phagocytosis defect of crqko plasmatocytes is less severe than what had been previously observed with two lethal crq deficiency mutants , Df ( 2L ) al and Df ( 2L ) XW88 [35] . A possible explanation is that these deficiencies may have deleted at least one other gene required for apoptotic cell clearance . Additionally , morphological defects associated with secondary mutations could have exacerbated the crq phagocytosis defect by preventing efficient plasmatocyte migration to apoptotic cells . These same deficiency mutants had been assessed qualitatively for phagocytosis of bacteria by injecting embryos with E . coli or S . aureus; their plasmatocytes had no obvious defect in their ability to engulf these bacteria [35] . However , a role for crq in phagocytosis of S . aureus , but not that of E . coli , was subsequently proposed based on S2 cell phagocytosis assays following knock-down of crq by RNAi [41] . Here , we show that crq is required in vivo for uptake and phagosome maturation of both S . aureus and E . coli . A simple explanation of this discrepancy with E . coli could be that knocking down crq by RNAi is not sufficient to affect its role in E . coli phagocytosis ( but sufficient to affect its role in S . aureus phagocytosis ) , and that completely abrogating crq expression by in vivo knock-out leads to a stronger phenotype with both bacteria . Our in vivo data in crqko flies further demonstrate that crq is required to resist multiple microbial infections , such as Ecc15 , E . faecalis , B . bassiana , and C . albicans . These data therefore argue that crq plays a more general role in microbial phagocytosis than was previously anticipated . Our previous experiments to test whether crq is required for bacterial phagocytosis in embryos were qualitative rather than quantitative , and did not allow us to identify a role for crq at that stage [53] . In contrast , the experiments we now report in adult crqko flies are quantitative and allowed us to identify a delay in phagocytosis , followed by a defect in bacterial clearance in crqko hemocytes . A possible explanation for this discrepancy would be that hemocytes may differ in their expression profile , behavior , and phagocytic ability at various developmental stages due to differences in their microenvironment and/or sensitivity to stimuli . Accordingly , it has recently been shown that the phagocytic activity of embryonic hemocytes acts as a priming mechanism , increasing the ability of primed cells to phagocytose bacteria at later stages [78] . It is therefore possible that embryonic , larval and adult hemocytes display very different levels of priming and bacterial phagocytic activity , and that crq is required mostly in larval/adult bacterial phagocytosis . Alternatively , a potential defect in phagocytosis of bacteria by embryonic hemocytes of the crq deficiencies may have been suppressed by the deletion of ( an ) other gene ( s ) in that genomic region . Because the immune competence of hemocytes varies during development [50 , 79 , 80] , we were prompted to re-examine the potential role for crq in innate immunity by knocking it out . Here , we show that Crq is a major plasmatocyte marker at all developmental stages of the fly . We have found that crqko flies are homozygous viable , but short-lived , and can hardly be maintained as a homozygous stock in a non-sterile environment; crqko pupae become susceptible to environmental bacteria and their microbiota during pupariation . In a recent study , Arefin and colleagues induced the pro-apoptotic genes hid or Grim in plamatocytes and crystal cells using the hml-gal4 driver ( Hml-apo ) and observed a similar pupal lethality , but also associated with an induction of lamellocyte differentiation , and the apparition of melanotic tumors of hemocyte origin [81] . The authors therefore concluded that the death of hemocytes triggered lamellocyte accumulation and melanotic tumor phenotypes [81] . In contrast , we did not observe any obvious melanotic tumors in crqko flies , despite observing a loss of hemocytes in aging crqko flies ( Fig 6C ) and crqko flies subjected to Ecc15 infection ( Fig 6D ) . One possible explanation is that hemocytes do not die of apoptosis in crqko flies , but of a distinct mechanism . Alternatively , crq mutation could affect more hemocytes than Hml-apo flies , as crq is expressed in all plasmatocytes , while Hml is only expressed in 72 . 4% of all plasmatocytes expressing crq ( from Fig 1C ) . Thus the 27 . 6% of non-Hml plasmatocytes ( thus non induced for apoptosis , which is hml-Gal4 dependent [81] ) may respond to the death of the other plasmatocytes by inducing a signal that triggers the induction of lamellocytes and the subsequent formation of melanotic tumors . Considering the role of crq in apoptotic cell clearance , this signal may require a functional crq , which could explain why crqko flies do not develop melanotic tumors . Strikingly , in the Arefin study , as well as in previous studies , targeted ablation of plasmatocytes also made resulting ‘hemoless’ pupae more susceptible to environmental microbes [23 , 24 , 81] . Extensive tissue remodeling takes place at pupariation , and plasmatocytes are essential to remove dying cells , debris , and bacteria . Thus , it was argued that this increased susceptibility was likely due to environmental bacteria invading the body cavity after disruption of the gut [82] . In addition , it was found that the gut microbiome of Hml-apo flies could influence pupal lethality , as the eclosure rate of Hml-apo flies varied depending on the quality of the food they were reared on [81] . Accordingly , our rescue of the crqko pupal lethality with antibiotics demonstrates that their premature aging and death are indeed due to infection by normally innocuous environmental bacteria . Altogether , these data suggest that phagocytes and crq are important actors regulating the interaction between a host and its microbiome . Hosts use both resistance and tolerance mechanisms to withstand infection and survive a specific dose of microbes [65 , 83] . crqko flies exhibit a shorter lifespan when compared to control flies , but they are equally tolerant to aseptic wounds and infections . The crqko flies are less resistant to infection , as crq is required to promote efficient microbial phagocytosis . crqko plasmatocytes can still engulf bacteria , albeit at a lower efficiency than their controls . Our data also demonstrate that crq plays a major role in phagosome maturation during bacterial clearance . This is in agreement with a recent study showing that crq promotes phagosome maturation during the clearance of neuronal debris by epithelial cells [36] . Thus , crq is required at several stages of phagocytosis . Similar observations have been made for the C . elegans Ced-1 receptor and for Drpr , as both promote engulfment of apoptotic corpses and their degradation in mature phagosomes [84 , 85] . ‘Hemoless’ , Hml-apo and crqko flies are all more susceptible to environmental microbes and their microbiota . While it is not known whether mutants of eater , which encodes a phagocytic receptor for bacteria but does not play a role in phagosome maturation , are more susceptible to environmental microbes during pupariation , both eater mutants and ‘hemoless’ flies showed either decreased or unaffected systemic responses [23 , 24 , 26] . Hml-apo larvae however , showed an upregulation in Toll-dependent constitutive Drs mRNA levels whereas Dpt expression was suppressed [81] . In contrast , crqko flies showed no significant difference in constitutive or infection induced expression of Drs , but showed an increased expression of Dpt with age , and infection induced an increased and chronic expression of Dpt . Altogether our results argue that phagosome maturation defects in crqko flies lead to persistence of bacteria and thus to an increased and persistent systemic immune response via the Imd pathway . Such defects in phagosome maturation are not present in hemocyte ablation experiments , which could explain different outcomes for the host immunity and survival . We have found that Crq acts in parallel to the Toll and Imd pathways . In the mealworm Tenebrio molitor , hemocytes and cytotoxic enzymatic cascades eliminate most bacteria early during infection , and AMPs are required to eliminate persisting bacteria [86] . These data suggest that AMPs act in parallel with hemocytes to fight infections . We have also found that crqko flies are more susceptible to infection with S . aureus than wild-type and Toll pathway-deficient flies . These results are consistent with S . aureus infection being mainly resolved via phagocytosis and Crq having a major role in this process . Surprisingly , we have observed the opposite for infection with other Gram-negative or positive bacteria and fungi . Drosophila mutants for AMP production were more susceptible to infection than crqko flies , arguing that AMPs are critical to eliminate the bulk of pathogens . Indeed , crq ( thus phagocytosis ) is not essential for Ecc15 elimination , but accelerates bacterial clearance . Our results also suggest that the defects in phagosome maturation may allow some bacteria to persist and grow within hemocytes , where they are hidden from systemic AMPs . Thus , depending on the microbe , humoral and cellular immune responses can act at distinct stages of infection . In this context , phagocytosis acts as a main defense mechanism against pathogens that may escape AMPs or modulate their production . Chronic activation of immune pathways can be detrimental to organismal health [13–15] . In Drosophila , multiple negative regulators of the Imd pathway , including PGRP-LB , act in concert to maintain immune homeostasis [14–16] . We have observed that crqko flies sustain high production levels of the AMP Dpt and the cytokine Upd3 , demonstrating that defects in phagocyte function can lead to chronic immune activation . Notably , the level of Dpt expression induced by activation of the Imd pathway in unchallenged conditions is stronger in crqko flies than was previously observed in mutants of three negative regulators of the Imd pathway , namely pirkEY , PGRP-SCΔ , and PGRP-LBΔ [15] , and over 1 , 000-fold higher in PGRP-LBΔ , crqko double mutants . This is despite the persistence of only a few hundred bacteria in these mutants . This phenotype may be due solely to the accumulation of these persistent bacteria , or Crq may also function in plasmatocytes to remove immunostimulatory molecules from the hemolymph . Nonetheless , our study shows that plasmatocytes , Crq , and phagocytosis are all key factors in the immune response , and that losing crq induces a state of chronic immune induction . The ability of a host to control microbes decreases with age , a phenomenon called immune senescence [71] . The causes of immune senescence remain elusive , but the loss of immune cells with age and a decline in their ability to phagocytose have been suggested [56 , 57] . Recent studies have argued that microbial dysbiosis and disruption in gut homeostasis contribute to early aging [76 , 77 , 87] . In addition , persistent activation of the JAK-STAT pathway in the gut has been linked to age-related decline in gut structure and function [88] . Aging crqko flies lose a greater number of hemocytes than wild-type flies after infection , which may be the result of accumulating bacteria in these hemocytes in which phagosomes fail to mature . The premature death of crqko flies could be partially rescued by the presence of antibiotics . This demonstrates that phagocytosis , and phagosome maturation in particular , plays a crucial role in managing the response to environmental microbes and potentially , the gut microbiota directly to promote normal aging . We have also found that chronic upd3 expression in crqko flies triggers premature midgut hyperplasia , which is known to alter host physiology and promote premature aging [72 , 76 , 89] . It has recently been proposed that plasmatocytes can influence gut homeostasis by secreting dpp ligands and modulating stem cell activity [90] . Our results reinforce the possibility of an interaction between plasmatocyte function and gut homeostasis , and suggests that cytokines derived from hemocytes can trigger cell responses in the gut . These results are also in agreement with a recent publication showing that Upd3 from hemocytes can trigger intestinal stem cell proliferation [69] . Altogether , these results demonstrate that the interaction between hemocytes and the gut tissue are central to host health , and our data demonstrate that phagocytic defects can be associated with chronic gut inflammation and aberrant intestinal stem cell turn-over . As gut aging and barrier integrity are in turn important to maintain bodily immune homeostasis [76] , we propose the following model: in crqko flies , plasmatocyte-derived cytokines accelerate gut aging promoting loss of gut homeostasis and microbial dysbiosis , with immune and plasmatocyte activation acting in a positive feedback loop ( S7 Fig ) . Collectively , our data show that Crq is essential in development and aging to protect against environmental microbes . Interestingly , the impact of mutating crq on host physiology is strikingly different from previously reported phagocytic receptor mutations . We speculate that this could be due to its dual role in uptake and phagosome maturation during phagocytosis . Crq is required for microbial elimination in parallel to the Toll and Imd pathways and acts to maintain immune homeostasis . This situation is surprisingly reminiscent of inflammatory disorders , such as Crohn’s disease , that result from primary defects in bacterial elimination and trigger chronic immune activation and disruption of gut homeostasis . Further characterization of the crq mutation in Drosophila will provide an interesting conceptual framework to understand auto-inflammatory diseases and their repercussions on immune homeostasis and host health .
All stocks were raised at 22°C on standard medium , unless otherwise specified . RelE20 , spzrm7 , and PGRP-LBΔ stocks were described in [15 , 61 , 91] . The crqko stock was generated by homologous recombination , which removed the majority of the crq open reading frame [36] and ( S1B Fig ) . For bacterial infections , males or females were pricked in the thorax with a needle previously dipped in a concentrated pellet of the tested pathogen . The following bacterial or yeast strains were used at the indicated optical density ( OD ) taken at 600 nm: Ecc15 ( OD = 200 ) , E . coli ( OD = 200 and OD = 10 ) , E . faecalis ( OD = 5 ) , S . aureus ( OD = 0 . 5 ) , C . albicans ( OD = 200 ) . For B . bassiana infection , flies were shaken in a petri dish with mature germinating Beauveria for spore coating . All infections and aging experiments were performed at 29°C . In antibiotic treatments , a cocktail of kanamycin , ampicillin , rifampicin , streptomycin , and spectinomycin ( 5mg/mL each ) was added to the fly medium . Axenic stocks were generated as described in [72 , 73] . Survival experiments represent at least 3 independent repeats with 20 flies ( 60–100 flies tested ) . Survival was analyzed by a Log-rank test using the statistical programs R and Prism . Flies were individually homogenized in 500 μl of sterile PBS using bead beating with a tissue homogenizer ( OPS Diagnostics ) . Dilutions of the homogenate were plated onto LB agar or MRS agar with a WASP II autoplate spiral plater ( Microbiology International ) , incubated at 29°C , and the CFUs counted . Results were analyzed using a Krustal-Wallis test in R . Flies were injected in their thorax with 69nl of pHrodo red or Alexa 488 bacteria ( Life Technologies Inc . ) using a nanoject injector ( Drummond ) . The fluorescence within the abdomen of the flies was then imaged at 45min , 3hrs , and 5hrs post-injection with a Leica MZFLIII fluorescent microscope and DFC300 FX camera and quantified using Image J 2 . 0 . 0-rc-30/1 . 49s ( NIH ) . For ex vivo imaging , flies were injected with 46nl of PBS at 45min , 3hrs and 5hrs after infection to release all hemocytes , and 10 flies were bled on a lysine-coated slide by mechanically scraping their hemocytes onto a drop of PBS . Once settled for 10min on the slide , hemocytes were quickly dried and mounted with AF1 mounting solution ( Citifluor Ltd ) . Slides were automatically scanned using a Zeiss LSM 700 confocal microscope , and the number of plasmatocytes and average fluorescence signal per plasmatocyte quantified . For immunostaining , flies were bled as described above and the hemocytes fixed in a solution of PBS , Tween 0 . 1% , PFA 4% for 30min . The samples were incubated in PBT with 1% normal goat serum and Crq [53] and GFP antibodies ( Roche ) at 1:500 overnight at 4°C . Samples were washed at RT three times for 5min in PBS , incubated with the appropriate secondary antibodies at 1:1000 in PBT for 2hrs at RT , and washed three additional times in PBT . Samples were imaged with a Zeiss LSM 700 confocal microscope . Total RNA was extracted from pools of 20 flies per time point using TRIzol ( Invitrogen ) . RNA was reverse-transcribed using Superscript II ( Invitrogen ) , and the qPCR was performed using SYBR green ( Quanta ) in a Biorad instrument . Data represent the ratio or relative ratio ( in % ) of mRNA levels of the target gene ( crq , Dpt , Drs or upd3 ) and that of a reference gene ( RpL32 also known as rp49 ) . The primer sequences used in this study are provided in the supplementary material . All experiments were performed at least 3 times .
|
Phagocytosis is a first-line host defense mechanism against microbes . Interactions between phagocytosis and other immune mechanisms , such as the humoral response , however , remain elusive . Defective phagocytosis can lead to immune deficiencies and chronic auto-inflammation . Here , we show that Croquemort ( Crq ) , a Drosophila member of the CD36 family of scavenger receptors , plays a role in microbial phagocytosis . Crq is required in phagocytes for efficient uptake of bacteria and fungi , and mutants for crq succumb to both environmental microbes and infections . Crq is also required for phagosome maturation , and crq mutants lack the ability to fully clear bacterial infection . As a result , crq mutant flies enter a state of chronic immune activation . Notably , they show increased production of the cytokine Upd3 that induces intestinal stem cell proliferation . Consequently , crq mutant flies show early signs of gut dysplasia , as well as a shortened lifespan . Altogether , our study demonstrates a new link between phagocytosis and tissue homeostasis , and illustrates how the chronic induction of cytokine production secondary to defective phagocytosis can alter gut homeostasis and shorten lifespan .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"vesicles",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"cell",
"processes",
"immunology",
"microbiology",
"staphylococcus",
"aureus",
"bacterial",
"diseases",
"physiological",
"processes",
"homeostasis",
"phagosomes",
"enterococcus",
"cellular",
"structures",
"and",
"organelles",
"bacteria",
"bacterial",
"pathogens",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"escherichia",
"coli",
"infections",
"medical",
"microbiology",
"staphylococcus",
"microbial",
"pathogens",
"enterococcus",
"faecalis",
"phagocytosis",
"hemocytes",
"cell",
"biology",
"physiology",
"apoptosis",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2016
|
The Drosophila CD36 Homologue croquemort Is Required to Maintain Immune and Gut Homeostasis during Development and Aging
|
Leprosy Type-1 Reactions ( T1Rs ) are pathological inflammatory responses that afflict a sub-group of leprosy patients and result in peripheral nerve damage . Here , we employed a family-based GWAS in 221 families with 229 T1R-affect offspring with stepwise replication to identify risk factors for T1R . We discovered , replicated and validated T1R-specific associations with SNPs located in chromosome region 10p21 . 2 . Combined analysis across the three independent samples resulted in strong evidence of association of rs1875147 with T1R ( p = 4 . 5x10-8; OR = 1 . 54 , 95% CI = 1 . 32–1 . 80 ) . The T1R-risk locus was restricted to a lncRNA-encoding genomic interval with rs1875147 being an eQTL for the lncRNA . Since a genetic overlap between leprosy and inflammatory bowel disease ( IBD ) has been detected , we evaluated if the shared genetic control could be traced to the T1R endophenotype . Employing the results of a recent IBD GWAS meta-analysis we found that 10 . 6% of IBD SNPs available in our dataset shared a common risk-allele with T1R ( p = 2 . 4x10-4 ) . This finding points to a substantial overlap in the genetic control of clinically diverse inflammatory disorders .
A clear temporal separation from the different stages of leprosy pathogenesis identifies the endophenotype Type-1 Reactions ( T1Rs ) as a well-delineated example for host pathological inflammatory responses in humans . An endophenotype , as defined by John and Lewis in 1966 , is a microscopic and internal phenotype that is not easily identified in the presence of an exophenotype , which is the dominating phenotype that is more easily recognized [1] . In the context of our study we refer to the term endophenotype as a condition ( T1R ) that occurs in some but not all persons displaying the necessary exophenotype ( leprosy ) diverging from the original concept of John and Lewis . Of note , T1R shares immune-pathological similarities with immune reconstitution inflammatory syndrome of HIV patients undergoing highly active antiretroviral therapy [2] , and paradoxical reactions in patients with Buruli ulcer undergoing anti-microbial therapy [3 , 4] . T1Rs are a major challenge of current leprosy control since the hyper-inflammatory immune response triggered by Mycobacterium leprae , the etiological agent of leprosy , frequently leads to permanent nerve damage [5] . A prompt identification of T1R cases and rapid clinical intervention are essential to prevent lasting neurological damage [6] . While acute neuritis is a hallmark of T1R , the detailed mechanisms that link hyper-inflammation to neuropathy are not known . Depending on the epidemiological setting , 30% to 50% of leprosy cases develop at least one T1R episode [5 , 7–10] . Why only a fraction of leprosy-infected individuals undergo T1R is not known but the description of a transcriptome signature in response to M . leprae antigen strongly supported a genetic predisposition to T1R [11] . In addition , genetic variants in a few number of candidate genes ( TLR1 , TLR2 , NOD2 , LRRK2 and TNFSF15/TNFSF8 ) were found to be associated with T1R [12–17] . Independently , variants in several of these genes had also been implicated in susceptibility to leprosy per se raising the possibility of an overlapping genetic control of intensity of pathway activation between protective and pathological host responses [18] . To contrast the genetic control of leprosy and its clinical subtypes from the genetic control of the pathological immune responses typical for T1R , we designed a genome-wide association scan ( GWAS ) to identify novel genes or variants associated solely with T1R . This may lead to predictive biomarkers for early recognition of T1R and possibly indicate novel pharmacological interventions that reduce the need for potentially adverse long-term corticoid treatment in T1R .
We evaluated the association of host genetic factors with T1R by conducting a family-based GWAS in 221 families with 229 T1R-affect offspring followed by stepwise replication in independent population-based case-control samples ( Fig 1 ) . For the discovery phase , approximately 6 . 3 million genotyped and imputed variants ( SNPs and INDELs ) that passed quality control were tested for association in both T1R-affected and T1R-free family sets . In T1R-affected families , a suggestive association with T1R was detected on chromosome region 10p21 . 2 ( Fig 2A and 2B ) . Among the 103 SNPs located in the interval and strongly associated with T1R leprosy ( pDiscovery < 0 . 001 ) , SNP rs7916086 ( pDiscovery = 8 . 2x10-7 ) displayed the strongest evidence of association . Applying a linkage disequilibrium ( LD ) threshold of r2 > 0 . 9 , the 103 SNPs located between the two recombination hot-spots in the 10p21 . 2 locus could be grouped into seven SNP bins ( Fig 2C , S1 Table ) . None of the SNPs in the 10p21 . 2 locus located outside this hot spot showed evidence for association below p < 0 . 001 . The tag SNP that presented the lowest p value for the association with T1R in each of the seven SNP bins was selected as the leading variant for its particular bin . When the 220kb region comprising the T1R-risk locus was evaluated in the T1R-free families no signal of association was detected ( Fig 2C , S1 Table ) . The formal heterogeneity test confirmed preferential association of T1R with the seven SNP bins reported in the discovery phase with p Heterogeneity ranging from 0 . 009 to 5 . 0x10-04 ( S1 Table ) . Of note , an additional 4372 variants located throughout the genome displayed p < 0 . 001 in the T1R-affected subset and are given in S2 Table . A multivariable analysis including the leading variant of each SNP bin ( r2 = 0 . 9 ) associated with T1R selected rs7916086 as the single signal of association in the 10p21 . 2 chromosomal region ( S1 Table ) . However , due to high LD among SNPs of the investigated bins , alternative models could not be excluded ( S1 Fig ) . Therefore , we selected the seven leading variants for each of the SNP bins ( r2>0 . 9 ) described above for further confirmative analyses in independent populations . The leading SNP in the discovery phase , rs7916086 , showed borderline evidence for association with T1R in the Vietnamese replication sample ( p = 0 . 04 ) . However , association of rs7916086 with T1R was not validated in the Brazilian sample ( p = 0 . 26 ) ( S3 Table ) . The leading SNPs in four additional SNP bins , namely rs10509110 , rs11006600 , rs10826329 and rs10763614 , did not show consistent evidence for significant association across the Vietnamese and Brazilian populations ( S3 Table ) . In contrast , SNP rs1875147 displayed strong replicated and validated evidence of association with T1R ( Table 1 ) . SNP allele “C” of rs1875147 was identified as global risk factor for T1R with an odds ratio ( OR ) = 1 . 37; confidence interval of the one-sided test ( uniCI ) 95% = 1 . 11; p = 0 . 006 , in the Vietnamese replication sample and , OR = 1 . 47; uniCI 95% = 1 . 15; p = 0 . 005 , in the Brazilian validation sample ( Table 1 ) . In addition , the tag SNP for a second bin , rs10826321 , was associated with T1R in the Vietnamese replication ( p = 0 . 003 ) and the Brazilian validation sample ( p = 0 . 04 Table 1 ) . In Vietnam , SNPs rs1875147 and rs10826321 were highly correlated ( r2 ≈ 0 . 8 ) capturing the same signal of association with T1R . However , compared to the Vietnamese , the LD between rs1875147 and rs10826321 was lower in Brazilians ( r2 = 0 . 21; S1 Fig ) . Since the 7 SNP were tested for replication and validation we did not apply a Bonferroni correction . To investigate the independent effect of rs1875147 and rs10826321 in Brazilians we performed a multivariable analysis . SNP rs1875147 maintained the association with T1R ( p = 0 . 009 ) while rs10826321 lost significance ( p = 0 . 49 ) . Next , we investigated the combined effect of rs10826321 and rs1875147 by conducting a haplotype analysis in the Brazilian sample . We found that the haplotype with the T1R-risk allele in both SNPs ( G-C alleles for rs10826321 and rs1875147 respectively ) was significantly associated with T1R ( p- = 0 . 04; S4 Table ) consistent with results obtained by multivariable analysis supporting the non-independent association of rs1875147 and rs10826321 with T1R . Interestingly , the haplotype ( A—C ) containing the alternative allele for rs10826321 and the T1R-risk allele for rs1875147 showed a trend towards association with T1R in Brazilians ( p = 0 . 06; S4 Table ) . This observation supported rs1875147 as the main cause of association of T1R with the 10p21 . 2 region . When a combined analysis was performed to summarise all study phases , only SNPs rs1875147 surpassed the genome wide threshold for significant association with T1R ( Table 1 , Fig 2D ) . In a fixed-effect meta-analysis SNPs rs1875147 presented an OR = 1 . 54; CI 95% = 1 . 32–1 . 80 , p = 4 . 5x10-08 for the C-allele . As modest levels of population heterogeneity were observed for the T1R-risk SNPs in a complementary fixed-effect model ( Table 1; S3 Table ) , we performed a random-effect meta-analysis . The seven SNPs showed similar levels of significance between the fixed and random-effect ( S5 Table ) . For the rs1875147 the random-effect model resulted in a risk-effect of OR = 1 . 54; CI 95% = 1 . 28–1 . 86 , p = 6 . 4x10-08 for the C-allele . The locus validated for association with T1R mapped within two recombinational hot spots where a single long non-coding RNA ( lncRNA ) was located ( Fig 2D ) . The novel lncRNA presented two isoforms , one encoded by the ENSG00000235140 ( a . k . a . RP11-135D11 . 2 ) gene and another encoded by the uncharacterized LOC105378318 ( Fig 2D ) . The two T1R-risk variants , rs1875147 and rs10826321 , are located at 6 . 5 kb and 8 . 7 kb , respectively , upstream of the transcription start site of the ENSG00000235140 gene . The rs10826321 variant alters the binding motif of a CTCF transcription factor in a CTCF binding site in 83 cell types ( S2A Fig ) . The rs10826321 T1R-risk G-allele is more commonly observed in CTCF binding than the alternative A-allele ( S2A Fig ) . SNP rs1875147 is reported as an expression quantitative trait locus ( eQTL ) for ENSG00000235140 in the transverse colon where the T1R-risk allele C is correlated with higher gene expression ( S2B Fig ) [19] . The eQTL effect for rs1875147 was also nominally significant in the terminal ileum of the small intestine and in the spleen in a smaller sample size ( S2C Fig ) . Both rs1875147 and rs10826321 are conserved loci across species [20] . Certain SNP alleles associated with T1R-risk had previously been shown to be susceptibility factors for inflammatory bowel disease ( IBD ) [21–23] . To investigate if there was an enrichment of risk alleles between T1R and IBD , we systematically compared evidence of association with T1R in the Vietnamese discovery set with evidence for association in a recent GWAS meta-analysis for IBD [24] ( Fig 3A ) . Of 232 independent top SNPs that had been associated with IBD by meta-analysis , 208 were available in the T1R-affected and T1R-free GWAS datasets [24] . For 22/208 SNPs ( 10 . 6% ) the IBD risk allele was associated at the 0 . 05 level with risk of T1R/leprosy . ( Fig 3A , S6 Table ) . This observed proportion of shared risk-alleles between T1R leprosy and IBD is significantly non-random ( p = 2 . 4x10-4 ) . Importantly , none of the 22 SNPs showed significant evidence of association with T1R-free leprosy while 9 SNPs displayed significant heterogeneity between leprosy and T1R indicating an enrichment of stringently defined T1R SNPs among IBD SNPs ( p = 1 . 9x10-3; Fig 3 , S6 Table ) . Similar analyses in T1R-free families , failed to detect an enrichment of leprosy risk alleles among IBD SNPs . Indeed , while several genes with known overlap of IBD and leprosy were detected ( i . e . RIPK2 , LACC1 and IL23R ) , there was no genome-wide statistical enrichment for IBD risk alleles in T1R-free leprosy ( p = 0 . 09; Fig 3A ) . As additional control , we evaluated three non-immunity phenotypes for which recent GWAS meta-analyses were available ( Schizophrenia [25] , human height [26] and human blood metabolites [27] ) for an overlap of genetic risk factors with T1R . There was no significant enrichment of either leprosy or T1R risk alleles with SNP alleles of any of the three control phenotypes ( Fig 3B to 3D ) . Among the 22 IBD SNPs associated with T1R leprosy , 17 are cis eQTL for one or more genes ( S3 Fig ) . Similarly , 7 of the 9 SNPs significantly heterogeneous between T1R and leprosy were eQTLs in either whole blood , rs3774937 ( NFKB1 ) , rs10065637 ( ANKRD55 ) , rs11150589 ( ITGAL ) and rs2836878 ( lncRNA ENSG00000235888 ) or multiple tissues , rs4664304 ( LY75 ) , rs113653754 ( HLA-DQB1 ) and rs4768236 ( LRRK2; S4 Fig ) [19 , 28] . SNPs that were eQTL in multiple tissues displayed some of the strongest associations with T1R ( S6 Table ) . Since the LY75 gene encodes a major endocytic receptor of dendritic cells and HLA-DQB1 gene expression is also modulated by a risk SNP , our results highlight the critical role of antigen presentation in dysregulated immunity of both IBD and T1R .
In summary , we have conducted the first GWAS for pathological inflammatory responses in leprosy using the largest collection of T1R-affected individuals to date . Our stepwise replication study in ethnically independent populations led to the description of an eQTL ( rs1875147 ) for the lncRNA gene ( ENSG00000235140 ) as a global risk-factor for T1R . Moreover , we have observed an enrichment of shared risk-alleles between leprosy/T1R and IBD but not for IBD and leprosy per se . We have shown previously for the PARK2 gene that testing only the leading SNP of the discovery phase in ethnically independent populations without considering population differences in the LD structure may result in false negative associations [29] . Here , the leading SNP in the Vietnamese discovery phase , rs7916086 , could not be validated for the association with T1R; but rather , two SNPs highly correlated with rs7916086 in the Vietnamese population ( namely rs1875147 and rs10816321 ) were T1R-risk factors in Brazilians . The lower LD conservation in Brazilians enabled us to narrow down the T1R association signal in the 10p21 . 2 region to a single SNP , rs1875147 , which presented a pre-established regulatory function . Since we used the 1000 Genomes data to impute SNPs for the analysis and chose a high r2 cut off for SNP bin definition , it is unlikely that another common SNP in strong LD with rs1875147 would provide a stronger signal of association . However , we cannot rule out a combination of rare variants as cause of the association signal . Combined , our results highlight the strength of employing different ethnicities in the validation phase since the genetic effects of rs7916086 , rs10826321 and rs1875147 could not be disentangled in the Vietnamese sample . An association with leprosy was previously reported for chromosome region 10p21 . 2 [30] . The reported peak of association with leprosy per se encompassed the ADO and EGR2 genes . The leading variant in the ADO/EGR2 locus , rs58600253 , is located at approximately three mega bases upstream of the T1R associated locus . When the imputed variant rs58600253 ( Info = 0 . 992 ) was evaluated in the T1R-affected and T1R-free families we observed no significant signal of association ( p = 0 . 25 and p = 0 . 22 , respectively ) . Moreover , no correlation of rs58600253 with the T1R signal tagged by rs187514 was detected using the best call genotypes ( r2 = 0 . 04 ) . These results indicated that the T1R locus on region 10p21 . 2 is independent of the leprosy per se ADO/EGR locus . Moreover , a recent GWAS meta-analysis by Wang et al . identified four novel loci associated with leprosy [31] . While none of the leading SNPs reported by Wang et . al . were significant in our T1R GWAS , we observed independent variants associated with leprosy in two out of the four newly reported loci . The rs4684104 SNP near the PPARG gene ( p = 2 . 4 x 10−6; p Heterogeneity = 5 . 4 x 10−4 ) and the rs10239102 near the BBS9 gene ( p = 4 . 2 x 10−4 , p Heterogeneity = 0 . 07 ) were T1R-specific and T1R-non-specific , respectively . The functional annotation for the rs1875147 T1R-risk alleles argues that upregulation of ENSG00000235140 transcription may contribute to T1R susceptibility . However , this lncRNA gene has not been found to be commonly expressed in all tissues . The ENSG00000235140 gene was detected mostly in the sexual organs , gastro intestinal tract , and in the lungs of healthy individuals [19 , 32] . These tissues usually do not harbor M . leprae , but are a reservoir for other mycobacteria such as M . avium paratuberculosis ( colon ) and M . tuberculosis ( lungs ) . The limited knowledge about the role of ENSG00000235140 in health and disease limits our understanding of this lncRNA in T1R pathogenesis . Notwithstanding , our data present the ENSG00000235140 gene as a prime candidate to unravel the riddle of pathological immune responses in T1R and possibly inflammatory disorders in general . An overlap regarding the genetic control of leprosy per se and IBD has been previously suggested [21–23 , 33] . Although the SNPs associated with IBD and leprosy are frequently the same the risk-allele are less consistent . This factor hinders the establishment of a shared biological mechanism for IBD and leprosy . As T1R affects a considerable proportion of leprosy cases it is possible that , at least partially , the genetic overlap proposed between IBD and leprosy is due to the T1R phenotype . Here our strategy was to evaluate if T1R and IBD shared additional risk-alleles . Although , our approach focusing only on the leading SNP per IBD locus was conservative , the enrichment for shared risk-alleles in IBD and T1R was strong and may represent only part of the shared biological mechanisms . The results reported here strongly support the view that susceptibility to IBD involves a genetic predisposition to mount dysregulated inflammatory immune responses as exemplified by the T1R phenotype in leprosy . In complex traits , precise phenotype definition is key for the detection of genetic associations . For example , we have previously shown for variants of the TNFSF15/TNFSF8 genes that leprosy patients with the T1R endophenotype are largely the cause of association with the leprosy exophenotype [16 , 17] . Consequently , the replication of the TNFSF15/TNFSF8 association in samples of leprosy patients with a low proportion of T1R is expected to display low power . Equally important , accurate phenotype definition directs the interpretation of detected associations . Assigning genes to the exophenotype leprosy that impact on the endophenotype T1R may lead to wrong conclusions about the pathology of leprosy . Hence , a notable strength of our study is the focus on a well-defined endophenotype which is directly connected to a major problem of current leprosy control . This increases the power for detection of genetic effects while at the same time opening a translational link for control of nerve damage . Despite these strengths , our study also had limitations . For example , we only tested an additive model , since T1R is highly prevalent in leprosy ( 30 to 50% of all cases ) ; dominant and recessive models of inheritance could unveil additional novel associations . Moderate levels of population heterogeneity were observed in the combined analysis ( I2 values ≈ 30 to 50; Table 1 , S3 table ) . The population heterogeneity was likely driven by a winner's curse phenomenon , a bias that inflates risk estimates for newly identified SNPs when a study lacks statistical power [34] . Because of the possible effect of winner’s curse , the combined risk effect should be consider as a summary of our study and the real risk-effect for variants in the 10p21 . 2 region are likely closer to the effect of the replication and validation phase . A second limitation is the pleiotropic analysis of IBD and leprosy/T1R . As a consequence of the T1R/leprosy sample size , intermediary to low frequency variants with modest genetic effect would not have been detected by our study . This might have led to an increased type II error and an under-estimation of the true overlap in the genetic control of IBD and T1R/leprosy . Hence , studies employing larger numbers of T1R/leprosy patients might provide better estimates of the overlap in the genetic control of these two inflammatory conditions .
The study was conducted according to the principles expressed in the declaration of Helsinki . Written informed consent was obtained for all adult subjects participating in the study . All minors assented to the study , and a parent or guardian provided the informed consent on their behalf . The study was approved by the regulatory authorities and ethics committees of the participating centers . Namely , Comissão Nacional de Ética em Pesquisa ( CONEP; 12638 ) for Goiania; The Research Ethics Committee at Fiocruz ( CEP-Fiocruz Protocol 151/01 ) for Rio de Janeiro; The Research Ethics Committee at Institute Lauro de Souza Lima for Rondonópolis ( 172/09 ) ; the Research Ethics Board at the RI-MUHC in Montreal ( REC98-041 ) , and the regulatory authorities of Ho Chi Minh City ( So3813/UB-VX and 4933/UBND-VX ) for the Vietnamese population . The subjects included in the study where followed up for a minimum of three years to confirm the presence or absence of T1R episodes . T1R-affected and T1R-free leprosy cases were mainly selected from the borderline class of Ridley and Jopling clinical scale of leprosy as T1R affects predominantly these cases that present an immunologically unstable immune response against M . leprae infection [7 , 35] . For the discovery phase , two sets of families of Vietnamese ( Kinh ) origin with leprosy-affected offspring were selected: the T1R-affected set comprised of 229 offspring belonging to 221 families and a T1R-free set comprised of 229 offspring in 209 families . The T1R-free set was matched to the T1R-affected set by the offspring’s leprosy clinical subtype . In the discovery phase , a transmission disequilibrium test ( TDT ) was applied to the T1R-affected and the T1R-free families independently . Next , the results of the individual TDTs were compared to investigate heterogeneity between both samples . The genetic heterogeneity test between T1R-affected and T1R-free subsets was tested by means of the FBATHet statistic and is detailed in the statistical approach section [36] . Variants that were associated in the T1R-affected set and showed heterogeneity with the T1R-free set were considered as T1R-specific and were investigated in the next phases of the study . The initial association results were followed up employing a replication and a validation phase . The replication sample was of Vietnamese ethnicity and encompassed 253 T1R-affected and 563 T1R-free leprosy patients . The validation sample comprised 471 T1R-affected subjects and 446 T1R-free leprosy patients as controls from the Central-west and South-east regions of Brazil as described previously [16 , 37 , 38] . In both replication and validation samples , cases and controls were matched for leprosy subtype . Genotypes of all subjects of the discovery phase were determined using the Illumina Human 660w Quad v1 bead chip . SNPs with call rate < 0 . 98 , more than two Mendelian errors in T1R-affected or T1R-free sets , minor allele frequency ( MAF ) < 0 . 01 or presented Hardy-Weinberg equilibrium ( p < 1 . 0 x 10−3 ) in 763 leprosy unaffected parents were removed from the analyses . Genotypes for the replication and validation phase samples were obtained through high-throughput SEQUENOM platform . The same quality control thresholds from the discovery phase were applied for SNP call rates and MAF exclusion to the replication and validation phase , with the exception of the HWE p value cut off which was restricted to p < 0 . 05 due to the lower number of tested SNPs compared to the discovery phase . A total of 38 , 753 genotyped A/T and C/G SNPs were removed prior to the phasing and imputation . The remaining 495 , 973 SNPs that passed the quality control filtering in the discovery phase were used to impute additional 11 . 5 million variants ( SNPs and INDELs ) in both T1R-affected and T1R-free family sets with SHAPEIT2 [39] and IMPUTE2 [40] software and the 1000 genomes Phase I v3 dataset containing 1092 individuals as the reference panel . Given the exploratory nature of the discovery phase , the threshold of imputation information measure ( Info ) > 0 . 5 was applied to capture most of the common variants ( MAF > 5% ) with reasonable confidence ( S5 Fig ) [41] , MAF > 0 . 001 and more than 10 informative families in both T1R-affected and T1R-free sets were used as a post-imputation quality control filtering for the association analyses . Imputed variants that were evaluated in the replication and validation phase had their genotypes confirmed in 440 subject of the discovery sample using the high-throughput SEQUENOM platform . In the discovery phase , a TDT was used to estimate non-random transmission of alleles from heterozygote parents to leprosy-affected offspring in both T1R-affected and T1R-free sets ( p Discovery ) . The analysis was carried out under a log-additive model using FBATdosage v2 . 6 for genotyped and imputed variants [42] . To contrast the TDT tests from the discovery phase a FBATHet test in T1R-affected and T1R-free sets was used ( p Heterogeneity ) . Briefly , heterogeneity of the allelic transmission rates in an endophenotype can be done in the FBATdosage framework by pooling the two subsets ( T1R-affected and T1R-free ) and contrasting the presence of the endophenotype T1R ( T1 = 1/V1 ) with the absence of T1R ( T2 = −1/V2 ) , where V1 and V2 denote the variance of the FBATdosage statistic for the each sample set , respectively [36] . Population-based association analyses were performed using logistic regression under a log-additive model and adjusting by the co-variables gender and age at leprosy diagnosis using PLINK v1 . 0 . 7 . The one-sided test was used with the alternative hypothesis that the T1R-risk alleles were also risk factors in the replication and validation samples . Multivariable analysis were performed with stepwise conditional logistic regression in SAS 9 . 3 . The haplotype analysis in the Brazilian sample was performed with THESIS v3 . 1 [43] . The linkage disequilibrium structure was evaluated with Haploview 4 . 1 [44] . To summarize the different steps of the study we used an inverse variance–weighted meta-analysis with a fixed-effect model and an alternative random-effect model proposed by Han and Eskin as implemented in the software METAL [45] and METASOFT [46] , respectively . To estimate the risk effect for the family-based design the un-transmitted allele from parents to T1R-affected offspring in the TDT was used as a pseudo-sib control . Briefly , up to three unaffected pseudo-sibs were created per family , one for each possible un-transmitted genotype . Subsequently , the original T1R-affected offspring were compared to the T1R-free pseudo-sibs in a matched case-control [47] . Under a log-additive model , TDT and pseudo-sibs analyses are equivalent [47] . Of note , METAL and METASOFT use standard errors and β coefficients to combine the statistics of each studied phase . In contrast to the replication and validation steps , a two-sided test was used in the combined analysis for the Vietnamese and Brazilian samples . To investigate if there was an enrichment of shared risk alleles between T1R and IBD , we used a hypergeometric test to systematically compare evidence of association with T1R in the Vietnamese . For instance , out of the 6 , 333 , 954 variants tested for association in our study 319 , 671 had p < 0 . 05 in the T1R-affected subset . Using the observed prior information of the number of variants with p < 0 . 05 , the hypergeometric test calculates the statistical significance of randomly selecting 22 variants with p < 0 . 05 when 208 variants ( number of variants from the IBD GWAS meta-analysis present in the T1R dataset ) were randomly drawn from a total of ~6 . 3 million . Here , the hypergeometric test corresponds to the one-tailed Fisher’s exact test . The same analytical approach was applied for the T1R specific variant in IBD , but in this analysis we used the number of variants with p < 0 . 05 and p heterogeneity < 0 . 05 out of a total of ~6 . 3 million variants of the GWAS . Since we tested for sharing of the same risk allele between T1R , and IBD , CD or Ulcerative Colitis ( UC ) one-tailed p values are reported . The same strategy was used in the three control phenotypes ( schizophrenia , height and blood metabolites . Since we tested for sharing of the same risk allele between T1R , and IBD , CD or one-tailed p values are reported . IBD meta-analysis data was freely available at the IBDgenetics website ( https://www . ibdgenetics . org/ ) [24 , 48] . Briefly , seven CD and eight UC collections with genome-wide data were combined with additional replication samples resulting in a total of 42 , 950 IBD cases and 53 , 536 health controls for the IBD meta-analysis [24 , 48] . Variants that surpassed p < 5 . 0 x 10−8 for association with IBD were reported as significant . Functional data for annotated SNPs were extracted from the GTeX ( http://www . gtexportal . org/home/ ) and Haploreg v4 http://www . broadinstitute . org/mammals/haploreg/haploreg . php databases . [19 , 20] The FBAT dosage is available at https://www . hgid . org/index . php ? menu=download
|
Leprosy still affects approximately 200 , 000 new victims each year . A major challenge of leprosy control is the prevention of permanent disability due to nerve damage . Nerve damage occurs if leprosy remains undiagnosed for extended periods or when patients undergo pathological inflammatory responses termed Type-1 Reactions ( T1R ) . T1R is a rare example where beneficial inflammatory responses are temporal separated from host pathological responses . There is strong experimental evidence that supports a role of host genetic factors in T1R susceptibility . Here , we employed a genome-wide association study ( GWAS ) to investigate susceptibility factors for T1R in Vietnamese families . We followed up the initial GWAS findings in independent population samples from Vietnam and Brazil and identified a set of cis-eQTL genetic variants for the ENSG00000235140 lncRNA as global risk factors for T1R . To test our proposal that T1R is a strong model for pathological inflammatory responses we evaluated if inflammatory bowel disease ( IBD ) genetic risk-factors were enriched among T1R risk factors . We observed that more than 10% of IBD-risk loci were nominally associated with risk for T1R suggesting a shared mechanism of excessive inflammatory response in the both disease etiologies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"alleles",
"long",
"non-coding",
"rnas",
"ethnicities",
"vietnamese",
"people",
"bacterial",
"diseases",
"mathematics",
"statistics",
"(mathematics)",
"gastroenterology",
"and",
"hepatology",
"genome",
"analysis",
"neglected",
"tropical",
"diseases",
"inflammatory",
"bowel",
"disease",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"mathematical",
"and",
"statistical",
"techniques",
"statistical",
"methods",
"genetic",
"loci",
"people",
"and",
"places",
"biochemistry",
"rna",
"nucleic",
"acids",
"meta-analysis",
"genetics",
"leprosy",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"physical",
"sciences",
"genomics",
"non-coding",
"rna",
"computational",
"biology",
"human",
"genetics"
] |
2017
|
A genome wide association study identifies a lncRna as risk factor for pathological inflammatory responses in leprosy
|
The fungus Cryptococcus neoformans can undergo a-α bisexual and unisexual reproduction . Completion of both sexual reproduction modes requires similar cellular differentiation processes and meiosis . Although bisexual reproduction generates equal number of a and α progeny and is far more efficient than unisexual reproduction under mating-inducing laboratory conditions , the α mating type dominates in nature . Population genetic studies suggest that unisexual reproduction by α isolates might have contributed to this sharply skewed distribution of the mating types . However , the predominance of the α mating type and the seemingly inefficient unisexual reproduction observed under laboratory conditions present a conundrum . Here , we discovered a previously unrecognized condition that promotes unisexual reproduction while suppressing bisexual reproduction . Pheromone is the principal stimulus for bisexual development in Cryptococcus . Interestingly , pheromone and other components of the pheromone pathway , including the key transcription factor Mat2 , are not necessary but rather inhibitory for Cryptococcus to complete its unisexual cycle under this condition . The inactivation of the pheromone pathway promotes unisexual reproduction despite the essential role of this pathway in non-self-recognition during bisexual reproduction . Nonetheless , the requirement for the known filamentation regulator Znf2 and the expression of hyphal or basidium specific proteins remain the same for pheromone-dependent or independent sexual reproduction . Transcriptome analyses and an insertional mutagenesis screen in mat2Δ identified calcineurin being essential for this process . We further found that Znf2 and calcineurin work cooperatively in controlling unisexual development in this fungus . These findings indicate that Mat2 acts as a repressor of pheromone-independent unisexual development while serving as an activator for a-α bisexual development . The bi-functionality of Mat2 might have allowed it to act as a toggle switch for the mode of sexual development in this ubiquitous eukaryotic microbe .
Sexual reproduction is generally considered bisexual , involving partners of the opposite sex . However , unisexual development ( self-fertile or inbreeding ) occurs in some eukaryotes without the need for an opposite mating partner [1–4] . Each reproduction mode has its own costs and benefits . Bisexual reproduction promotes outcrossing , but finding a compatible mating partner can be challenging . Unisexual reproduction promotes inbreeding/selfing , but it avoids the cost associated with locating a mating partner . The latter might be challenging for species that are immobile or species where compatible partners are rare . Fungi are classified as homothallic or heterothallic based on their requirement for a compatible mating partner for sexual reproduction . The heterothallic life cycle or the a-α bisexual reproduction was elaborated for the fungus Cryptococcus neoformans in 1970s and it includes the yeast-to-hypha morphological transition and sporulation [5–7] . It was noticed that some isolates ( mostly α ) underwent a similar cellular differentiation process ( filamentation and sporulation ) on their own in the absence of cells of the opposite mating type . This process was termed “haploid fruiting” and was originally considered a mitotic or asexual event [8] . Later , “haploid fruiting” was found to be a meiotic process [9–11] . Sporulation during this process requires the key meiosis-specific genes and the resulting progeny show high levels of recombination frequencies [9 , 11–14] . Such unisexual reproduction can occur by fusion between cells of the same mating type ( the same sex mating ) [9 , 15] , or more frequently , by endoreplication [16 , 17] . Both modes of sexual development in Cryptococcus involve the yeast-to-hypha morphological transition and the formation of fruiting bodies known as basidia and meiotic basidiospores ( Fig 1A ) . Same environmental factors , such as nutrient limitation , dehydration , and copper , are known to influence unisexual and bisexual reproduction [18 , 19] . Bisexual reproduction generates equal number of a and α progeny with comparable fitness [20–22] , with the obvious exception that α isolates have enhanced ability to undergo unisexual development [9 , 18 , 21] . This offers a plausible explanation for the predominance of the α mating type ( 99% ) among the clinical and environmental isolates [15 , 16 , 20 , 23] . However , compared to bisexual reproduction , unisexual development is far less efficient in terms of robustness in filamentation ( hyphal growth ) and sporulation under laboratory conditions [9 , 24 , 25] , presenting a challenge to the proposed importance of unisexual reproduction in nature . Unisexual reproduction and bisexual reproduction have both shared and unique features ( Fig 1A ) . However , no genetic factor that specifies cryptococcal commitment to unisexual development has been identified . Bisexual mating proceeds through cell fusion , the formation of dikaryotic hyphae , the fusion of two parental nuclei at basidium heads , and meiosis and sporulation . Unisexual development forms monokaryotic hyphae ( or self-filamentation ) . Diploidization is highly plastic and can be derived from cell-cell fusion or endoreplication at any development stage prior to meiosis and sporulation at basidium heads [3 , 26 , 27] . Thus , the decision in choosing bisexual development must have occurred prior to cell fusion . The Sxi1α-Sxi2a homeodomain proteins are specific to bisexual reproduction , and they are critical in maintaining the dikaryotic state of the hyphae post α-a cell fusion [28 , 29] . Because Sxi1α/Sxi2a are not necessary for α-a-conjugation [28 , 29] , this homeodomain complex functions after the choice of bisexual reproduction mode has been made . The master regulator of filamentation , Znf2 , is required for filamentation and subsequent development to complete the life cycle during both unisexual and bisexual reproduction [30 , 31] ( Fig 1A ) . Thus , Znf2 is not specific to any reproduction mode . The transcription factor Mat2 activates the expression of multiple components of the pheromone sensing pathway by binding to the pheromone response elements ( PREs ) in the promoter region of the respective genes [30 , 32] . Consequently , the mat2Δ mutant cannot undergo cell fusion even in the presence of a compatible wild-type mating partner [30] . We showed previously that Mat2 also plays an important role in deciding which parental mitochondrial DNA will be inherited in the progeny even prior to cell fusion ( prezygotic control of uniparental mitochondria inheritance ) [33] . Hence , Mat2 greatly influences the behavior of the mating partners prior to actual cell fusion , making it a likely candidate involved in early decision-making process for the mode of sexual development . In sexual reproduction , meiosis ( reduction division ) follows polyploidization , which is achieved by a-α cell fusion during bisexual development . However , unisexual reproduction can proceed through cell fusion or endoreplication [3 , 9] . We previously found that the frequency of cell fusion during unisexual mating was quite low [9] and evidence from population genetics studies indicates that endoreplication might be the major route for ploidy increase [16] . Thus it is conceivable that the pheromone pathway , which is designed for non-self-recognition between a and α cells but not for cells of the same mating type [34 , 35] , might not be as important for unisexual development . Some intriguing observations from previous studies support this notion [36–39] . For instance , deletion of the pheromone receptor gene CPRα ( STE3 ) almost abolished bisexual mating ( measured by quantifying a-α cell fusion events ) , but the abundance of hyphae produced by wild type and the ste3Δ mutant alone ( self-filamentation ) was similar [36] . Likewise , bisexual mating was impaired when the pheromone transporter gene STE6 was disrupted , but the STE6 gene deletion did not affect self-filamentation [38] . Similarly , when the G-protein α subunit genes GPA2 and GPA3 involved in sensing pheromone were both deleted , the gpa2Δgpa3Δ mutant was sterile in bisexual mating but still robust in self-filamentation [39] . However , the finding that Mat2 was necessary for both unisexual and bisexual development [30 , 31] is contradictory to the notion that the pheromone pathway might not be critical for unisexual development . We set out this study to resolve the conflict and to investigate if one of the roles of the pheromone pathway is priming Cryptococcus to choose bisexual over unisexual reproduction . Our results indicate that unisexual reproduction can be completed independent of the pheromone pathway under multiple disparate conditions , thus resolving previously contradictory findings .
Mat2 regulates the pheromone response pathway and is required for cell fusion during bisexual ( a-α ) mating [30–32] . The failure in a-α cell fusion blocks further development into dikaryotic hyphae ( Fig 1A ) . When non-self-filamentous strains are used , successful a-α cell fusion during bisexual mating will lead to production of dikaryotic hyphae ( Fig 1B ) . As expected , no hypha differentiation was observed when the mat2Δ mutant was co-cultured with an opposite mating partner under all mating-inducing conditions tested ( bottom panel , Fig 1B ) , consistent with the established role of Mat2 in the pheromone pathway . XL280 is a self-filamentous strain commonly used to study unisexual development [10 , 12–14 , 18] . The deletion of MAT2 in XL280 abolished self-filamentation under all mating-inducing conditions tested ( V8 , Yeast Nitrogen Base , Filament Agar , and Murashige and Skoog media ) ( top panel , Fig 1B ) [30] . Thus , Mat2 was considered crucial for both unisexual and bisexual reproduction . If the hypothesis of the non-essentiality of the pheromone pathway for unisexual development is valid , then we would predict that the disruption of Mat2 , the master regulator of the pheromone sensing pathway , should not abolish the ability of Cryptococcus to undergo self-filamentation . We previously found that copper ( ≤ 100 μM ) enhances self-filamentation [18] . Copper is also one of the components in the V8 juice medium that promote mating [19] . Thus , we decided to test the impact of copper at various concentrations on the mat2Δ mutant . The colony of the mat2Δ mutant grown on V8 medium was smooth and round due to the presence of only yeast cells ( Fig 2A and 2B ) , as expected based on previous studies [30 , 32] . We found that copper at 400 μM induced self-filamentation in the mat2Δ mutant ( Fig 2A ) . Accordingly , the colony of the mat2Δ mutant grown on the V8+copper medium appeared fluffy due to profuse production of filaments ( Fig 2B ) . Consistently , the addition of the copper chelator BCS ( bathocuproinedisulfonic acid ) to the V8+copper medium ( 400 μM ) reduced the robustness of filamentation shown by the mat2Δ mutant in a dose-dependent manner ( Fig 2C ) . None of the other metal ions tested , namely iron , magnesium or zinc , induced filamentation in the mat2Δ mutant ( Fig 2D ) . Addition of copper to defined media such as Filament Agar medium or Yeast Nitrogen Base medium also triggered filamentation in the mat2Δ mutant ( S1 Fig ) , indicating that the effect of copper on Mat2-independent filamentation is not limited to the complex V8 juice medium . To examine if the ability of the mat2Δ mutant to filament in response to copper is specific to the XL280 background , we tested the mat2Δ mutant made in the JEC21 background . Wild type JEC21 produced sporadic filaments on V8 medium and filamentation was slightly increased on V8+copper medium ( S2 Fig ) . The mat2Δ mutant in the JEC21 background filamented more robustly on V8+copper medium ( S2 Fig ) . The mat2Δ mutant was also able to produce basidia and spores on V8+copper medium , although at a lower frequency ( S3 Fig ) . We also tested the mat2Δ mutants made in the backgrounds of two congenic pairs: XL280α/XL280a ( S1 Fig ) and JEC21α/JEC20a ( S4 Fig ) . The mat2Δ mutants , either of the mating type α or the mating type a , filamented on V8+copper and FA+copper media . Taken together , the mat2Δ mutant made in different genetic or mating type backgrounds can undergo self-filamentation and sporulation in response to copper . Thus , unisexual reproduction does not depend on this transcription factor of the pheromone pathway . Mat2 controls multiple components of the pheromone sensing pathway and deletion of the MAT2 gene abolishes the ability of cells to produce or to respond to pheromone [30 , 32] . Under mating-inducing condition , Mat2 activates Znf2 , the master regulator of filamentation [30 , 31] . However , Znf2 itself is not critical for the pheromone sensing or response [30] . To determine if filamentation in the mat2Δ mutant evoked by copper still requires Znf2 , we generated the mat2Δznf2Δ double mutant . Although the znf2Δ mutant does not self-filament , the znf2Δ mutant can mate with a wild-type partner of a compatible mating type during bisexual mating on V8 medium and the fused heterokaryon will produce filaments ( S5 Fig ) [30] . The mat2Δznf2Δ double mutant , however , failed to filament when co-cultured with the compatible wild-type partner on V8 medium ( S5 Fig ) , consistent with the essential role of Mat2 in cell fusion . We then tested self-filamentation of this znf2Δmat2Δ double mutant on V8+copper medium . As expected , the mat2Δ mutant filamented on V8+copper medium . The znf2Δmat2Δ double mutant failed to filament on V8+copper medium , similar to the znf2Δ single mutant ( Fig 3A ) . These results indicate that filamentation in the mat2Δ mutant stimulated by copper still requires Znf2 . The finding further corroborates Znf2 as the essential regulator of filamentation . To examine if filaments produced by the mat2Δ mutant on V8+copper share the same molecular features as filaments produced by the wild type , we examined the expression pattern of a few proteins known to be expressed in wild-type hyphae . We first examined the hypha-specific protein Cfl1 [31] . Cfl1-mCherry expressed under the control of CFL1’s native promoter was detected in hyphal subpopulations when wild-type cells were cultured on V8 medium , but not in yeast cells when the wild-type strain was cultured in YPD medium ( Fig 3B ) . As expected , no Cfl1-mCherry could be detected when the mat2Δ mutant was cultured on YPD or on V8 medium ( Fig 3B ) , consistent with the mat2Δ mutant forming only yeast cells under these conditions . By contrast , Cfl1 could be clearly detected on the hyphae produced by the mat2Δ mutant grown on V8+copper medium ( Fig 3B ) . Another protein , Dha1 , which is enriched in basidia produced by wild-type hyphae [40] , showed similar localization in the basidia produced by either the wild type cultured on V8 or the mat2Δ mutant cultured on V8+copper medium ( Fig 3C ) . These results suggest that filaments produced by the mat2Δ mutant shares the same molecular features with that of the wild type . The ability of the mat2Δ mutant to filament on V8+copper medium could be caused by cryptic activation of pheromone . To test this , we examined the transcript level of the pheromone gene MFα1 in wild type and in mat2Δ cultured on V8 or V8+copper medium for the indicated time periods . CFL1 , the hyphal specific gene downstream of Znf2 was used as a marker for filamentation [31] . As expected , the CFL1 transcript level was increased when wild type was cultured on V8 medium compared to that on YPD medium ( Fig 4A ) . The MFα1 transcript level was also induced in wild type cultured on V8 medium ( Fig 4A ) , consistent with the known induction of pheromone under this condition [18 , 30] . By contrast , there was no induction in the CFL1 transcript level or the MFα1 transcript level in the mat2Δ mutant on V8 medium ( Fig 4A ) , which corroborates the established role of Mat2 in the pheromone pathway and the non-filamentous phenotype of the mat2Δ mutant under this condition . Consistent with the real-time PCR results , northern blot analysis also indicated no induction for MFα1 in the mat2Δ mutant cultured on V8 medium ( Fig 4C ) . On V8+copper medium , the transcript level for both CFL1 and MFα1 were increased in wild type compared to that in YPD medium ( Fig 4B ) . For the mat2Δ mutant cultured on V8+copper medium , the transcript level of CFL1 was increased ( Fig 4B ) . This is consistent with the expression of Cfl1 protein ( Fig 3B ) and the filamentous phenotype of the mat2Δ mutant cultured under this condition . However , there was no induction of MFα1 . Similarly , no MFα1 was detected in the mat2Δ mutant on V8+copper medium by northern blot ( Fig 4C ) . Thus , regardless of the conditions used , there was no detectable induction of pheromone in the mat2Δ mutant . To further attest pheromone-independent nature of filamentation in the mat2Δ mutant , we compared the transcript level of other components of the pheromone sensing pathway in the wild type and the mat2Δ mutant on V8 and V8+copper medium using RNA-seq . Regardless of the condition used , we found that the transcript levels for many components in the pheromone pathway were low in the mat2Δ mutant , including the pheromone transporter gene STE6 and the pheromone receptor gene STE3 ( Fig 4D ) . By contrast , the transcript level of CFL1 was induced in the mat2Δ mutant on V8+copper medium , consistent with the filamentous phenotype of the mat2Δ mutant under this condition . Phenotypic examination of the triple pheromone mutant mfα1–3Δ and the pheromone transporter ste6Δ mutant revealed slightly enhanced self-filamentation of these mutants when cultured on V8+copper medium ( S6 Fig ) . Collectively , these results demonstrate that filamentation in mat2Δ on V8+copper medium is independent of pheromone or the components of the pheromone sensing pathway . As the mat2Δ mutant can filament independent of the pheromone pathway on V8+copper medium , one would predict that factors affecting pheromone production should not affect mat2Δ filamentation . Light is known to repress the pheromone genes through the light sensor complex [41] . Here , we cultured the wild type and the mat2Δ mutant strains on V8 or V8+copper medium in the dark or under constant light . Wild type cultured on V8 medium showed drastic reduction in filamentation when it was exposed to constant light ( Fig 5A ) , consistent with the idea that filamentation on V8 medium is primarily driven by the pheromone pathway . The mat2Δ mutant grown on V8 medium did not show any filamentation either under constant light or in the dark . Both wild type and the mat2Δ mutant filamented equally well on V8+copper medium in the dark or in presence of light ( Fig 5A ) . These results indicate that light does not inhibit filamentation on V8+copper medium , consistent with our prediction of its pheromone independent nature . A previous study showed that prior growth at a high temperature can prime self-filamentation in C . neoformans once cells are transferred to filamentation-inducing conditions [42] . We decided to test if prior growth at a high temperature can induce filamentation in the mat2Δ mutant . We cultured the wild type , the mat2Δ mutant , and the znf2Δ mutant at 22°C , 30°C , and 37°C on YPD medium before transferring them onto V8 medium for additional incubation at 22°C . As expected , wild type filamented and the znf2Δ mutant showed no filamentation under all three conditions tested ( Fig 5B ) . The mat2Δ mutant transferred from prior cultures at 37°C , but not at 22°C or 30°C , filamented ( Fig 5B ) . This finding indicates that high temperature-induced filamentation can be independent of pheromone . To identify components required for filamentation in the mat2Δ mutant , we decided to perform forward genetic screen in the mat2Δ mutant background via Agrobacterium-mediated insertional mutagenesis . We generated 77 , 000 T-DNA insertional mutants in the mat2Δ background and screened these mutants on V8+copper medium for isolates that failed to filament ( S7A Fig ) . We isolated 47 insertional mutants that showed only yeast growth on V8+copper medium . Genomic DNA of these selected insertional mutants were pooled into four groups and sequenced , and the T-DNA insertion sites were identified using a recently described approach [43] . A total of 93 insertion sites were identified ( S7B Fig ) , indicating that multiple insertions occurred in some strains and that not all insertions were responsible for blocking filamentation . Indeed , T-DNA insertions into CNB05100 and CNC04530 were found to be not responsible for abolishing filamentation on copper medium based on our independent gene deletion experiments . Of the identified insertions , 32 occurred in intergenic regions and the rest inserted within ORFs . A couple of insertions occurred in the same regions . Genetic loci affected by the T-DNA insertions from the 47 mutants and their annotated gene functions are included in S3 Table . These genes encode membrane proteins , mitochondrial proteins , transporters , and hypothetic proteins . Some of the insertion sites were also identified through inverse PCR as previously described [30] . Because the mat2Δ parental strain is sterile in bisexual mating , we could not perform genetic linkage analysis by crossing to separate the insertions that are linked to the non-filamentous phenotype from those that are unlinked . By combining the T-DNA insertional mutagenesis data and the RNA-seq data , we selected a few candidate genes of interest for targeted gene deletion . The deletion of the genes recapitulated the phenotype caused by the insertions . A total of 1068 genes were upregulated by more than 2 fold in the mat2Δ mutant cultured on V8+copper ( filamentous colony ) compared to that on V8 ( yeast colony ) based on our RNA-seq data ( S4 Table ) . To narrow down the genes that are unique for filamentation in the mat2Δ mutant , we compared transcripts of the mat2Δ mutant and wild type grown on V8+copper to those grown on V8 that were upregulated for more than 3 fold . A total of 196 transcripts were shared between wild type and the mat2Δ mutant that were upregulated on V8+copper and 264 transcripts were unique to the mat2Δ mutant ( S8A Fig ) . We found that multiple components of the calcineurin pathway were upregulated in the mat2Δ mutant on V8+copper medium ( S8B Fig ) . One of the candidate genes identified through DNA sequencing as well as through inverse PCR sequencing of the insertional mutants was CNB1 . CNB1 encodes the regulatory subunit of calcineurin , a serine threonine specific phosphatase [44–46] . Calcineurin is a heterodimer composed of the catalytic subunit Cna1 and the regulatory subunit Cnb1 . Calcineurin is required for fungal adaptation to different environment conditions such as ion stress , pheromone response , morphogenesis , and growth at 37°C [45] . To confirm that the disruption of CNB1 abolished filamentation , we deleted the CNB1 gene in the wild-type XL280 background . The deletion of CNB1 caused severe growth defect at 37°C ( S9 Fig ) , consistent with what was reported previously of the cnb1Δ mutant made in the JEC21 background [44] . The cnb1Δ strain failed to produce any filament either on V8 medium or V8+copper medium ( Fig 5C and 5D ) . This observation suggests that the Cnb1 is required for filamentation regardless of the pheromone response . To test if the pheromone response pathway is functional in the cnb1Δ mutant made in the XL280 background , we crossed the cnb1Δα cells with JEC20 ( a ) . Because JEC20 is a mating type a strain that does not self-filament , any filaments observed from the cross would be the result of cell fusion events between cnb1Δ α mutant and JEC20 ( a ) . Filamentation was observed from the cross between cnb1Δ α and JEC20 ( a ) ( Fig 5C ) , indicating that Cnb1 is not critical for the pheromone response . This is again consistent with previous studies in the JEC21 background , indicating that the deletion of CNB1 does not have any defect in pheromone production or cell fusion [47] . To further confirm that the deletion of CNB1 is responsible for the blocked filamentation of mat2Δ on V8+copper medium identified by our insertional mutagenesis screen , we generated the cnb1Δmat2Δ double mutant . The cnb1Δmat2Δ double mutant did not produce any filaments when crossing with JEC20 ( a ) ( Fig 5C ) , consistent with the cell fusion defect caused by the deletion of MAT2 [30] . The cnb1Δmat2Δ double mutant failed to undergo self-filamentation either on V8 or V8+copper medium ( Fig 5C and 5D ) , supporting the idea that Cnb1 is required for both pheromone-dependent and pheromone-independent filamentation . Similar to the deletion of CNB1 , the deletion of the catalytic subunit of calcineurin CNA1 did not prevent crossing with a compatible wild-type mating partner during bisexual mating , but it blocked self-filamentation ( Fig 5C ) . The double mutant cna1Δmat2Δ , similar to the double mutant cnb1Δmat2Δ , was unable to filament either on V8 or V8+copper medium ( Fig 5C and 5D ) . Since calcineurin is activated in response to increased calcium levels [48] , we examined if addition of CaCl2 could induce filamentation in the mat2Δ mutant . Indeed , we found that addition of CaCl2 at high concentrations ( between 1000–2000 μM ) induced filamentation in the mat2Δ mutant after prolonged incubation in the dark on V8 medium ( S8C Fig ) . We showed earlier that prior exposure to high temperature also induced filamentation in the mat2Δ mutant . Given that calcineurin is activated in response to calcium and also to environmental stresses such as high temperature [49 , 50] , we reasoned that filamentation induced by high temperature may also require calcineurin . We used the calcineurin inhibitor FK506 to test this hypothesis as the deletion of the CNA1 gene or the CNB1 gene causes severe growth defect at 37°C ( S9 Fig ) [47] . Here , we cultured wild type and the mat2Δ mutant at 37°C and then transferred the cells onto V8 medium containing FK506 ( 1 μg/ml ) . No filamentation was observed in either wild type or the mat2Δ mutant ( Fig 6A ) , indicating that treatment with FK506 abolished filamentation . Hence , thermo-induced filamentation also requires calcineurin . Collectively , these data confirm that filamentation , be it pheromone-dependent or pheromone-independent , requires calcineurin . Because Crz1 is a known transcription factor downstream of calcineurin and because its transcript level was induced in the mat2Δ mutant on V8+copper medium ( S8B Fig ) , we decided to examine if calcineurin acts on copper-induced filamentation through Crz1 . For this purpose , we generated the crz1Δ mutant as well as the crz1Δmat2Δ double mutant . Either the crz1Δ mutant or the crz1Δmat2Δ double mutant showed obvious growth defect at 37°C ( S9 Fig ) , indicating that Crz1 is not as critical as calcineurin for thermal-adaptation . Such phenotype is expected based on the crz1Δ mutant phenotype reported for the serotype A background [49] . However , the deletion of CRZ1 in the wild-type background did not affect filamentation on V8 or V8+copper medium . Likewise , the deletion of CRZ1 in the mat2Δ mutant background did not affect the ability of mat2Δ to filament on V8+copper medium ( Fig 6D ) . This result suggests that Crz1 is dispensable for filamentation on copper medium . Thus , downstream factors of the calcineurin pathway other than Crz1 [49] must be required for filamentation induced by copper . Znf2 is a specific regulator for hyphal morphogenesis and it is not critical for adaptation to many other stresses tested [30 , 31] . Consistently , no enrichment of stress regulatory elements ( STEs ) was observed within 2 kb sequences upstream of the open reading frame of ZNF2 . Although the core sequence of metal regulatory elements ( 5’-GCTG-3’ ) [51] are enriched in this region , the copper specific sensing element ( 5’-ATATTGCTGT-3’ ) is absent . This morphogenesis regulator is expressed at low levels except under conditions that induce filamentation . So far no factor has been identified that can override the need for Znf2 in terms of filamentation . It is thus not surprising that overexpression of ZNF2 can enable filamentation in various mutants ( e . g . mat2Δ ) . Like Znf2 , calcineurin is also essential for Mat2-dependent or Mat2-independent filamentation ( this study and previous studies [30 , 47] ) . In contrast to Znf2 , calcineurin is a general stress response regulator in C . neoformans and in other fungi [52–56] . We hypothesize that this general stress adaptation regulator cooperates with Znf2 in controlling filamentation in Cryptococcus . If this hypothesis is valid , then the defect in filamentation caused by the disruption of calcineurin would not be overcome by overexpressing ZNF2 . To test this hypothesis , we introduced the ZNF2 overexpression construct into the mat2Δ and the mat2Δcnb1Δ double mutant . Overexpression of ZNF2 in the mat2Δ mutant enabled filamentation ( Fig 6B ) , as we reported previously [31] . However , overexpression of ZNF2 in the mat2Δcnb1Δ mutant failed to confer filamentation ( Fig 6B ) . Consistently , deletion of the CNA1 gene abolished filamentation in the ZNF2oe strain on V8 medium or V8+copper medium ( Fig 6C ) . These results indicate that both Znf2 and calcineurin are required for pheromone-dependent and pheromone-independent filamentation . To examine if calcineurin affects the localization or the stability of Znf2 , we tested the effect of the calcineurin inhibitor FK506 on mCherry-tagged Znf2 that was controlled by the promoter of the copper transporter CTR4 [51] . The expression of mCherry-Znf2 was suppressed by copper and induced in the presence of the copper chelator BCS ( Fig 7A ) , consistent with our previous studies [57] . When cells expressing mCherry-Znf2 were exposed to FK506 , the fluorescence intensity was sharply reduced ( Fig 7B , 7D and 7F ) . However , signals of mCherry-Znf2 could still be detected in the nucleus ( Fig 7B ) , suggesting that treatment with FK506 primarily affected the protein level of Znf2 rather than its nuclear localization . As a control , we used a strain carrying a mCherry tagged Phd11 , a protein with two PHD finger domains ( plant homeodomains ) that are conserved readers of histone modifications in eukaryotes [58] . Treatment with FK506 did not affect either the nuclear localization or the intensity of Phd11 ( Fig 7C , 7E and 7F ) . This suggests that reduced mCherry-Znf2 signal is not a general effect caused by FK506 . Reduction in the Znf2 protein level after FK506 treatment was also observed by Western blot analyses ( Fig 7E ) . Collectively , these results suggest that calcineurin may regulate filamentation through controlling the stability of the master regulator Znf2 .
The eukaryotic microbe C . neoformans can undergo bisexual reproduction as well as unisexual reproduction . Unisexual reproduction in this fungus is proposed to have given rise to the sharply skewed population towards the α mating type in the current natural population [9 , 11 , 15 , 16 , 59] . However , the inefficiency of unisexual development compared to bisexual development under mating-inducing laboratory conditions challenges this view . One explanation for the conflicting observations is that the laboratory conditions currently used for sexual reproduction might not mimic natural niches of Cryptococcus that favor unisexual reproduction . An alternative but not mutually exclusive hypothesis is that there is bifurcation of genetic regulation for unisexual and bisexual development . Certain environmental conditions may favor unisexual development while suppressing the activation of genetic pathways required for bisexual development . To test this hypothesis , we used the mat2Δ mutant , which is blocked for bisexual mating , to search for conditions that specifically promote unisexual development in Cryptococcus . We found that C . neoformans could indeed undergo unisexual development without Mat2 . We further showed that filamentation in the mat2Δ mutant is independent of pheromone or other components of the pheromone sensing pathway . Furthermore , we found that several disparate conditions ( prior heat exposure , copper , or calcium ) can induce pheromone-independent filamentation . These conditions are likely encountered by this fungus in nature , as C . neoformans is known to be associated with soil and vegetation . Some leafy greens such as kale contain copper ( ~240 μM ) and calcium ( ~37 mM ) at concentrations high enough to stimulate unisexual development . Thus , it is conceivable that some natural niches will provide suitable conditions for pheromone-independent unisexual development in Cryptococcus . The pheromone sensing pathway is crucial for bisexual mating in Cryptococcus and also in a wide variety of fungal species across different phyla , such as Saccharomyces , Candida , and Ustilago [60–63] . In essence , pheromone triggers non-self-recognition ( pheromone produced by α cells preferentially binds to the pheromone receptor present in a cells and vice-versa ) , which then promotes cell fusion . Accordingly , the efficiency of bisexual mating is often quantified by cell fusion events [36] . However , unlike bisexual reproduction , unisexual reproduction can proceed either through cell-cell fusion or endoreplication , and the latter is likely the major route for ploidy increase [3 , 9 , 16] . Increase in ploidy by endoduplication often occurs as a response to stress in eukaryotes [64–66] , and sexual reproduction in many fungi , including Cryptococcus , takes place under stressful conditions . After increased ploidy through endoduplication , Cryptococcus could complete sexual reproduction by meiosis to return to the haploid state . Contradictory to the non-essentiality of pheromone for unisexual development in Cryptococcus , the transcription factor of the pheromone pathway Mat2 was shown to be required for both unisexual and bisexual development under all the mating-inducing conditions tested so far . Thus , this study resolves the conflicting observations for and against the crucial role of the pheromone pathway in unisexual development [36–38 , 61] . Collectively , this and previous studies firmly establish that the very process that defines bisexual mating is not crucial for unisexual development . Some of the important morphological features are conserved in both pheromone-dependent and pheromone-independent sexual development . Both proceed through the formation of hyphae , basidia , and spores . Both require the master regulator of filamentation Znf2 and general stress regulator calcineurin . Filamentation marker Cfl1 is produced by hyphae generated during pheromone-dependent sexual development [31] and during pheromone-independent unisex . The cell surface protein Dha1[40] shows similar localization at basidia produced by pheromone- dependent or independent sexual development . Hence , the key downstream features seem to be conserved in both unisexual and bisexual development . Our findings demonstrate the possibility of sexual development under conditions and environments where pheromone is not activated , or in strains in which the pheromone pathway is genetically deactivated/non-functional . The observations that the pheromone ( mfα1–3Δ ) and pheromone transporter mutants self-filamented more robustly than the wild type on V8+copper medium , and that the mat2Δ mutant showed the extreme phenotypes of being non-filamentous on mating-inducing V8 medium while filamented robustly on V8+copper medium , suggest that the pheromone pathway might even exert an inhibitory effect on unisexual development under certain occasions . Our RNA-seq data suggest that the V8+copper medium is inhibitory to the expression of pheromone and other components of the pheromone pathway . It is possible that Mat2 itself represses genes that promote unisexual development . We are interested in interrogating this hypothesis in the future . The finding that disparate conditions could trigger pheromone-independent unisexual development reflects the robustness of sexual reproduction in Cryptococcus . Likewise , multiple cues such as temperature , nutrition starvation , N-acetyl glucosamine , oxidative stress , and genotoxic agents can induce the commensal and pathogenic fungus Candida albicans to switch from the “white” state to the mating competent “opaque” state [67 , 68] . Adding to the complexity is that both pheromone-dependent and pheromone-independent sexual reproduction modes likely occur simultaneously in different subpopulations in a wild-type cryptococcal community . The heterogeneity in the sexual mode in the community might reflect or be enforced by the stochastic expression of Mat2 within the population . Consistent with this idea , we noticed that not all cells responded to the calling of pheromone even under a strong mating-inducing condition [40] . Given that Cryptococcus is a ubiquitous environmental fungus and an opportunistic pathogen to a wide range of hosts , this fungus might have developed various tactics to propagate sexually under different conditions rather than the risky sole dependence on the pheromone pathway and the rare presence of a compatible mating partner . Hence , sexual reproduction , a well-known evolution feature for adaptation , is itself an adaption to varying selective pressures .
The strains used in this study are listed in S1 Table . Cryptococcus strains were maintained as glycerol stocks at -80°C . Yeast cells were grown on YPD medium ( 1% yeast extract , 2% Bacto peptone , 2% dextrose , 2% Bacto agar ) . Mating assays were performed on V8 pH7 solid medium ( 0 . 5 g/litre KH2PO4 , 4% Bacto agar , 5% of V8 juice from Campbell Soup Co . ) in the dark at 22°C with or without the addition of copper , other metal ions , or the copper chelator BCS as indicated in the texts and figures . Copper at 400 μM was used in most experiments unless indicated otherwise . Other media used for filamentation assays include MS medium ( Murashige and Skoog medium minus sucrose , Sigma-Aldrich ) , YNB medium ( 6 . 7 g/L Yeast Nitrogen Base without amino acids , 2% Bacto agar , 2% dextrose ) , and Filament Agar medium ( 6 . 7 g/L Yeast Nitrogen Base without ammonium sulphate , 0 . 5% glucose , and 4% Bacto-agar at pH 5 . 0 ) as described previously [8 , 69] . Gene deletion was carried out using the split marker recombination approach as we previously described [70] . Briefly , the 5’ and 3’ sequences of approximately 1 kb that flank the open reading frame of the gene of interest were fused to two third of the NAT or NEO drug resistance marker respectively using overlap PCR . Dominant drug markers NAT and NEO were amplified from the plasmids pPZP-NATcc and pPZP-NEO1 [71] . The gene deletion construct ( mixture of two fragments ) was introduced into the recipient strain through biolistic transformation as described previously [72] . Gene deletion was confirmed through diagnostic PCR as we described previously [70] . For the mutant strains when bisexual mating was possible , the linkage between the gene deletion and the observed mutant phenotypes was established through genetic linkage assay of micro-dissected basidiospores as described previously [73] . Briefly , meiotic progeny from the cross between a mutant and the congenic wild type of a compatible mating type was dissected . These progeny were analyzed for the mutant phenotype , mating type , and the presence of the gene deletion at the correct genetic locus to establish whether the phenotypes observed are genetically linked to the gene deletion . For complementation of the deletion strains , the ORF and 1–1 . 5 kb upstream sequence of the gene was amplified using the wild-type genomic DNA as the template . The amplicon was then cloned into the pPZP-NEO1 plasmid . The wild-type allele of the gene together with the drug selection marker was then introduced into the corresponding gene deletion strain through biolistic transformation . To construct overexpression strains , ORFs of the genes of interest were amplified by PCR , digested , and then ligated into the pXL1 plasmid after the GPD1 promoter region as we described previously [31] . This generates constitutive expression of the gene . For the inducible system , we replaced the PGPD1 region of the plasmid with the PCRT4-2 to generate the copper inducible system as described previously [31 , 51] . All primers used for generating gene deletion , complementation , or overexpression are listed in S2 Table . To generate mCherry tagged proteins , the mCheery gene was fused to the C-terminus of CFL1 , PHD11 , and DHA1 as we described previously [31] . To drive the gene expression with their native promoter ( e . g . PCFL1-CFL1-mCherry ) , 1 kb upstream of the gene’s ORF together with its ORF was amplified through PCR and ligated into the pXL1 vector containing mCherry such that mCherry is fused in frame with the C-terminus of the gene [31] . To drive the gene expression with the GPD1 promoter ( e . g . PGPD1-DHA1-mCherry ) , ORF of DHA1 fused with mCherry was introduced into pXL1 vector after the PGPD1 region [40] . The plasmids were then digested and introduced into the mat2Δ mutant using biolistic transformation to obtain the respective strains ( mat2Δ PCFL1-CFL1-mCherry and mat2Δ PGPD1-DHA1-mCherry ) . For self-filamentation , WT and mutant cells of equal optical density measured at 600 nm ( OD600 = 3 ) were plated onto V8 or V8+copper medium and incubated in the dark at 22°C for 6 days or as specified in the figures and texts . For most of the phenotypic assays , 400 μM of CuSO4 was added to the V8 juice medium . For testing different metal ions , 400 μM of the CaCl2 , FeCl3 , and ZnSO4 were used . For sporulation and protein localization studies , 150–200 μM of CuSO4 was used . Spores were visualized after incubation on V8 medium or V8 medium with copper for 2–3 weeks . For examining the ability of the mutant cells to undergo bisexual mating , non-self-filamentous reference strain of the mating type a , JEC20 , was used [74] . Colony images were acquired through a GO21 camera connected to the stereoscope Olympus SZX16 as we described previously [57] . Yeast , hyphae , basidia , and spores were captured by a Zeiss Axiocam 506 camera connected to a Zeiss Imager M2 epifluorescence microscope as we described previously [75] . The filter used for visualizing mCherry was the FL filter set 43 HE cy3 ( Carl Zeiss Microscopy ) . RNA extraction and qPCR was performed as we described previously [31] . For extraction of total RNA from cells grown on V8 or V8+copper medium , equal number of cells were plated onto the medium and cells were incubated in the dark at 22°C . For the control samples , cells were grown overnight in liquid YPD . At the indicated time points , cells were collected , washed with cold ddH2O , and lyophilized . Total RNA was extracted using Purelink RNA minikit ( Life Technologies ) followed by DNase treatment ( Ambion ) . The quality and quantity of the RNA samples were analyzed by electrophoresis on a denaturing formaldehyde agarose gel as we described previously [57] . First strand cDNA synthesis was performed using Superscript III cDNA synthesis kit ( Life Technology ) following the manufacturer’s instructions . Transcript levels were normalized using the constitutively expressed housekeeping gene TEF1 as we described previously [31] . Primers for real-time PCR are listed in S2 Table . The procedures for northern blot analyses were the same as we described previously [30] . Briefly , total RNA was extracted from cells grown on V8 or V8+copper medium for 16 hours . Poly ( A ) tailed RNAs were purified using PolyATtract mRNA isolation System IV ( Promega ) following manufacturer’s instructions . The random primers DNA labeling System ( Life technologies ) was used for generating probes for MFα and ACT1 . The primers used are listed in S2 Table . The WT strain XL280 and the mat2Δ mutant were cultured on YPD , V8 , and V8+copper ( 400 μM ) media . Cells were collected at 16 hours after inoculation for the extraction of total RNA . The total RNA samples were submitted to the TAMU AgriLife Centre for bioinformatics and genomic systems engineering for strand specific RNA-seq following the standard protocol for Illumina Genome Analyzer IIx ( http://www . txgen . tamu . edu/ ? s=sequencing&search . x=0&search . y=0&search=Search ) as we described previously [57] . Sequence reads were aligned to the XL280 reference sequence [13] as pairs with Tophat2 [13 , 76 , 77] . Genes with differential expression was investigated using DESeq [78] and edgeR [79] with default settings . A gene was considered as significantly differentially expressed only when it was identified by both DESeq and edgeR , and passed the false discovery rate cutoff ( FDR< = 0 . 05 ) . IGV software was used for viewing the transcripts [80] . RNA-seq data are deposited at NCBI ( BioProject ID: PRJNA344667 ) as the following SRA files: SRR5272486 , SRR5272476 , SRR5272474 , SRR5272472 , SRR5272484 , SRR5272482 , SRR5272480 , and SRR5272478 . Insertional mutagenesis was performed in the mat2Δ mutant made in the XL280 background ( NATR ) . Agrobacterium tumefaciens strain EHA105 containing the Ti-plasmid pPZP-NEO1 was used for the insertional mutagenesis as described previously [30 , 71] . A . tumefaciens was grown overnight in Luria-Bertani medium containing kanamycin at 22°C . Cells were washed twice with sterile water and grown on induction medium containing 100 μM Acetosyringone for additional 6 hours . Cryptococcus mat2Δ cells grown overnight in liquid YPD was washed and resuspended in induction medium to obtain the density of 1x107 cells/ml . Equal aliquots of fungal and bacterial cells were mixed and co-cultured on the induction medium ( 200 μl per drop ) for 3 days at 22°C in the dark . The cocultured cells were then collected and plated onto V8+cefotaxime+G418+NAT+copper medium . Agrobacterium cells were killed on this selective medium ( antibiotic cefotaxime ) and transformants with T-DNA ( confer G418 resistance ) would be able to survive . Most colonies were able to filament on the V8+copper medium , similar to the mat2Δ parental strain . Mutants that only grew in the yeast form after 2 weeks of incubation were selected . To identify insertion sites through inverse PCR coupled with sequencing , genomic DNA from the selected mutants was digested with a restriction enzyme , purified , and self-ligated as we described previously [30] . Primers AI076/Ai077 were used for inverse PCR and sequencing as we described previously [30 , 71] . After sequencing , flanking region sequences were used for BLAST search against C . neoformans serotype D genome database at Genebank to identify the insertion sites and the genetic loci affected by the insertion . A total of 47 candidate insertional mutants which could not filament on V8+copper medium were selected . These candidate strains were grown overnight in YPD liquid medium . Genomic DNA was extracted following cetyltrimethylammonium bromide ( CTAB ) extraction protocol as previously described [81 , 82] . Genomic DNA from 47 individual selected mutants was pooled into four groups , with each group consisting DNA from 11–12 mutants . 10 μg of total DNA from each pool was submitted to the TAMU AgriLife , Centre for bioinformatics and genomic systems engineering for sequencing ( Illumina Miseq 175 bp x 175 bp , paired end reads ) . Each group yielded 1 . 1x108 to 1 . 75x108 reads pairs ( 129 . 5 M pairs ) , yielding an average 12x coverage of Cryptococcus genome per strain . Analysis of the insertion sites was performed using the AIMHII approach as recently described for Cryptococcus T-DNA insertional mutants [43] . Strains with mCherry tagged Znf2 or Phd11 ( control ) were grown overnight in inducing condition in liquid YPD culture for 10 hours and cells were examined under a Zeiss Imager M2 epifluorescence microscope to visualize the mCherry signal . Cells of equal density ( OD600 = 3 ) were then treated with or without FK506 ( 2 μg/ml ) for additional 2 hours . Cells were then briefly washed with ddH2O and their images were taken with Zeiss Imager M2 epifluorescence microscope using same exposure time for each sample . Data for intensity plot was obtained through ZEN image software ( Carl Zeiss Microscopy , NY ) . Protein extraction and western blotting was performed as described previously [83–85] . PCTR4 –mCherry-Znf2 strains were grown in liquid YPD+BCS for 10 hours . Half of the samples were grown for 2 additional hours with the addition of FK506 ( 2 μg/ml ) whereas the other half of the samples were grown without FK506 . The cells were lyophilized after washing with cold phosphate-buffered saline ( PBS ) . The dried cells mixed with glass beads were disrupted by a Cell Disrupter ( Next Advance ) . Total protein extraction was carried out using the lysis buffer ( 300 mM NaCl , 2mM EDTA , 25mM HEPES ( pH 7 . 5 ) , proteinase inhibitor ) . The sample was denatured with SDS containing loading buffer . Samples were separated on SDS -12% PAGE gel and were transferred to a membrane using TE 70 ECL semidry transfer unit ( GE healthcare ) for 1 hour at 30V . Anti-mCherry primary antibody ( 1/2000 dilution ) and rabbit anti-mouse secondary antibody ( 1/10 , 000 dilution ) were used to detect mCherry signals . Signal detection was carried out using chemiluminescence ( ECL ) system according to the manufacturer’s instruction ( Pierce ) .
|
Like many higher eukaryotes , the fungus Cryptococcus neoformans is known to undergo bisexual mating that involves α and a partners . Surprisingly , the α mating type is dominant in nature ( >99% ) . Unisexual reproduction was previously discovered in Cryptococcus and the mating type α allele enhances this sexual reproduction mode . However , bisexual mating is far more robust than unisexual development under all current laboratory conditions tested . This challenges the predicted prevalence and importance of unisexual development in cryptococcal population . In this study , we identified a condition that favors unisexual reproduction and suppresses bisexual reproduction . We showed that unisexual reproduction occurring under this condition is independent of pheromone or the key components of the pheromone pathway , despite the fact that this pathway is crucial for bisexual development . Our findings demonstrate that the pheromone pathway is primarily used for identifying compatible mating partners through non-self recognition , and it is not required for pheromone-independent unisexual reproduction . This discovery opens the door to investigate how this eukaryotic microbe specifies different modes of sexual reproduction and the bifurcation of genetic control of these reproduction modes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussions",
"Methods",
"and",
"materials"
] |
[
"cell",
"physiology",
"cryptococcus",
"neoformans",
"medicine",
"and",
"health",
"sciences",
"cryptococcus",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"rna",
"extraction",
"microbiology",
"sexual",
"reproduction",
"developmental",
"biology",
"fungi",
"regulator",
"genes",
"gene",
"types",
"fungal",
"pathogens",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"mycology",
"medical",
"microbiology",
"microbial",
"pathogens",
"modes",
"of",
"reproduction",
"biochemistry",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"pheromones",
"organisms",
"cell",
"fusion"
] |
2017
|
Pheromone independent unisexual development in Cryptococcus neoformans
|
Aluminum ( Al ) toxicity is a primary limitation to crop productivity on acid soils , and rice has been demonstrated to be significantly more Al tolerant than other cereal crops . However , the mechanisms of rice Al tolerance are largely unknown , and no genes underlying natural variation have been reported . We screened 383 diverse rice accessions , conducted a genome-wide association ( GWA ) study , and conducted QTL mapping in two bi-parental populations using three estimates of Al tolerance based on root growth . Subpopulation structure explained 57% of the phenotypic variation , and the mean Al tolerance in Japonica was twice that of Indica . Forty-eight regions associated with Al tolerance were identified by GWA analysis , most of which were subpopulation-specific . Four of these regions co-localized with a priori candidate genes , and two highly significant regions co-localized with previously identified QTLs . Three regions corresponding to induced Al-sensitive rice mutants ( ART1 , STAR2 , Nrat1 ) were identified through bi-parental QTL mapping or GWA to be involved in natural variation for Al tolerance . Haplotype analysis around the Nrat1 gene identified susceptible and tolerant haplotypes explaining 40% of the Al tolerance variation within the aus subpopulation , and sequence analysis of Nrat1 identified a trio of non-synonymous mutations predictive of Al sensitivity in our diversity panel . GWA analysis discovered more phenotype–genotype associations and provided higher resolution , but QTL mapping identified critical rare and/or subpopulation-specific alleles not detected by GWA analysis . Mapping using Indica/Japonica populations identified QTLs associated with transgressive variation where alleles from a susceptible aus or indica parent enhanced Al tolerance in a tolerant Japonica background . This work supports the hypothesis that selectively introgressing alleles across subpopulations is an efficient approach for trait enhancement in plant breeding programs and demonstrates the fundamental importance of subpopulation in interpreting and manipulating the genetics of complex traits in rice .
Aluminum ( Al ) toxicity is the major constraint to crop productivity on acid soils , which comprise over 50% of the world's arable land [1] . Under highly acidic soil conditions ( pH<5 . 0 ) , Al is solubilized into the soil solution as Al3+ , which is highly phytotoxic , causing a rapid inhibition of root growth that leads to a reduced and stunted root system , thus having a direct effect on the ability of a plant to acquire both water and nutrients . Cereal crops ( Poaceae ) have been a primary focus of Al tolerance research [2] . This research has demonstrated that levels of Al tolerance vary widely both within and between species [3]–[8] . Of the major cereal species that have been extensively studied ( rice , maize , wheat , barley and sorghum ) , rice demonstrates superior Al tolerance under both field and hydroponic conditions [3] , [8] . Although rice is 6–10 times more Al tolerant than other cereals , very little is known about the genes underlying this tolerance . Based on its high level of Al tolerance and numerous genetic and genomic resources , rice provides a good model for studying the genetics and physiology of Al tolerance . In wheat , sorghum , and barley , Al tolerance is inherited as a simple trait , controlled by one or a few genes [9]–[11] . However , in maize , rice , and Arabidopsis , tolerance is quantitatively inherited [12] , [13] . Al tolerance genes have been cloned in wheat and sorghum . The wheat resistance gene , ALMT1 , encodes an Al-activated malate transporter [14] . The sorghum resistance gene , SbMATE , encodes a member of the multidrug and toxic compound- extrusion ( MATE ) family and is an Al-activated , root citrate efflux transporter [15]–[17] . Four mutant genes that lead to Al sensitivity in rice have recently been cloned , STAR1 ( Sensitive to Al rhizotoxicity1 ) , STAR2 ( Sensitive to Al rhizotoxicity2 ) , ART1 ( Aluminum rhizotoxicity 1 ) , and Nrat1 ( Nramp aluminum transporter 1 ) [18]–[20] . The products of STAR1 and STAR2 are expressed mainly in the roots and are components of a bacterial-type ATP binding cassette ( ABC ) transporter . Both are transcriptionally activated by exposure to Al and loss of function of either gene results in hypersensitivity to Al . STAR1 and STAR2 are similar to two Al sensitive mutants in Arabidopsis , als1 and als3 , also encoding ABC transporters [21] , [22] . ART1 is a novel C2H2-type zinc finger transcription factor that interacts with the promoter region of STAR1 . ART1 is reported to regulate at least 30 down-stream genes , some of which are involved in Al detoxification and serve as strong candidate genes controlling rice Al tolerance [19] . Nrat1 is one of the genes that is regulated by ART1 and was recently demonstrated to be an Al transporter that is localized to the root cell plasma membrane [18] , [20] . It is hypothesized that Nrat1 confers Al tolerance by transporting Al into the cell and reducing the concentration of Al in the cell wall [20] . None of the four cloned rice genes described above have been demonstrated to be involved in natural genetic variation of Al tolerance in rice and only one ( Nrat1 ) maps to a previously reported Al tolerance QTL [23] , suggesting that these genes may be involved in basal Al tolerance [19] , [20] , [24] . A more thorough analysis is necessary to determine whether there might be natural variation associated with these loci that would help trace their evolutionary origins and clarify their contribution to the high levels of Al tolerance observed in rice . Seven QTL studies on Al tolerance have been reported in rice using 6 different inter- and intra-specific mapping populations [13] , [25]–[29] . Together , these studies report a total of 33 QTLs , located on all 12 chromosomes , with three intervals ( on chromosomes 1 , 3 , and 9 ) being detected in multiple studies . In all of the QTL studies , Al tolerance was estimated based on relative root growth ( RRG ) , and specifically on inhibition of the growth ( elongation ) of the longest root ( elongation of the longest root in Al treatment/root growth of controls ) . Rice has a very fine and fibrous root system without dominant seminal roots . We recently showed that there is a weak correlation between rice Al tolerance based on RRG of the longest root and RRG of the total root system ( R2 = 0 . 17 ) [8] . This raises the question whether mapping Al tolerance QTL using total root and longest root RRG indices independently might identify novel loci , helping to integrate QTL studies with studies based on induced mutations . Historically , O . sativa has been classified into two varietal groups , Indica and Japonica , based on morphological characteristics , ecological adaptation , crossing ability and geographic origin [30] . These two varietal groups are believed to represent independent domestications from a pre-differentiated ancestral gene pool ( O . rufipogon ) , followed by significant gene flow among and between subpopulations [17] , [31]–[39] . These two varietal groups ( names are italicized with an upper case first letter , i . e . , Indica and Japonica ) have been further divided into five major subpopulations ( subpopulation names are italicized using all lower-case letters ) ( indica , aus , tropical japonica , temperate japonica , and aromatic [group V] ) based on DNA markers ( SSR , SNPs , indels ) [40]–[42] . Genotypes that share <80% ancestry across subpopulations or varietal groups are classified as admixed varieties [42] , while smaller groups adapted to specific ecosystems may be recognized as upland , deep water , or floating varieties [43] , [44] . Upland varieties , which are generally grown at high altitudes on dry ( non-irrigated ) soils , are those most commonly exposed to acidic , Al-toxic soil conditions . These varieties are almost invariably of tropical japonica origin , suggesting a priori that the tropical japonica subpopulation would be a likely source of superior alleles for Al tolerance in rice . Diverse panels of O . sativa are reported to have similar , or slightly elevated levels of linkage disequilibrium ( LD ) compared to species such as Arabidopsis , maize and human . The average extent of LD in rice has been estimated at between 50–500 kb [45]–[49] , depending on the germplasm evaluated , compared to 10–250 kb in Arabidopsis and human [50]–[57] , 100–500 kb in commercial elite maize inbreds and 1–2 kb in diverse maize landraces [58] , [59] . The inbreeding nature of O . sativa , coupled with its demographic history , are major determinants of genome-wide patterns of LD . Strong selective pressure over the course of rice domestication has also lead to deep population substructure ( Fst = 0 . 23 to 0 . 57 ) [40] , [42] , which sets it apart from Arabidopsis , in which population structure is gradual across geographic distances [60] , [61] . Population substructure can lead to false-positives in association mapping studies , and must be taken into account [61]–[63] . The mixed-model has been demonstrated to work well in both maize and Arabidopsis [61] , [63] , and it has also shown its ability to greatly reduce the false positive rates in rice when used within a single subpopulation [64] , though it may introduce false negatives when used on a diversity panel representing all domesticated subpopulations [65] . A diversity panel consisting of 413 O . sativa accessions , representing the genetic diversity of the primary gene pool of domesticated rice [66] , was recently genotyped with 44 , 000 SNPs ( ∼10 SNPs/kb ) [65] , [67] , [68] as the basis for GWA studies . The slow decay of LD , while facilitating GWA analysis , limits the resolution of association mapping in rice . The first targeted association mapping study in rice [45] demonstrated that LD decay in the aus subpopulation was approximately 90 kb ( ∼5 genes ) in a region on chromosome 5 containing the xa5 resistance gene . LD is expected to decay more quickly in O . rufipogon ( <50 kb , or 1–3 genes ) [48] , providing higher resolution for LD mapping , and more slowly in the japonica subpopulations [47]–[49] . Nonetheless , when compared to the resolution of a typical QTL study ( 250 lines ) ( ∼10–20 cM resolution , where 1 cM = ∼250 kb ) , association mapping is expected to provide between 10–200 times higher resolution for a population of similar size as long as sufficient marker density is obtained to exploit the historical recombination . Thus , an association mapping study that uses markers densities similar to a QTL study will not have the increased resolution and will increase the risk of type-2 error . For both GWA and QTL analysis in rice , fine-mapping and/or mutant analysis is generally required to identify the gene ( s ) underlying a QTL of interest . However , the fine-mapping phase can generally be focused on a smaller target region following GWA analysis . In this study , the genetic architecture of rice Al tolerance was investigated via bi-parental QTL analysis in two mapping populations using relative root growth of the longest root , the primary root system , and the total root system quantified with the digital root phenotyping methods described previously for rice Al tolerance [8] . Subsequently , genome wide association ( GWA ) analysis was undertaken using 36 , 901 high quality SNPs that had been genotyped on the rice diversity panel [65] . Regions identified by GWA were compared with regions identified as QTLs in bi-parental mapping populations for both this and previous studies , as well as with Al sensitive mutants and/or candidate genes . Phenotypic outliers identified in the diversity panel were further investigated to identify regions of subpopulation-admixture that accounted for extreme Al tolerance phenotypes .
Three hundred eighty three diverse O . sativa accessions from the rice diversity panel [42] , [67] ( Table S1 ) were evaluated for Al tolerance using an Al3+ activity of 160 µM in a hydroponic nutrient solution . This Al3+ activity had been previously determined to be optimal for evaluating a wide range of Al tolerance in diverse rice germplasm [8] . In the diversity panel , Al tolerance , measured as the relative root growth of the total root system ( TRG-RRG ) , was normally distributed around a mean of 0 . 59 +/−0 . 24 ( SD ) and ranged from 0 . 03–1 . 35 ( Figure 1A ) . Some varieties were inhibited by as much as 97% , while 16 varieties ( representing three subpopulations ) showed enhanced root growth in the presence of 160 µM Al3+ ( Table S1 ) . When accessions were grouped based on varietal group ( >80% ancestry ) the Japonica varietal group ( consisting of the temperate japonica , tropical japonica and aromatic subpopulations ) was significantly more Al tolerant than the Indica varietal group ( indica and aus subpopulations ) ( p<0 . 0001 ) ( Figure 1B ) . The Japonica varieties had a mean Al tolerance value of RRG = 0 . 72 , an interquartile range of 0 . 61–0 . 82 , and ranged from 0 . 13–1 . 35 . The Indica varieties had a mean Al tolerance value of RRG = 0 . 36 , an interquartile range of 0 . 27–0 . 43 , and ranged from 0 . 03–1 . 15 ( Figure 1B ) . Eleven accessions were classified as “admixed” between varietal groups , and these had a mean Al tolerance equal to the mean of all 372 accessions ( TRG-RRG = 0 . 59 ) with >80% ancestry to either varietal group . A one-way ANOVA demonstrated that subpopulation explained 57% of the phenotypic variation observed for Al tolerance ( TRG-RRG ) among the 274 accessions that carried a subpopulation classification . Despite the differences in mean TRG-RRG between subpopulations , considerable variation was also detected within each subpopulation ( Figure S1 ) . Two immortalized QTL mapping populations were analyzed for Al tolerance . One consisted of 134 recombinant inbred lines ( RIL ) derived from the cross IR64/Azucena [69] , and the other was comprised of 78 backcross inbred lines ( BIL ) derived from the cross Nipponbare/Kasalath//Nipponbare [70] . These populations were used to evaluate Al tolerance using three different indices of relative root growth ( RRG ) , ( 1 ) longest root growth ( LRG-RRG ) , ( 2 ) primary root growth ( PGR-RRG ) and ( 3 ) total root growth ( TRG-RRG ) ( see Materials and Methods for details ) . The phenotypic distribution was approximately normal for each population , no matter which root screening index was used ( illustrated for TRG-RRG in Figure S2A and S2B ) . The QTL mapping populations allowed us to determine which of the three root evaluation methods would be most useful for evaluating the diversity panel as a whole . The method of phenotyping , specifically , the RRG index used to estimate Al tolerance , directly impacted the significance of QTLs detected by composite interval mapping ( Figure 2A–2C and Figure S3A–S3C ) . In the RIL population , three Al tolerance ( Alt ) QTL were detected using total root growth ( the TRG-RRG index ) , AltTRG1 . 1 on chromosome 1 , AltTRG2 . 1 on chromosome 2 , and AltTRG12 . 1 on chromosome 12 ( Figure 2A–2C Table 1 ) . The Azucena allele conferred increased tolerance at the loci on chromosomes 1 and 12 and reduced tolerance at the locus on chromosome 2 . QTLs were detected in the same positions on chromosomes 1 and 12 using RRG based on primary root growth ( the PRG-RRG index ) , although with lower LOD scores ( Figure 2A–2C; Table 1 ) . Using longest root growth ( the LRG-RRG index ) , a single QTL was detected on chromosome 9 , AltLRG9 . 1 , and this QTL was not detected when the other root indices were used . The major QTL on chromosome 12 ( AltTRG12 . 1 ) , which explained >19% of the variation in Al tolerance based on TRG-RRG , is located between 2 . 69–5 . 10 Mb and encompasses the Al sensitive rice mutant art1 , which is located at 3 . 59 Mb [19] . In the BIL population , two QTL were detected using the TRG index , AltTRG1 . 2 on chromosome 1 , which co-localized with the AltTRG1 . 1 QTL identified in the RIL population , and AltTRG12 . 2 on chromosome 12 , which did not overlap with the AltTRG12 . 1 identified in the RIL population ( Figure 2A–2C , Figure S3A–S3C , Table 1 ) . The Nipponbare allele conferred tolerance at the chromosome 1 locus and the Kasalath allele conferred tolerance at the AltTRG12 . 2 locus . No QTLs were detected on chromosome 2 in the BIL population . Using the PRG-RRG index , one QTL was detected on chromosome 6 , where the Kasalath allele conferred resistance . No QTLs were detected using the LRG-RRG index in the BIL population . The Al tolerance index used for evaluating the phenotype directly affected both the identity and the significance of the QTLs detected . Al tolerance index-specific QTLs were detected in both populations and no QTL locus was detected across all three indices . Based on number of QTL detected , significance of QTL , and variance explained by the QTL , total root growth ( TRG ) proved to be the single most powerful Al tolerance index . However , rice QTLs detected using different evaluation methods are likely to confer Al tolerance by different mechanisms , such as tolerance of primary , secondary , lateral , or all roots , and thus they are complementary and together provide a robust evaluation of the genetic architecture of Al tolerance than any single index alone . To identify Al tolerance loci based on genome-wide association ( GWA ) mapping , we used an existing genotypic dataset consisting of 36 , 901 SNPs [65] , and the total root growth ( TRG-RRG ) Al tolerance phenotype generated on 373 O . sativa accessions over the course of this study . GWA mapping was conducted , using SNPs with a MAF>0 . 05 , across all 373 genotypes as well as independently within the indica , aus , temperate japonica , and tropical japonica subpopulations ( Figure 3 ) . The Efficient Mixed-Model Association ( EMMA ) [71] model was used in each analysis ( both within and across subpopulations ) to correct for confounding effects due to subpopulation structure and relatedness between individuals . As the subpopulation structure was highly correlated with Al tolerance , it was observed that analyzing all samples ( 373 ) together with the EMMA model resulted in an overcorrection ( causing type 2 error ) and a corresponding reduction in SNP significance ( Figure S4 ) . To address this problem , a PCA approach was also employed when analyzing all ( 373 ) samples together . However , the PCA approach resulted in a slight under-correction for population structure ( Figure S4 ) , demonstrating that results from each GWA method has limitations when used across all germplasm in this highly structured diversity panel . A total of ∼48 distinct Al tolerance genomic regions were identified by GWA mapping ( Figure 3 ) . Twenty-one regions were detected ( p<0 . 0001 ) across all ( 373 ) accessions using the PCA model ( Figure 3 ) , while only two SNPs were above the significance threshold when all ( 373 ) accessions were analyzed together using the EMMA model ( Figure 3 ) , both of which were also detected by PCA . The threshold of p<1 . 0E-04 was determined based on the upper-limit false discovery rate ( FDR ) , determined from the candidate genes in the same approach as in Li et al . [72] ( Table S2 ) . Thirty-two regions were significantly associated with Al tolerance in the indica subpopulation ( Figure 3 ) , including five regions that were also detected across all ( 373 ) samples using the PCA model . In the aus subpopulation , a single , highly significant , region was detected on chromosome 2 that was unique to this subpopulation and contained the Nrat1 candidate gene LOC_Os02g03900 ( Figure 3 ) . No significant SNPs ( MAF>0 . 05 ) were detected in the temperate japonica or tropical japonica subpopulations . The GWA mapping results indicate that the majority of significant loci are subpopulation-specific and that phenotypic variation for Al tolerance within given subpopulations is largely controlled by alleles that are unique to that subpopulation . SNPs identified by GWA were also compared to a set of 46 a priori candidate genes as well as to positions of QTL regions identified through bi-parental mapping ( this study and previous reports ) ( Table 1 and Figure 3 ) . Two regions of highly significant SNP clusters , one within the aus ( 8 SNPs; p = 2 . 8E-07 ) subpopulation on chr . 2 and one within the indica ( 32 SNPs; p = 2 . 9E-07 ) subpopulation on chr . 3 , co-localized to previously reported QTLs in populations in which an aus and indica parent served as the susceptible parents , respectively [17] , [23] . The list of 46 a-priori Al tolerance candidate genes ( Table 2 ) was compiled based on published information on Al sensitive mutants from rice and Arabidopsis [20]–[22] , [24] , cloned Al tolerance genes from wheat and sorghum [14] , [15] , expression profiles from Al treated maize and rice roots [19] , [73] , and an association study on specific candidate Al tolerance genes of maize [74] . Significant SNPs ( p<1 . 0E-04 ) within a 200 kb window of the a priori candidate genes were enriched 2 . 4 times compared to other SNPs ( p>0 . 0001 ) outside of the a priori and QTL regions . The 200 kb window was selected to fall within the estimated window of LD decay in rice ( ∼50–500 kb [45]–[49] and the upper-limit false discovery rate for the a priori genes was 42% . In addition , four of the 46 gene candidates ( ∼9% ) were located within a 200 kb window enriched for GWA SNPs in this study ( Figure 3 and Table 2 ) . One of the candidate genes ( Nrat1 ) on chr . 2 , co-localized with both GWA SNPs and a previously reported QTL ( Figure 3 ) . The relationship between the four candidates that co-localized with GWA SNPs are discussed in order of their positions on the rice genome below . A cluster of eight highly significant SNPs ( p-values = 2 . 3×10−5–2 . 8×10−7 ) on chromosome 2 between 1 . 536 Mb–1 . 675 Mb was associated with Al tolerance within the aus subpopulation ( Figure 3 and Table 2 ) . Previously , a QTL had been reported in the same location ( 0 . 536–1 . 9 Mb ) where the susceptible parent was of aus origin [26] . The LD decay in the aus subpopulation at this region was calculated to be 150 kb and a strong candidate gene was identified within the target region . The gene ( LOC_Os02g03900 located at 1 . 66 Mb ) encodes a Nramp6 metal transporter and was demonstrated to have altered expression patterns in Al-treated roots of the Al sensitive art1 rice mutant [19] . This Nramp6 metal transporter was recently reported as Nrat1 , a plasma membrane-located transporter for Al with enhanced sensitivity to Al in the knockout mutant [20] . As was the case with the ART1 gene itself , the Nrat1 metal transporter has not been associated with natural variation for Al tolerance prior to this study . On chromosome 5 , a significant region was detected across all samples ( 373 genotypes ) by PCA , co-localizing with the STAR2 gene ( LOC_Os05g02750 ) ( Figure 3 and Table 2 ) . The LD decay across this region was estimated at >500 kb , and encompassed two significant regions detected across all samples ( PCA ) , one of which was also detected within the indica subpopulation . STAR2 is the rice ortholog of the Arabidopsis Al sensitive mutant als3 [21] . It encodes the transmembrane domain of a bacterial-type ATP binding cassette ( ABC ) transporter and the star2 mutant is Al sensitive [24] . STAR2 was also found to be part of a gene network showing altered expression in response to Al in the art1 mutant compared to the ART1 wild type [19] . This study provides the first evidence that there may be natural variation for Al tolerance in rice at the STAR2 locus; however it is important to recognize that the PCA approach may under-correct for the effect of subpopulation in this study , thus it will be necessary to confirm the effect of the STAR2 alleles identified in this diversity panel . A significant GWAS region identified in the indica subpopulation on chromosome 7 co-localized with LOC_Os07g34520 , a rice ortholog of a maize isocitrate lyase a priori candidate gene associated with Al tolerance in maize [73] , [74] . The LD decay across this region within the indica subpopulation was 250 kb . Three highly significant regions detected within indica were further investigated to identify whether any clear Al tolerance candidate genes were located within these SNP clusters . The first region was a cluster of 32 significant SNPs ( p = 3 . 0E-7 ) between 28 . 782–27 . 863 Mb on chr . 3 that co-localized with a previously reported QTL ( Nguyen et al . , 2002 ) . Two clear candidates were identified among the 13 genes in this cluster; a nucleobase-ascorbate transporter ( LOC_Os03g48810 ) and a chloride channel protein ( LOC_Os03g48940 ) . The second region was a 10 SNP cluster ( p = 9 . 3E-12 ) between 26 . 986–27 . 479 Mb on chr . 7 . Of the 80 genes in this region , 34 of which were retrotransposons , there were three strong candidate genes; a glycosyl transferase protein ( LOC_Os07g45260 ) , a cytochrome P450 protein ( LOC_Os07g45290 ) and a zing finger RING type protein ( LOC_Os07g45350 ) . This region on chr . 7 was also identified in the introgression analysis as a localized introgressed region from Japonica into the highly tolerant Indica outliers ( discussed below ) . The third region was an 8 SNP cluster between 4 . 892–5 . 164 Mb on chr . 11 . Among the 48 genes in this region , there were two major classes of candidate genes observed , including 12 F-box proteins and a zinc finger CCHC protein . We chose to further investigate the variation in and around the Nrat1 gene on chromosome 2 because multiple independent lines of evidence supported the existence of a gene ( s ) in this region responsible for a significant portion of the variation for Al tolerance in rice . Evidence included a strong GWA peak in the aus subpopulation , a previously reported QTL [26] , and the localization of the Nrat1 Al transporter gene . Using the 44 K SNP data , LD in this region was calculated to be ∼150 kb in the aus subpopulation and 11 distinct haplotypes were observed in the entire diversity panel across a 139 kb region around the Nrat1 gene ( 1 . 536 Mb–1 . 675 Mb on chr . 2 ) ( Figure 4A ) . Haplotype 1 ( Hap . 1 ) , which was unique to the aus subpopulation , was found in 8 Al sensitive aus accessions and one Al sensitive aus/indica admixed line . These 9 genotypes were among the least Al tolerant ( 7th percentile , mean RRG = 0 . 16 ) of the 373 accessions screened ( Table S1 ) . Haplotype 1 explained 40% of the phenotypic variation for Al tolerance within the aus subpopulation ( Figure S5 ) . In addition , four aus accessions that were highly or moderately Al tolerant were found to contain a tropical japonica introgression across this region ( described in the section on Introgression analysis below ) . Haplotype 2 ( Hap . 2 ) was found in one aus and one indica accession , and was most similar to Hap . 1 , differing at only 2/14 SNPs ( Figure 4A ) . The two lines containing haplotype 2 had very different levels of Al tolerance; the aus variety , Kasalath ( ID 85 ) , was highly susceptible , with a RRG = 0 . 2 , while the indica variety , Taducan ( ID 163 ) , was tolerant , with a RRG = 0 . 8 , suggesting that this extensive 14-SNP haplotype across the 139 kb region was not predictive of Al tolerance . However , when the haplotype was built using only the four SNPs immediately flanking the Nrat1 gene , a group of 16 accessions sharing the same haplotype at these four SNPs was clearly identified . These 16 accessions , included the 10 susceptible aus accessions ( including one aus/indica admixed line ) carrying haplotype 1 and haplotype 2 and six indica accessions ( of varying Al tolerance ) carrying haplotype 2 and haplotype 3 ( Figure 4A ) . To determine if the four-SNP haplotype flanking the Nrat1 gene could be further resolved , we focused more deeply on the Nrat1 gene itself . We sequenced all 13 exons ( including introns ) of Nrat1 ( 1874 bp ) in 26 susceptible and tolerant varieties representing the aus , indica , tropical japonica and temperate japonica subpopulations ( Figure 4B ) . The accessions carried haplotypes 1 , 2 , 3 , 6 and 11 , as described in Figure 4A; where haplotype 1 was aus-specific and corresponded to the most sensitive group of accessions in the diversity panel; haplotype 2 was found in phenotypically divergent aus and indica accessions as described above; haplotype 3 was found in moderately tolerant indica varieties; haplotype 6 , which appeared to be the ancestral haplotype , was the most common haplotype in all subpopulations and was associated with moderately high levels of tolerance; and haplotype 11 , which was found in a majority of tropical japonica varieties , all of which were Al tolerant . Based on the 22 SNPs and/or indels identified across the 1 , 874 bp of Nrat1 sequence , highly resolved , gene haplotypes were constructed ( Figure 4B ) . The gene haplotypes corresponded fairly well to the extended haplotype groups that had been constructed using the data from the 44 K SNP chip , except in the case of haplotype 2 , where varieties differed at 10/22 ( 45% ) of the SNPs across the Nrat1 gene . This fully resolved haplotype at the Nrat1 gene resulted in the susceptible Kasalath clustering with the other highly susceptible aus varieties and the tolerant Taducan clustering with other highly tolerant varieties ( Figure 4 ) . Three non-synonymous SNPs ( polymorphisms 4 , 16 , 17 ) were shared among the 9 highly susceptible aus accessions . When the Eukaryotic Linear Motif resource ( http://elm . eu . org ) was used to identify functional sites in the Nrat1 gene , polymorphism 16 was identified as a functional site where a C→T SNP caused an amino acid change from valine→alanine ( amino acid 500 ) . This protein site was predicted to be involved in PKA-type AGC kinase phosphorylation , with the functional site spanning amino acids 497–503 . Thus , polymorphism 16 was identified as a strong functional polymorphism candidate underlying natural variation in Nrat1 . The fact that polymorphism 16 was also observed in two Al tolerant temperate japonica and one moderately tolerant tropical japonica accession ( haplotype 11 ) suggested that SNP 16 alone was not predictive of Al tolerance . However , a combination of polymorphisms 4 , 16 , and 17 was entirely predictive of Al susceptibility . This study demonstrates the power of whole genome association analysis to integrate divergent pieces of evidence from independent bi-parental and mutant studies , enabling us to associate gene-based diversity with germplasm resources and natural variation that is of immediate use to plant breeders . There is a clear difference in the degree of Al tolerance found in the Japonica varietal group and the Indica varietal group , with the 10th percentile of Al tolerance of Japonica ( 0 . 53 ) being nearly equal to the 90th percentile of Indica ( 0 . 55 ) ( Figure 1B ) . However , there are clear outliers within each varietal group . Five Indica accessions are highly Al tolerant ( ID 30 , 66 , 142 , 163 , 337 ) , ranging from 2 . 1–3 . 2 times the mean Indica Al tolerance , and three Japonica accessions ( ID 12 , 52 , 112 ) are highly susceptible , each approximately 0 . 19 of the mean Japonica Al tolerance ( Figure 1B and Table S1 ) . To determine if these outliers were the result of introgressions across varietal groups , we calculated the allele ancestry of 5 , 467 SNPs distributed throughout the genome and identified specific genomic regions where historical Indica×Japonica admixture was detected only in the respective Indica or Japonica outlier lines . To do this , Japonica introgressions identified in highly Al tolerant Indica lines were used to query all other Indica accessions and only those Japonica introgressions that were uniquely present in the highly Al tolerant outlier Indica lines were considered as candidate regions underlying the outlier phenotype . When the five Indica outliers were used for this analysis , a few , well-defined regions comprising 2 . 4–4 . 9% of the genome corresponded to regions of Japonica introgression ( Table 3 ) . In the case of the three highly Al susceptible Japonica varieties , the genetic background was highly heterogeneous and the small number of lines precluded doing any admixture analysis . Therefore , the admixture analysis was conducted only on the five highly tolerant Indica outliers . In the five outlier Indica accessions , 6 Japonica introgressions ( median size = 780 kb ) were identified that were specific only to these 5 lines . Three of these introgressions were present in two genotypes , two of the introgressions were present in three genotypes , and one introgression was present in four of the outliers ( Table 3 ) . Three introgressions encompass SNPs identified by GWA analysis and two co-localized with bi-parental QTL . The introgression that was present in four of the indica outlier genotypes was located on chromosome 7 between 27 . 05–28 . 62 Mb and contained 94 annotated genes . This introgression included a cluster of GWA SNPs that were highly significant within the indica subpopulation ( p = 2 . 6×10−5 , MAF = 0 . 10 ) and was one of the top 100 most significant SNPs identified when the diversity panel as a whole was analyzed .
In this study , we utilized bi-parental QTL mapping and GWA analysis to examine the genetic architecture of Al tolerance in rice and to identify Al tolerance loci . Phenotyping of the diversity panel provided valuable information about the range and distribution of Al tolerance in O . sativa and offered new insights into the evolution of the trait . The mean Al tolerance in Japonica was twice that of Indica ( p<0 . 0001 ) , and 57% of the phenotypic variation was explained by subpopulation . The relative degree of Al tolerance in the five subpopulations ( temperate japonica>tropical japonica>aromatic>indica = aus ) was consistent with the level of genetic relatedness among them [42] , [44] and suggests that temperate and tropical japonica germplasm contain alleles that would be useful sources of genetic variation for enhancing levels of Al tolerance within indica and aus . This is supported by the identification of highly tolerant indica varieties from the rice diversity panel that contain introgressions from Japonica in regions characterized by GWA peaks . The highly tolerant Indica outliers demonstrate the feasibility of using a targeted approach to increase Al tolerance in Indica varieties by introgressing genes from Japonica . While less obvious , our QTL analysis demonstrated the ability to increase Al tolerance in Japonica using targeted introgressions from Indica . This was demonstrated within both QTL populations by the identification of two loci in which alleles from the highly susceptible Kasalath parent conferred enhanced levels of Al tolerance in the Nipponbare genome ( temperate japonica ) and one locus where the moderately susceptible IR64 parent conferred enhanced tolerance in crosses with Azucena ( tropical japonica ) ( Table 1 ) . To date , only a few indica and aus accessions have been used in QTL mapping populations and the identification of a large number of GWA loci in indica , coupled with the fact that indica is significantly more diverse than all other O . sativa subpopulations [40] , [42] suggests that there are likely to be many novel alleles that could be mined from the indica subpopulation . Further evidence of the value of this approach in the context of plant breeding comes from the transgressive variation observed in both QTL populations , where some RILs and BILs exceeded the Al tolerance observed in the tolerant tropical and temperate japonica parents , Azucena and Nipponbare , respectively , due to alleles derived from the susceptible indica ( IR64 ) or aus ( Kasalath ) parents , respectively . The significant differences in Al tolerance among varietal groups and subpopulations , and evidence that different genes and/or alleles contribute to Al tolerance within the major varietal groups , is consistent with Indica and Japonica domestication from pre-differentiated , wild O . rufipogon gene pools that differed in Al tolerance . Future experiments will test this hypothesis by comparing levels of Al tolerance found in wild populations of O . rufipogon . The inherently higher levels of Al tolerance found in the Japonica varietal group may help explain why tropical japonica varieties are so often found in the acid soils of upland environments . Compared to QTL mapping , GWA significantly increases the range of natural variation that can be surveyed in a single experiment and the number of significant regions that are likely to be identified . Furthermore , GWA provides higher resolution than QTL mapping , facilitating fine-mapping and gene discovery . This was illustrated by the two highly significant regions detected by GWA that overlapped with previously reported QTLs . GWA detected a highly significant cluster of 32 SNPs ( p = 2 . 9E-07 ) on chr . 3 within the indica subpopulation , defining the candidate region to 81 kb window containing 13 genes , while the previously reported QTL interval was 1 , 720 kb [17] , containing 260 genes . Similarly , the Nrat1 locus identified within the aus subpopulation on chromosome 2 initially narrowed the target region to 139 kb containing 27 genes by GWA , while the previously reported [26] QTL interval was 1 , 360 kb and contained 234 genes . Surprising , the Nrat1 region was not significant in the BIL population , in which the resistant parent ( Nipponbare ) contained a resistant haplotype at Nrat1 and the susceptible parent ( Kasalath ) contained the susceptible haplotype at Nrat1 . The fact that a significant signal was not detected in the BIL population can likely be explained by one or more of the following: 1 ) the bias inherent in the small population size ( 78 BILs ) , 2 ) the backcross population structure in which only 11 individuals ( 14% of BILs ) contained the Kasalath allele at the Nrat1 locus and/or 3 ) the effects of genetic background on the Nrat1 QTL region . The Nrat1 QTL region was detected in one previous QTL study by Ma et al . [23] where a BIL population consisting of 183 lines was used , with Kasalath as the susceptible aus parent and Koshihikari as the tolerant temperate japonica parent [23] . In that study , the Nrat1 QTL region was of minor significance ( LOD = 2 . 81; R2 = 7% ) , and it is noteworthy that the two other ( more significant ) QTLs detected in that study were the two QTLs detected in our BIL population using only 78 lines . The fact that the Nrat1 QTL region was not detected in our BIL mapping population and was of low significance in the Ma et al . QTL study suggests that the effect of the Kasalath allele is likely to be influenced by genetic background effects ( GXG ) . In an aus genetic background , the Nrat1 susceptible haplotype explains 40% of the phenotypic variation , and the diversity panel contains enough aus varieties for this to be statistically significant using GWA; however , in the BIL population where Nipponbare served as the recurrent parent , the aus alleles exist in a largely temperate japonica background . Given the extent of GXG observed in inter-sub-population crosses , and the small size of our BIL population , this appears to be the most likely explanation as to why the Nrat locus was not detected in our QTL experiment . Although GWA significantly increased the power and resolution of QTL detection , nearly all the significant loci detected were subpopulation-specific . This is entirely consistent with the strong subpopulation structure in rice and the high correlation of Al tolerance with subpopulation , justifying our GWA analysis on each subpopulation independently . So the question might be asked as to why it is also necessary to conduct GWA in the diversity panel as a whole ? The answer to this question lies in the complex biology and demographic or breeding history of O . sativa . In this study GWA was conducted both within and across subpopulations , and it demonstrated that GWA on the diversity panel as a whole leveraged power to detect alleles that were segregating across multiple subpopulations , even if they were rare within any one subpopulation group , while when used on independent subpopulations , it was useful in detecting alleles that segregated only within one or two subpopulations but tended to be fixed in others . This is what would be expected from what we know about the evolutionary history of rice with its examples of shared domestication alleles [35] , [75] coupled with myriad subpopulation-specific alleles [41] , [48] , [76]–[78] that provide each subpopulation with its specific identity and spectrum of ecological adaptations . There are cases in which QTLs discovered by bi-parental mapping are not detected by GWA analysis . One reason for this is that QTL mapping can readily detect alleles that are rare in a diversity panel , are subpopulation-specific , or where the phase of the allelic association differs across subpopulations , while GWA analysis has limited power to do so . This is important in the case of rice , because of the degree of differentiation between the subpopulations and the significant evolutionary differences between the Indica and Japonica varietal groups , as discussed above . Thus , while variation that is strongly correlated with subpopulation structure is undetectable by GWA analysis , these loci can be easily detected by QTL analysis if crosses between sub-populations are used . This is illustrated by the identification of the Al tolerance QTL , ( AltTRG12 . 1 ) encompassing the ART1 locus on chromosome 12 . This large-effect QTL ( LOD = 7 . 85 , R2 = 0 . 193 ) was clearly detected in the RIL population but was not detected by GWA analysis . The QTL mapping populations utilized in this study were of limited population size and thus largely underpowered [79] . As a result it is likely that some QTL effects were overestimated and that other small effect QTL were not detected . Although we cannot be certain of the exact amount of variance explained by a particular QTL , it is reasonable to conclude that the major QTL detected ( AltTRG12 . 1 ) is , in fact , the most significant QTL in the population . GWA mapping also provides a valuable link between functional genomics and natural variation , and in the case of rice , highlights the subpopulation-specific distribution of specific alleles and phenotypes . We implicate the involvement of the STAR2 ( chr . 6 ) /ALS3 ( Arabidopsis Al sensitive mutant ) gene , previously identified as induced mutations in rice and Arabidopsis , respectively [22] , [23] , and document the detection of highly resolved , novel Al tolerance loci in the indica and aus subpopulations . This is a critical bridge for germplasm managers and plant breeders who look for alleles of interest in germplasm collections rather than as sequences in GenBank . Our strongest example of the value of linking functional genomics and natural variation is illustrated by the GWA region on chromosome 2 , where we demonstrate that the aus-specific susceptible haplotype in this region is functionally related to an Nramp gene . This gene was previously identified to have altered expression in the art1 ( transcription factor ) Al sensitive mutant [19] and was recently reported as Nrat1 ( for Nramp aluminum transporter ) , an Al transporter localized to the plasma membrane of root cells , which when knocked out , enhances Al susceptibility . This is consistent with this transporter serving to mediate Al uptake by moving it directly into root cells , presumably into the vacuole , and away from the root cell wall [20] . Our haplotype analysis of the GWA region on chromosome 2 and sequence analysis of the Nrat1 gene identified putative sensitive and tolerant haplotypes that implicate the Nrat1 gene , and further identified two putative functional polymorphisms specific to the Al sensitive aus accessions . These data provides valuable information for identifying Nrat1 alleles that can be used to test the hypothesis put forth by Xia et al . [20] , namely that Al tolerance is conferred by reducing Al concentrations in the cell wall . It will be interesting to see if the sensitive alleles of this gene encode an Nramp transporter that is less effective at mediating Al uptake . Furthermore , the observation that three of the four most Al tolerant aus accessions contain tropical japonica introgressions across this gene region strongly suggests that Al tolerance of aus genotypes can be increased by the targeted introgression of tropical japonica DNA at the Nrat 1 region . One of the objectives of this study was to determine if the Al tolerance index employed ( longest root growth [LRG] , primary root growth [PRG] , or total root growth [TRG] ) affected the detection and/or significance of Al tolerance QTL . In a recent publication from our research team , it was demonstrated that significantly different Al tolerance scores were obtained with the different indices [8] . In all previous QTL studies , Al tolerance was determined based on relative root growth ( RRG ) of the longest root . This study demonstrated that the Al tolerance index has a direct effect on the detection and significance of QTLs . Total root growth ( TRG ) was the single most powerful Al tolerance index , based on number of QTL detected , significance of QTL and variance explained by the QTL . However , it is relevant to point out that LRG-RRG identified a large-effect QTL ( AltLRG9 . 1 ) in the RIL population that was not detected using any other index , and PRG-RRG identified a unique QTL on chromosome 6 where the susceptible Kasalath variety carried the resistance allele . These observations suggest that different root evaluation methods are likely to identify Al tolerance QTLs that confer tolerance mediated by different types of roots , or possibly by different patterns of gene expression detectable only when specific phenotypic evaluation protocols are used . The strongest example of the importance of utilizing the TRG-RRG index is demonstrated by the identification of the AltTRG12 . 1 QTL in the RIL mapping population . The ART1 gene , a C2H2-type zinc finger-type transcription factor that causes Al hypersensitivity when mutated , is located close to the center of the Alt12 . 1 QTL peak . When this gene was first identified , it was suggested that it was not involved in natural variation of Al tolerance in rice , as no QTL had ever been identified in the region [19] . Based on our results , it is likely that this QTL was not previously identified because relative root growth was measured only based on LRG , rather than on TRG-RRG . Further fine-mapping of this locus , along with sequence and expression analysis , is underway to determine whether the ART1 locus underlies this QTL and to understand the mechanism by which it contributes to natural variation for Al tolerance . Previous studies in other cereals have reported that the correlation of Al tolerance between hydroponics and field conditions is >70% [80] and studies on rice Al tolerance mutants have demonstrated that tolerance/susceptibility observed in hydroponics screens is also observed under soil conditions [24] . To accurately assess the value of the loci detected in this study as targets of selection in rice breeding programs , we are currently developing experiments to determine the effect of the key loci detected in this work under Al-toxic field conditions . Furthermore , four sets of reciprocal NILs ( 8 NILs total ) for the four QTLs detected in the RIL population are being developed to determine the effect of each QTL under both hydroponic and field conditions . Finally , field experiments will be conducted to determine which hydroponic root measurement phenotype ( TRG , PRG , or LRG ) is the best for predicting a genotypes Al tolerance under field conditions . This study provides the most comprehensive analysis of the genetic architecture of Al tolerance in rice to date . It demonstrates the power of whole genome association analysis to identify phenotype-genotype relationships and to integrate disparate pieces of evidence from QTL studies , mutant analysis , and candidate gene evaluation into a coherent set of hypotheses about the genes and genomic regions underlying quantitative variation . By tracing the origin of Al tolerance alleles within and between rice subpopulations , we provide new insights into the evolution and combinatorial potential of different alleles that will be invaluable in breeding new varieties for acid soil environments . This work demonstrates how genetic and phenotypic diversity is partitioned by subpopulation in O . sativa and provides support for the hypothesis that the most efficient approach to enhancing many quantitative traits in rice is to selectively introgress genes/alleles from one subpopulation into another . Our study also lays the foundation for understanding the genetic basis of Al tolerance mechanisms that enable rice to withstand significantly higher levels of Al than do other cereals . It not only facilitates more efficient selection of tolerant genotypes of rice , but it points the way toward using this knowledge to enhance levels of Al tolerance in other plant species .
Plants were grown hydroponically in a growth chamber as described by Famoso et al . [8] . Al tolerance was determined based on relative root growth ( RRG ) after three days in Al ( 160 µM Al3+ ) or control solution . The hydroponic solution used in this study was chemically designed and optimized for rice Al tolerance screening; for a detailed comparison of the phenotypic procedures employed in this work compared to previously published rice Al tolerance work see Famoso et al . ( 2010 ) . To obtain uniform seedlings , 80 seeds were germinated and the 30 most uniform seedlings were visually selected and transferred to a control hydroponic solution for a 24 hour adjustment period . After the 24 hour adjustment period , root length was measured with a ruler and the 20 most uniform seedlings were selected and distributed to fresh control solution ( 0 uM Al3+ ) or Al treatment solution ( 160 uM Al3+ ) . Plants were grown in their respective treatments for ∼72 hours and the total root system growth was quantified using an imaging and root quantification system as described by Famoso et al . ( 2010 ) . The mean total root growth was calculated for Al treated and control plants and RRG was calculated as mean growth ( Al ) /mean growth ( control ) . The 373 genotypes screened for Al tolerance and used in the association analysis are part of a set of 400 O . sativa genotypes that have been genotyped with 44 , 000 SNPs as described by Zhao et al . [65] . The QTL populations consisted of a population of 134 recombinant inbred lines ( RILs ) derived from a cross between Azucena ( tolerant tropical japonica ) and IR64 ( susceptible indica ) [67] , [70] and a population of 78 backcross introgression lines ( BILs ) derived from a cross between Nipponbare ( tolerant temperate japonica ) and Kasalath ( susceptible aus ) and backcrossed to Nipponbare . The Al3+ activity at which Al tolerance was screened was determined by identifying the Al3+ activity that provided the greatest difference in tolerance between the parents . The tolerant parent of the RIL population , Azucena , and the tolerant parent of the BIL population , Nipponbare , are similar in Al tolerance , whereas the susceptible parent of the RIL population , IR64 , is significantly more tolerant than the susceptible parent of the BIL population , Kasalath ( Figure 1A ) . To ensure that a normal distribution was obtained in each population , a different Al3+ concentration was used for each mapping population . The RIL population was screened at 250 µM Al3+ because the Azucena parent is very Al tolerant and the IR64 parent is only moderately susceptible . The BIL population was screened at 120 µM Al3+ because the Kasalath parent is extremely Al sensitive , though the Nipponbare parent is very Al tolerant . Figure 1 displays the Al tolerance of each mapping parent in reference to the 373 genetically diverse rice accessions screened at 160 µM Al3+ . The genetic component of the phenotypic variance was calculated as VarG = VarG+Var ( GxE ) +error . QTL analysis was conducted using composite interval mapping ( CIM ) function in QTL Cartographer [81] . The significance threshold was determined by 1000 permutations . Genome-Wide Association Analysis was performed using three approaches in all samples ( 373 ) with phenotypes . The first approach was the naïve approach , which is simply the linear regression of phenotype on the genotype for each SNP marker . The second approach was principle component analysis ( PCA ) , where we obtained the four main PCs ( principle components ) that reflect the global main subpopulations in the sample to correct population structure estimated from software EIGENSOFT . [82] . The first four PCs are included as cofactors in the regression model to correct population structure: . Here β and γ are coefficient vectors for SNP effects and subpopulation PCs respectively . and are the corresponding SNP vector and first 4 PC vectors , and is the random error term . The third approach was the linear mixed model proposed by [62] , [63] , implemented in the R package EMMA [71] , which models the different levels of population structure and relatedness . The model can be written in a matrix form as: y = Xβ+Cγ+Zμ+e where β and γ are the same as above , both of which are fixed effects , and is the random effect accounting for structures and relatedness , is corresponding design matrices , and is the random error term . Assume μ∼N ( 0 , σ2gK ) and e∼N ( 0 , σ2eI ) , and K is the IBS matrix , as in [62] . We also conducted GWA using both the naïve approach and the mixed model approach in each of the four main subpopulations ( IND , AUS , TEJ , TRJ ) . For the mixed model , the model was changed to y = Xβ+Zu+e , since there was no main subpopulation division within each subpopulation sample . Linkage disequilibrium decay and haploblocks were calculated at specific chromosome/gene regions using Haploview software [83] . Population structure was analyzed employing Expectation-Maximization techniques on an HMM model of per-marker ancestry along a chromosome with a weak linkage model between adjacent markers on the same chromosome induced by the HMM's state dependence on the previous marker's subpopulation assignment ( M . Wright , Cornell University , personal communication ) . The 5 , 467 SNPs used for admixture analysis were a subset of the 36 , 901 high quality SNPs on the 44 K chip , and were selected based on their information content and ability to distinguish genetic groups , rather than individuals . The two main criteria used to select the subset of SNPs were a ) good genomic distribution and minimal LD among those used in the analysis , and b ) MAF>0 . 05 in at least one subpopulation . The state of the HMM at each marker corresponds to the subpopulation of origin for the marker ( and by extension , the region containing the marker and its adjacent markers ) . The number of a priori distinct subpopulations was K = 5 , consistent with that reported previously by Garris et al . 2005 and Ali et al . , 2011 [40] , [66] . A set of 50 standard non-admixed “control” lines , 10 representing each of the Garris et al . subpopulations , that were genotyped on the 44 K rice SNP array were used to develop and evaluate the method . All 50 lines were correctly assigned to each of the subpopulations and concordant with previous results using STRUCTURE [84] , with little or no admixture or introgressions detected . The EM/HMM method was favored over the corresponding “linkage model” of recent versions of STRUCTURE because the EM/HMM model explicitly modeled inbreeding and estimated the inbreeding coefficient for each line independently , permitting lines in various stages of purification or inbreeding to homozygosity to be analyzed . The lines phenotyped in this study that were also genotyped on the 44 K SNP array were then analyzed , combined with these 50 control lines and the local ancestry along chromosomes were assigned by maximizing the state path of the HMM while simultaneously estimating subpopulation specific allele frequencies using the forward-backward algorithm . Using this method , introgressions from a foreign subpopulation into a line with a vast majority of the genetic background originating from a single subpopulation were detected .
|
While rice ( Oryza sativa ) is significantly more Al tolerant than other cereals , no genes underlying Al tolerance in rice have been reported . Using genome-wide association ( GWA ) and bi-parental QTL mapping , we investigated the genetic architecture of Al tolerance in rice . Japonica varieties were twice as Al tolerant as indica and aus varieties . Overall , 57% of the phenotypic variation was correlated with subpopulation , consistent with observations that different genes and genomic regions were associated with Al tolerance in different subpopulations . Four regions identified by GWA co-localized with a priori candidate genes , and two highly significant regions co-localized with previously identified quantitative trait loci ( QTL ) . Haplotype and sequence analysis around the candidate gene , Nrat1 , identified a susceptible haplotype explaining 40% of the Al tolerance variation within the aus subpopulation and three non-synonymous mutations within Nrat1 that were predictive of Al sensitivity . Using Indica × Japonica mapping populations , we identified QTLs associated with transgressive variation where alleles from a susceptible indica or aus parent enhanced Al tolerance in a tolerant japonica background . This work demonstrates the importance of subpopulation in interpreting and manipulating complex traits in rice and provides a roadmap for breeders aiming to capture genetic value from phenotypically inferior lines .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biotechnology",
"marker-assisted",
"selection",
"agricultural",
"biotechnology",
"agriculture"
] |
2011
|
Genetic Architecture of Aluminum Tolerance in Rice (Oryza sativa) Determined through Genome-Wide Association Analysis and QTL Mapping
|
Combination therapies are often needed for effective clinical outcomes in the management of complex diseases , but presently they are generally based on empirical clinical experience . Here we suggest a novel application of search algorithms—originally developed for digital communication—modified to optimize combinations of therapeutic interventions . In biological experiments measuring the restoration of the decline with age in heart function and exercise capacity in Drosophila melanogaster , we found that search algorithms correctly identified optimal combinations of four drugs using only one-third of the tests performed in a fully factorial search . In experiments identifying combinations of three doses of up to six drugs for selective killing of human cancer cells , search algorithms resulted in a highly significant enrichment of selective combinations compared with random searches . In simulations using a network model of cell death , we found that the search algorithms identified the optimal combinations of 6–9 interventions in 80–90% of tests , compared with 15–30% for an equivalent random search . These findings suggest that modified search algorithms from information theory have the potential to enhance the discovery of novel therapeutic drug combinations . This report also helps to frame a biomedical problem that will benefit from an interdisciplinary effort and suggests a general strategy for its solution .
To understand the motivation for our work it is important to consider that , even if simulations might play a role , the intended use of the algorithms is not entirely in silico , but partially in vivo or in vitro , using high-throughput biological measurements in organisms or isolated cells , respectively . This approach becomes increasingly relevant because high-throughput measurement technology , initially developed by drug companies for the screening of large libraries of compounds in multi-well plate formats , is now more and more available to the scientific community . It is useful to regard the information processing by our experimental systems as parallel biological computations , since the algorithms we are using are indeed derived from algorithms that were implemented in silico in other scientific fields . Parallel measurements are suitable for multi-well high-throughput technology . There are requirements regarding the computational complexity of the algorithms that limit the choice of suitable approaches . These requirements are discussed in more detail in the Results . Both the number of operations and computational costs unique to in vivo/in vitro algorithms should be considered . Algorithm design requires the application of an appropriate structure to the data . Although there are many options to represent the space of possible drug combinations , we used a tree representation with drug combinations as nodes linking to all possible additions of one drug in the next level . Individual drugs form the base of the tree and combinations of maximum size are at the top ( see Algorithms section in the Results ) . When exploring the drug combination tree going from smaller to larger combinations , as in the algorithms we suggest , we are giving more weight to lower-order drug interactions . This is consistent with data available on adverse drug interactions , which are reported mostly for two-drug combinations [6] , [7] . Estimating the optimal size of a combination is a different problem , examined in detail in the Discussion . The beneficial effect of a combination is also due to additive components ( not depending on interactions ) and to multiple higher-order effects . The search algorithms we suggest are derived from sequential decoding algorithms . These were chosen in part because of similarities among the data trees to be searched in the biological and decoding applications ( see again the Algorithms section in the Results ) . Sequential decoding algorithms are used for convolutional codes , in which nearby nodes in the data tree are related , similarly to different but partially overlapping combinations of drugs . Another feature of sequential algorithms that fit our purposes is the use of a list-based memory of the path taken to reach each node . We provide in the Discussion a detailed argument suggesting that a suitable algorithm should be able to integrate all available information on the state of the system with that obtained by iterative measurements . The integration should take place at every iteration within the algorithm , rather than being a weighted average of different methods applied separately . The presence of the updated list as a guide for each iteration provides our algorithms with a natural mean of information integration . Both the fully factorial dataset we show in Figure 1 and the complex structure of the biological networks that are being reconstructed in systems biology supports this expectation of frequent non-linearities in phenotype measurements along the data tree . Therefore we are interested in algorithms that can search within a solution space presenting substantial non-linearities . If the relation among drugs in a combination were linear , the best algorithm would simply determine the best dose in single drug measurements and use these to obtain the best combination . If , on the contrary , non-linearities were extreme , the use of stochastic algorithms might be preferable . Stochastic algorithms ( see also Discussion ) can cope with multiple local minima in the solution space , but they do so by incorporating a random element . This requires a price in terms of computational cost , and the performance of stochastic algorithms is therefore often not as good as that of more tailored algorithms [8] , [9] . The algorithms we suggest can cope with moderate and variable non-linearities by going back to previous nodes in the tree . Starting with the stack sequential algorithm , which was developed to search for optimal decoding in the field of digital communications [10] , we describe and test algorithms that can be used to search for an optimal combination of a sizeable number of drugs , by testing only a small subset of all possible combinations . The algorithms are useful for large combinations , where collecting fully factorial datasets is not feasible . We present results obtained from simulations in a computational model of cell death and from experiments using two models with complementary biological properties: ( i ) restoring the decline with age in heart function and exercise capacity in Drosophila melanogaster;and ( ii ) selective killing of human cancer cells . The first in vivo experimental model has the advantage of including the complexity of whole organism interventions , while the second in vitro model has the potential for markedly higher throughput testing . These models are also representative of two different general types of multi-drug interventions: one type aims at improving function , while the other is based on the induction of cell death , a selective disruption of network function . Results suggest that optimal or near-optimal combinations of compounds can be found in these systems with only a small fraction of the number of tests as a fully factorial design , and with significantly higher efficacy than random searching . In summary the contributions of this work are:
A fully factorial dataset is a dataset where all possible combinations of drugs for the selected doses are tested . The dataset was obtained from biological measurements in a living organism , Drosophila melanogaster ( the fruitfly ) . A detailed account of the Drosophila cardiac aging model was presented previously [11] . We performed an initial screen of compounds for their effects on cardiac aging in Drosophila , selected for their general effects on multiple biological functions , previously demonstrated low toxicity and , for some compounds , known effects on aging in other models . After screening 44 compounds individually at multiple doses ( a total of 300 groups , each composed of 10–20 flies ) , we chose two doses each of four compounds for more comprehensive measurements of their combined effects on three age-related phenotypes: the declines in maximal heart rate , exercise capacity and survival . The selected compounds ( see Methods for doses in the fly food ) were: doxycycline , a broad spectrum antibiotic and inhibitor of mitochondrial protein synthesis [12]; sodium selenite , an essential trace mineral and cofactor of many metabolic enzymes; zinc sulfate , another trace mineral and cofactor of many metabolic enzymes; and resveratrol , a phenolic antioxidant with an action on proteins linked to aging [13] . The compounds were fed to flies from the age of 7 days to the age of 30 days . We have previously shown cardiac physiological changes with age in 30 day-old flies [11] . The maximal heart rate was measured at the age of 30 days . Climbing velocity was measured every 5 days between the ages of 15 to 30 days , using a non-invasive procedure . We studied 10 male flies for climbing and 10 female flies for the cardiac measurements . Survival to 30 days was also measured in these flies . Figure 1 illustrates the fully factorial dataset consisting of 81 combinations of 4 drugs , using 2 doses for each drug ( 1 control , 8 individual tests , 24 groups of 2 combined drugs , 32 groups of 3 combined drugs and 16 groups of 4 combined drugs ) . The number on the right of each combination in Figure 1 is a summary score ( z-score ) obtained from the three phenotypes mentioned: the declines with age in maximal heart rate , climbing velocity and survival . Each value was normalized by dividing by a weekly control , then for each group subtracting the population mean and diving by the population standard deviation . The z-scores from the three phenotypes were then averaged to yield a summary z-score that equally weights each of the three measurements . Analysis of Figure 1 shows that , with a larger number of drugs in the combination , there is a statistically significant increase ( p<0 . 05 ) in the percentage of treatments that have an improved z-score compared with untreated controls of the same age . The landscape ( see section in Discussion on control landscapes ) obtained from this dataset has 7 local maxima and 1 global maximum in the phenotype z-score . The maxima correspond to drug-doses configurations for which the z- score decreases by changing any of the drugs by a single dose . We have also calculated the basin of attraction , i . e . the number of drug-doses configurations that will end up in a given maximum by following the maximal increase in z-score , and found that the global maximum corresponds to the largest basin . This is an example of how landscape terminology can be used to define moderate non-linearities suitable for the algorithmic approach we suggest . The fully factorial dataset of Figure 1 was used to test the SS′ and SS-TD′ algorithms . Both algorithms were successful in finding the best combination ( and 3 of the 5 best combinations ) with a lower cost compared to an exhaustive search ( 24 and 27 tests out of 81 for the SS′ and SS-TD′ respectively ) . We performed computational simulations of multiple interventions on the apoptosis network using the two algorithms described above . The computational model is based on the apoptosis network , hsa04210 , of the KEGG database ( http://www . genome . jp/kegg/ ) . We used the discrete apoptosis model described in our previous publication [1] , where the discrete state of proteins at each node is determined by the strength of a signal from the neighboring nodes according to a logarithmic rule . In this model , the final life/death signal is calculated following the signaling in the directed network up to a final output node . The effect of a drug on a given node is modeled by changing the activity on that node and calculating the corresponding change in the output life/death node . We simulated selective killing of cells caused by drugs acting on the apoptosis network . All possible interventions on 6 , 7 , 8 , and 9 proteins , using 3 doses , were simulated . We used the dataset containing all possible interventions to study the efficacy for selective killing of the two algorithms ( SS′ and SS-TD′ ) , compared with randomly selected combinations of the same size ( see Figure 8 ) . Both algorithms were significantly more efficient than random tests ( p<0 . 0001 ) . The SS-TD′ was clearly superior in the frequency of identification of the very best combination , but the SS′ also performed well ( Figure 8 ) . If a purely additive strategy were the optimal one , the SS′ would find it , with no backtracks . However , this does not seem to be the case . In the fully factorial tests , larger combinations of up to 9 interventions were more effective than single or two-drug interventions in finding the most selective solution ( p<0 . 0001 ) . We also performed an alternative simulation changing a large number of parameters ( see Methods section ) , to test the robustness of these findings , and were able to confirm the behavior shown in Figure 8 . As suggested by the number of top combinations found by random sampling in Figure 8 , these fully factorial datasets contained multiple maxima . We investigated a different group of 30 fully factorial datasets ( using 8-drug combinations ) where maxima were very few ( less than 0 . 05% of the total ) . Not surprisingly , in these simulations , random tests never found the top combinations . However , top combinations were found in 30% of the tests by the SS′ algorithm and in 80% of tests by the SS-TD′ algorithm . Furthermore , the distances of the best solutions found from the real maxima ( expressed in % of the optimal value ) were: 9 . 2±1 . 4 ( mean±SEM ) for random tests , 4 . 7±1 . 2 for SS′ and 0 . 3±0 . 1 for SS-TD′ . All differences between groups were statistically significant ( p<0 . 01 ) . Two lymphoma cell lines , RS 11846 and DoHH2 , were used . These cell lines were chosen for the simplicity of the culture conditions , aiming to validate the method . Future tests will explore selectivity including also normal cells and cells with different tumorigenic potential . The number of viable cells was measured using a luminescence test for ATP ( ATPlite , PerkinElmer ) . We used three different doses ( for 60 hours ) of six drugs affecting cell viability: Vincristine , Etoposide , Rituximab , Apogossypol , Dexamethasone and qVD-OPH . The first five drugs can induce cell death as individual interventions while the last is an inhibitor of cell death . We compared the SS-TD′ algorithm with random combinations . After 36 tests for each cell line using individual doses we measured 91 combinations using the SS-TD′ algorithm and 107 randomly chosen ones ( 107 was the maximum theoretical number of tests required by the algorithm ) . The steps followed the order: couples , triplets , sextuplets , quintets , quartets . The SS-TD′ selectivity ( mean 21 . 3±2 . 4% ) was markedly better than that in the random approach ( mean 1 . 9±2 . 5% , p<0 . 0001 ) . Furthermore , none of the five most selective combinations could have been found with the traditional approach of combining only drugs that are cytotoxic individually , since these five combinations all contained qVD-OPH . The cancer cell results are shown in Figure 9 .
Several concepts ( e . g . synergy ) have been developed in the past for the study of combinations of mainly two drugs [3] , [18] , [19] . Synergy is useful but it is not a necessary property for the optimal combination In any case , our algorithm objective ( finding the best combination ) includes the case where this optimal result is due to synergy . A PubMed search for algorithm and “combination therapy” identified 101 papers . All of the abstract and the papers that appeared relevant were reviewed . Most papers describe sets of clinical rules derived from clinical experience or from randomized clinical trials , relevant to combinations of 2–3 drugs , to be implemented by physicians . These approaches were called therapeutic , diagnostic , treatment , management or decision algorithms . A few papers [20]–[22] describe algorithms to be implemented in silico and providing guidance for some drug combinations of small size , using disease specific information . None of these papers described search algorithms suitable for partially in vivo or in vitro searches as those we describe . A recent interesting paper [23] describes the use of stochastic algorithms for the search for optimal drug combinations . The methods described are not directly suitable for parallel biological measurement but stochastic methods , for example genetic algorithms , can certainly be adapted for this purpose . Several current cancer chemotherapy regimens are composed of combinations of 6 or more drugs . Examples , indicated by their acronyms and followed by the respective number of drugs , are: BEACOPP 7 , ChlVPP/EVA 7 , MACOP-B 6 , ProMACE-CytaBOM 9 , MOPPEBVCAD 10 , m-BACOD 6 [24]–[27] . When the algorithms suggested here search within a pool of drugs the best combination found can be of any size . In other words when searching within all possible combinations of different doses of 10 drugs , it is possible that the best combination emerging might be composed of only 3 drugs , as for example in the Drosophila dataset of Figure 1 . An important question whether we can determine the maximum number of compounds that a combination should have . Our opinion is that such a maximum limit cannot be set as a general rule , based on the following considerations: We can think of drug interventions as transmitting information to biological networks . When we search for optimal drug combination the efficiency of transmission of information ( the domain of information theory ) is important , and it is therefore not surprising that some modified algorithms from digital communications , which are used to efficiently decode signals in the presence of noise , might be applicable . There are , however , also several differences that require modifications to these algorithms . Among the similarities with the digital communication applications of sequential decoding algorithms are the following: the partial exploration of a tree of possible solutions , the dependence of the score on the previous steps of the algorithm , the objective of maximizing the score and minimizing the cost , and the use of an ordered list to store the solutions . Among the differences are the following: the partially different data structure to be explored and the related possibility of jumping to different parts of the tree and even ignoring some steps ( for example , SS-TD class algorithms are not used for decoding and are unlikely to be useful ) , and the tendency of the largest combinations to have higher scores . The computational cost is also partially different . For example memory is not a limiting factor but the number of tests and the time required by each step are . We would also like to discuss the drug safety implications of the use of drug combinations in general and of our approach more specifically . Of the two main types of adverse drug events , type A adverse events represent the majority [30] . These are dose-related and arise from the pharmacological action of the drug [30] . Type A adverse events are not necessarily increased in combinations if we use reduced doses of each drug . Furthermore , the objective metric of the algorithm can incorporate the reduction of adverse effects . An example is the choice of selective cell death for the cancer cell measurements we report , rather than just the killing of cancer cells . If we were to find a large therapeutic combination that had an extremely selective action only on cancer cells ( or on a particular cancer ) , this would have a greatly improved safety profile compared to any of the existing chemotherapeutic regimens . The second major type of adverse drug events , type B , is much more rare and not dose-related . These adverse events are at least in part genetic [30] and should be ameliorated by including genomic data as one of the components of our algorithms in future implementations . As for single drugs , medication safety is always a balance of risks and benefits . Some types of cancer have a prognosis of only a few months . Hence the risk-benefit analysis cannot be discussed for drug combinations in general , but depends on the type of disease , the type of drugs involved and the condition and informed choice of each patient . Drug interactions are a known cause of adverse events , but , given that multi-therapy is common and essential for many patients ( most hospitalized patients receive at least six drugs [31] ) , it is preferable to develop formal methods of assessment , as we suggest , rather than leaving the development of multi-therapies to the empirical decision of individual physicians . We have mentioned in the Introduction that one of the desirable features of these algorithms is the capacity of dealing with non-linearities in drug combinations . The most suitable measures of non-linearity can be obtained by building an n-dimensional “control landscape” , where the dimensions are the drugs , at different doses . The notion of landscape represents a commonly used concept in the analysis of many complex systems encountered in physics , biology , computer science and engineering [8] . Several features can provide a quantitative characterization of these landscapes , such as the density of optima [32] or the ruggedness [8] . The ruggedness measures the correlation of the biological score to be optimized in “neighboring” positions and can be obtained by defining random walk processes in the drug configuration space , and by calculating the correlation length of the score in such processes [32] . The landscape could also be modified using system-wide biological data ( omic data ) to reduce non-linearities . This omic warping is analogous to approaches commonly used in physics . While the tests in cancer cell lines reported here do add evidence supporting the efficacy of the suggested algorithms , it would be desirable in future experiments to give priority to the collection of fully factorial datasets . Comparisons with random samples have several limitations , including the fact that the true optimum is unknown . Fully factorial datasets are , when experimentally feasible , more informative , allowing the characterization of the landscapes and the evaluation of alternative search algorithms . There is a more general rationale supporting the use of algorithms integrating information on the state of the system with iterative measurements . The Artificial Intelligence community realized at the beginning of the 90s that robots could not manage a complex environment utilizing only explicit models of reality [33] . An alternative approach that started from simpler stimulus-response algorithms was more successful and was later integrated with the older models in hybrid architectures [33] . The proponents of this approach ( among them Rodney Brooks ) argued that this process was similar to the evolution of the nervous system , which is based on stimulus-response mechanisms of increasing complexity in lower invertebrates , integrated ( but not replaced ) by representations of reality within the brain of higher organisms . See also Figure 1 of Pfeifer et al [34] . Similarly we can start from “stimulus-response” algorithms and then improve them using progressively more detailed and mechanistic models of biological networks . The algorithms we have described are composed of several iterations , each depending on the previous response of the system . As pointed out [35] , control and optimization algorithms do contain information about the system , when effective , but only in an implicit form . This approach , used to control very complex and partially unknown systems by natural evolution and by possibly the most ambitious attempt to emulate evolution , building intelligent machines , is a general strategy that motivated the development of our algorithms . It is useful to consider how system-wide molecular data ( such as genomic , proteomic , metabolomic and transcriptomic data ) could be used in the context of our searches . These omic datasets could affect the ranking in two ways: as objects of multivariate analysis and as parameters of mechanistic network models , as in Figure 10 . Pattern recognition methods and multivariate statistics can be used to analyze system-wide molecular data [36] . With these models , it could be possible to distinguish the groups studied in a multi-dimensional representation . For example , it might be possible to test whether a combination brings the metabolic and transcriptional profiles of treated cells or organisms closer to that of the target state and by how much . A similar approach was used in a recent publication by Lamb et al [37] , where a single score was obtained to represent the response of a breast cancer cell line to drugs . The score was a summary of multivariate biological data ( microarrays ) . This statistical approach is justified by the fact that not all molecular information is included in the network models , but is expected to play a lesser role as the comprehensiveness of the models improves . Metabolic models similar to that described in our recent paper on Drosophila hypoxia may also play a role [38] . Gene expression data of metabolic enzymes and NMR measurements of metabolites for individual treatments could be added to the model and the effect of combining the interventions can be simulated . The model can provide summary measures that have an important effect on function , such as ATP production , and are ideally suited as weighted modifiers of the algorithm rankings . For the cancer experiments we could iteratively modify the apoptosis computational networks described in our recent paper [1] . To reflect the results of intervention experiments , one could add to the model the targets of all the drugs used , and use microarray data specific for the cell types to modify the simulations . As our biological knowledge improves , mechanistic models should play an increasing role . The algorithms described here are suitable as frameworks to integrate imperfect information from different sources . The information can be used to modify the rankings and fully factorial datasets can be used to assign weights to different types of information . For example , if the cytoprotective protein Bcl-2 is overexpressed in a target cell type or if network simulations indicate that it is an important control node , one could modify the ranking metric of combinations including drugs acting on it and test whether this improves the efficiency of the algorithms within our fully factorial datasets . There is great interest in personalized medicine and it is clear that personalized therapy requires combinations , since we cannot develop a different drug for each patient . The information on the state of the system that we suggest should be incorporated in the algorithms can at the same time provide a molecular profile corresponding to each effective combination . In other words an omic-combination dictionary could be built listing the untreated genomic , transcriptomic , proteomic and metabolomic profile optimally responding to a drug combination , and this information could guide therapy in individual patients . The algorithms could be used not only to find optimal combinations for specific diseases but also for individual patients when repeated sampling is feasible , for example in studies of chemosensitivity of cells from the blood of leukemia patients [39] . Novel technology for high-throughput screening and for omic data measurements might allow us to develop new combined pharmacological interventions adapting algorithmic and theoretical approaches from more quantitative sciences We report data from computational simulations and from biological experiments in vivo and in cell culture , suggesting that modified search algorithms from information theory have the potential to enhance the discovery of novel optimal or near-optimal therapeutic combinations . It would be desirable to obtain a larger number of fully factorial datasets , for different biological systems . This would allow a direct comparison of the algorithms reported here with other reasonable alternatives , such as stochastic algorithms . Fully factorial datasets would be even more useful if they were to include system-wide molecular ( omic ) data , at least for the single drug and for untreated cases . While this might require a considerable experimental effort , it would allow this area of research to be firmly established and provide a resource for scientists with different algorithmic backgrounds to test their ideas . Several colleagues have already pointed out analogies with other computational problems within their fields of expertise that might lead to useful alternative approaches . For example a colleague has suggested that exploring alternatives within the class of “online algorithms” is a promising area of future work . Other colleagues have proposed that modern biologically-inspired heuristic methods , such as “particle swarm optimization” , might also be used to search for optimized drug combinations . In the next few years we plan to obtain and make fully available on the web additional fully factorial datasets for drug-induced selective cell death , and we hope that this will stimulate interdisciplinary interest in this approach to the problem of multi-drug therapy .
A detailed account of the Drosophila cardiac aging model was presented previously , in which an age-dependent decline in Drosophila cardiac rate under stress was reported [11] . We developed new methods for imaging rapidly and non-invasively the adult Drosophila heart and for automated measurement of heart rate and its variability . To assess exercise capacity in Drosophila and changes with age , climbing velocity was measured using a method described by Gargano et al . [40] , modified to include image processing that allowed individual flies to be studied . The flies were transferred into 15-ml tubes and the operator tapped the top of the tube . Owing to their capacity for geotaxis orientation , flies tend to climb upwards . A digital imaging system/camera ( Motionscope PCI , Redlake Imaging MASD , Inc . ) with an attached Vivitar wide-angle lens , was used to capture video sequences at 60 frames per second of the flies as they climbed the tube . Images were analyzed with software ( MotionScope 2 . 21 . 1 ) and for each fly within the tube an individual velocity was obtained . The selected compounds and doses ( in the fly food ) were: doxycycline , with concentrations at 0 . 5 mg/mL and 1 mg/mL; sodium selenite , at 0 . 005 mg/mL and 0 . 0125 mg/mL; zinc sulfate , at 0 . 5 mg/mL and 1 mg/mL; and resveratrol , at 0 . 25 mM and 0 . 5 mM . The data sets , used to test the SS′ and SS-TD′ algorithms , were created using the apoptosis model [1] , with some changes concerning the search procedure and the output value . Instead of performing an exhaustive search on all the nodes of the network , we limited the search to a randomly chosen subgroup of nodes . We also used as output value the difference of the cubic value of one individual compared to the average of the cubic sum of the remaining population , to reward the individuals with the highest values . Confirmatory simulations were also performed , to test the robustness of our findings by changing several parameters . The parameters were the number of states for nodes and links , the starting values for the states , the ranges of the output of the simulation , and the nodes selected for the interventions . The software was written in C++ and implemented on 32-nodes of a 64-bit Linux cluster with 2GB of memory per node . The longest searches required about 30 minutes of computation . The analysis of the collected data consisted of three separated steps: sort , search algorithm and statistical analysis . For the first step , a quick sort implementation was used creating different ranks for each individual . In the second step , all the algorithms and random execution returned information for each rank . These were used in the last step , where we collected the statistical analysis data , dividing the resulting population into different samples , to compare each algorithm with the others . Owing to the dimension of the data , it was necessary to limit the number of analyzed nodes to a maximum of 9 . Computational time was significant only for the sorter , requiring several hours for the largest files on an entry-level Linux workstation . ATP is a marker for cell viability because it is present in all metabolically active cells and the concentration declines very rapidly when the cells undergo necrosis or apoptosis . Human tumor cells DOHH2 and RS11846 were maintained as suspension cultures at standard conditions: humidified atmosphere with 5% carbon dioxide , at 37°C in an incubator , using RPMI-1640 medium , supplemented with 10% heat-inactivated fetal calf serum and 2 mM L-glutamine . Cells were kept in log phase via replacement of cellular suspension aliquots by fresh medium two or three times weekly . Stock solutions of the 6 chosen drugs were freshly prepared in water ( Vincristine ) , physiological saline solution ( Rituximab ) or DMSO ( Etoposide , Q-VD-Oph , Apogossypol and Dexamethasone ) . The stock solutions were diluted with RPMI-1640 in order to obtain the desired final concentrations . Less than 0 . 5% of the solvent was present at the final dilutions . All the procedures related to cell culture , drug preparation , and treatment were carried out in a laminar flow cabinet . Briefly , exponentially growing cells were seeded in 96-well plates ( 90 µL aliquots/well ) at a density of 5 . 55 104 cells/mL and 10 µL of drug solution were added . Final concentrations of the drug were the following: Vincristine ( 0 . 01 , 0 . 1 , or 0 . 5 nM ) , Etoposide ( 0 . 01 , 0 . 1 , or 1 µM ) , Apogossypol ( 1 , 2 . 5 , or 4 µM ) , Q-VD-OPh ( 5 , 10 , or 25 µM ) , Rituximab ( 5 , 15 , or 20 µg/mL ) , Dexamethasone ( 0 . 1 , 1 , or 25 µM ) . Plates were incubated for 60 hours . After the incubation , 30 µL aliquots of ATPlite reconstituted reagent ( Perkin-Elmer ) were added to every well . The plates were shaken for 3 minutes at 750 rpm ( Eppendorf MixMate ) . The absorption of the samples was measured using a monolight 3096 microplate luminometer ( BD ) . Ten µL of a 10 mM ATP solution was added to every well as internal standard . The plates were shaken for 2 minutes at 750 rpm and read . Selectivity was defined as the difference in % survival between the two cell types . All results are expressed as mean±standard error of the mean . For comparisons of 2 groups unpaired t tests were used ( non-parametric tests were also significant ) and for comparison of more than 2 groups we used one-way analysis of variance with Bonferroni correction for post-test comparisons . The Drosophila data presented in Figure 1 were analyzed using the chi square test for trends and results were confirmed using one-way analysis of variance with linear test for trend . The number of combinations in the introduction was obtained using Newton's Binomial series up to the 6th order . The statistical software used was Prism ( GraphPad ) .
|
This work describes methods that identify drug combinations that might alleviate the suffering caused by complex diseases . Our biological model systems are: physiological decline associated with aging , and selective killing of cancer cells . The novelty of this approach is based on a new application of methods from digital communications theory , which becomes useful when the number of possible combinations is large and a complete set of measurements cannot be obtained . This limit is reached easily , given the many drugs and doses available for complex diseases . We are not simply using computer models but are using search algorithms implemented with biological measurements , built to integrate information from different sources , including simulations . This might be considered parallel biological computation and differs from the classic systems biology approach by having search algorithms rather than explicit quantitative models as the central element . Because variation is an essential component of biology , this approach might be more appropriate for combined drug interventions , which can be considered a form of biological control . Search algorithms are used in many fields in physics and engineering . We hope that this paper will generate interest in a new application of importance to human health from practitioners of diverse computational disciplines .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"pharmacology/personalized",
"medicine",
"pharmacology",
"computational",
"biology/systems",
"biology"
] |
2008
|
Search Algorithms as a Framework for the Optimization of Drug
Combinations
|
Elongator is a six subunit protein complex , conserved from yeast to humans . Mutations in the human Elongator homologue , hELP1 , are associated with the neurological disease familial dysautonomia . However , how Elongator functions in metazoans , and how the human mutations affect neural functions is incompletely understood . Here we show that in Caenorhabditis elegans , ELPC-1 and ELPC-3 , components of the Elongator complex , are required for the formation of the 5-carbamoylmethyl and 5-methylcarboxymethyl side chains of wobble uridines in tRNA . The lack of these modifications leads to defects in translation in C . elegans . ELPC-1::GFP and ELPC-3::GFP reporters are strongly expressed in a subset of chemosensory neurons required for salt chemotaxis learning . elpc-1 or elpc-3 gene inactivation causes a defect in this process , associated with a posttranscriptional reduction of neuropeptide and a decreased accumulation of acetylcholine in the synaptic cleft . elpc-1 and elpc-3 mutations are synthetic lethal together with those in tuc-1 , which is required for thiolation of tRNAs having the 5′methylcarboxymethyl side chain . elpc-1; tuc-1 and elpc-3; tuc-1 double mutants display developmental defects . Our results suggest that , by its effect on tRNA modification , Elongator promotes both neural function and development .
Regulation at the level of translation is one important way in which gene activity is controlled in metazoans . Several different mechanisms have previously been identified by which translation can be regulated during development or memory formation [reviewed in 1] . During anterior-posterior patterning of the Drosophila embryo , the translation of hunchback mRNA in the posterior region of the embryo is inhibited by binding of a protein complex to the Nanos response element in the hunchback 3′UTR [2] . In Caenorhabditis elegans , developmental timing is controlled by the small temporal RNAs , lin-4 and let-7 , which act by forming heteroduplexes with their target mRNAs and , at least in some cases , suppressing their translation [3] . Translation efficiency is also regulated by phosphorylation of translational components at the initiation and elongation steps [4] , [5] . For example , during memory formation in mice , translation of ATF4 mRNA is regulated by phosphorylation of initiation factor eIF2α [6] . Another way in which the efficiency of translation can be modulated is by covalent modification of nucleosides in the anticodons of tRNAs . In the decoding of mRNA , modified nucleosides in the anticodon region , especially position 34 ( wobble position ) and position 37 , have been suggested to be important for restriction or improvement of codon-anticodon interactions [7]–[10] . In S . cerevisiae , 25% of the tRNA species are covalently modified by the addition of either carbamoylmethyl ( ncm ) or methoxycarbonylmethyl ( mcm ) side chains to the 5′carbon of U34 [11]–[14] . A subset of these tRNAs contains a further modification on wobble uridines , addition of a thio group at the 2′position ( Figure 1 ) [11] , [13] , [14] . In vivo , presence of an 5-carbamoylmethyluridine ( ncm5U ) , an 5-methoxycarbonylmethyluridine ( mcm5U ) or an 5-methoxycarbonylmethyl-2-thiouridine ( mcm5s2U ) improves reading of both A- and G-ending codons [14]–[16] . In S . cerevisiae , formation of ncm and mcm side chains present at 5′position of wobble uridines requires the Elongator complex [12] , which is composed of six subunits Elp1p – Elp6p [17] , [18] . Yeast cells lacking Elongator activity are viable but display multiple defects including those in PolII transcription and exocytosis [16] , [18]–[22] . However , these defects all appear to result from a primary defect in tRNA modification [16] . Elongator complex is conserved in eukaryotes and has also been purified from humans [23] . Inactivation of Elongator subunits in multicellular organisms causes multiple defects including those in development , cell proliferation , cell migration and neuron projection [24]–[27] . Recently , Elongator in mice has been reported to acetylate α-tubulin [27] . However , it is presently unclear whether Elongator in higher eukaryotes functions directly in multiple processes or acts on a small number of targets whose absence leads to pleiotropic defects . Mutations in the human homologue of yeast ELP1 , IKBKAP/hELP1 , have been shown to cause Familial Dysautonomia ( FD ) , a genetic disorder primarily affecting the sensory and autonomic nerve systems [28]–[30] . Human IKAP/hELP1 protein is part of a complex of six proteins that also contains the human homologues of yeast Elongator proteins [23] . Whether Elongator in humans or other metazoans promotes tRNA modification has not been reported . The aim of the present study was to investigate the function of the Elongator homologues , ELPC-1 and ELPC-3 in the nematode , C . elegans . In particular , we were interested to determine first , whether Elongator in metazoans is required for modification of wobble uridines , and second , whether C . elegans could be established as a model to study the role of Elongator in modulating translation within neurons and other tissues . We demonstrate that Elongator is required in C . elegans for the formation of modified nucleosides in tRNA , and that Elongator mutants have defects in neurological and developmental processes associated with reduced translation . We believe our results also have important implications for the etiology of FD disease .
Searches of the C . elegans protein sequence database with the yeast or human Elp1p and Elp3p sequences revealed that C . elegans contains single Elp1p and Elp3p homologues , named ELPC-1 and ELPC-3 , which are encoded by Y110A7A . 16 and ZK863 . 3 respectively [see Materials and Methods for an explanation of gene nomenclature] . To investigate the function of elpc-1 and elpc-3 in C . elegans , we used elpc-1 ( tm2149 ) and elpc-3 ( tm3120 ) , deletion mutants kindly supplied by S . Mitani of the National Bioresource Project , Japan . The elpc-1 ( tm2149 ) deletion removes 275 bp of sequence spanning parts of exons 7 and 8 ( Figure 2A ) , whereas the elpc-3 ( tm3120 ) removes 356 bp spanning the first half of exon 3 and contains as well an insertion of four nucleotides in the second half of exon 3 ( Figure S1A ) . The elpc-3 ( tm3120 ) deletion removes part of a sequence sharing significant homology to the Radical S-adenosylmethionine ( SAM ) superfamily [31] . Members of this family of proteins contain an FeS cluster and use S-adenosylmethionine ( SAM ) to catalyse a variety of radical reactions . The Elp3p Radical SAM domain has been found to be required for iron binding in Methanocaldococcus jannaschi [32] , and for integrity of the Elongator complex in yeast [33] . In yeast , Elp1p and Elp3p are required for the formation of mcm5 and ncm5 side chains of modified nucleosides mcm5U , ncm5U , ncm5Um and mcm5s2U present at the wobble position in tRNA [12] . To determine whether their homologues in C . elegans , ELPC-1 and ELPC-3 , also function to promote wobble uridine tRNA modification , we examined if the mcm5U , ncm5U or mcm5s2U modified nucleosides were present in tRNA isolated from wild-type and elpc mutants . Total tRNA isolated from wild-type worms contained ncm5U and mcm5s2U nucleosides ( Figure 2B and 2D , Figure S1B , S1D ) . However , no mcm5U was detected ( Figure S2D ) , implying that modification of uridine in C . elegans tRNA differs in at least one respect from that in S . cerevisiae . In contrast to wild-type worms , no mcm5s2U or ncm5U nucleosides were present in tRNA isolated from elpc-1 ( tm2149 ) mutants ( Figure 2C and 2E ) . Instead , 2-thio uridine ( s2U ) was detected in tRNA isolated from the elpc-1 ( tm2149 ) mutant but not from wild-type worms ( Figure 2F and 2G ) . This nucleoside arose from a failure in the mutant to add the mcm5 side chain of the mcm5s2U nucleoside . The tRNA modification defect in the elpc-1 ( tm2149 ) mutant was rescued by elpc-1 activity provided by a transgene ( Figure 2C and 2I ) . Thus , like yeast Elp1p , C . elegans ELPC-1 is required for the formation of mcm5 and ncm5 side chains in tRNA . Consistent with the tRNA modification defect in the elpc-1 ( tm2149 ) mutant , tRNA isolated from elpc-3 ( tm3120 ) mutants lacked the mcm5s2U and ncm5U nucleosides and instead contained s2U ( Figure S1 ) . Synthesis of the s2 group of mcm5s2U in yeast requires Tuc1p [15] , [34]–[38] . The homologue of Tuc1p in C . elegans is encoded by open reading frame F29C4 . 6 [39] . In this paper we will refer the F29C4 . 6 gene as tuc-1 . We analyzed tRNA from tuc-1 ( tm1297 ) mutant worms by HPLC and confirmed that it lacked the mcm5s2U modification and instead had mcm5U , a nucleoside not normally found in C . elegans tRNA ( Figure S2B , S2C , S2D , S2E ) . Furthermore , a transgene containing wild-type tuc-1 DNA restored formation of mcm5s2U in tRNA ( data not shown ) . Consistently , tRNA isolated from an elpc-1 ( tm2149 ) ; tuc-1 ( tm1297 ) double mutant lacked both the 5′- and 2′ side-chains of wobble uridines and no ncm5U or mcm5s2U nucleosides were observed ( Figure S3 ) . To investigate the expression pattern of C . elegans ELPC-1 in various tissues , we examined worm strains harboring a transgene encoding functional , full length ELPC-1 protein fused to GFP . The transgene contained 435 bp of the promoter region and all 11 introns ( Figure 2A ) . The transgene rescued the tRNA modification defect in the elpc-1 mutants ( Figure 2C and 2I ) . The fusion protein encoded by the transgene was preferentially detected in several tissues including the nervous system ( Figure 3 ) . However , its presence was not uniform . Within the nervous system , ELPC-1::GFP was seen predominantly in a pair of neurons that control egg-laying , the HSNs ( Figure 3F and 3G ) , and in chemosensory neurons in the head ( Figure 3A–3E ) . Within the latter class of neurons , the ELPC-1::GFP level was particularly high in the ASE , ADF and ASK pairs of neurons ( Figure 3B–3E . For nomenclature of neurons , see Materials and Methods ) . Expression was seen both within the cell bodies ( Figure 3B ) and along the entire lengths of the neuronal processes ( data not shown ) . Outside of the nervous system , a strong ELPC-1::GFP signal was seen in the pharynx ( the feeding organ ) ( Figure 3A ) and the vulva ( Figure 3N and 3O ) , part of the egg-laying apparatus in the hermaphrodite . In all animals examined , ELPC-1::GFP expression was also seen in the two CAN cells ( Figure 3H and 3I ) , which are associated with the excretory canals and are required for proper function of the excretory system . In all cells in which ELPC-1::GFP was seen , fluorescence was restricted to the cytoplasm ( Figure 3A ) . The ELPC-3::GFP fusion was expressed in the same set of cells ( data not shown ) . In S . cerevisiae , defects in wobble uridine tRNA modification are associated with reduced translation efficiency [14]–[16] , [40] . The yeast elp3 tuc1 double mutant , in which modifications at both the 5′and 2′positions of the uridine moiety are absent , is lethal [15] . To investigate the influence of wobble uridine modifications on the efficiency of translation in C . elegans , we examined the effect of elpc-1 , elpc-3 and tuc-1 mutations on β-galactosidase expression in worms harboring a lacZ transgene driven by heat shock-responsive elements from the hsp16-1 gene . The induction of lacZ mRNA upon heat shock was not reduced in strains lacking wild-type elpc-1 , elpc-3 or tuc-1 gene activity , or in elpc-1; tuc-1 double mutant worms ( Table 1 ) . However , β-galactosidase activity was 28% lower in protein extracts from heat shocked elpc-1; tuc-1 double mutants than in those from wild-type worms subjected to the same heat shock regime ( Table 1 ) . A modest ( ∼14–18% ) but significant reduction in β-galactosidase activity was also seen elpc-1 ( tm2149 ) , elpc-3 ( tm3120 ) or tuc-1 ( tm1297 ) single mutant worms ( Table 1 ) . To monitor cell and tissue specific protein synthesis , we used an established technique , fluorescence recovery after photobleaching ( FRAP ) [41] . The rate of protein synthesis in different cells and tissues was measured using GFP reporters . We used gcy-5::gfp and mec-4::gfp which are expressed in ASER and 6 touch cell neurons respectively , and myo-3::gfp which is expressed in the body wall muscle . In all reporter fusions examined , photobleached wild type animals showed a significant recovery of GFP signal within 5 hours ( Figure 4 , Figure S4 ) . However , animals with the elpc-1 ( tm2149 ) or elpc-3 ( tm3120 ) mutations had a slower GFP signal recovery , indicating a reduced rate of protein synthesis ( Figure 4 , Figure S4 ) . Cycloheximide , an inhibitor of translation , was used to confirm that the recovered GFP signal was due to newly synthesized protein . In animals treated with cycloheximide , no significant recovery of GFP signal was observed ( Figure 4 , Figure S4 ) . Together , these experiments demonstrate that an absence of uridine modification in tRNA is associated with a reduction in translation efficiency in C . elegans . elpc-1 and elpc-3 single mutants were viable and fertile and they were able to move normally on the bacterial lawn . Furthermore , the chemosensory neurons in which ELPC-1::GFP and ELPC-3::GFP are strongly expressed are present at their normal positions and have normal morphology ( Figure S5 ) . Among these neurons ( Figure 3A–3E ) , the ASE pair of sensory neurons is required for experience-dependent behaviors elicited by different salt concentrations [42] . Wild-type worms normally chemotax towards NaCl . However , pre-incubation in normal salt concentrations in the absence of nutrients elicits an aversion response to NaCl when worms are subsequently tested in chemotaxis assays [43] . In this salt learning assay , worms that have grown at normal salt concentrations and in the presence of abundant nutrients are first starved for four hours in the presence or absence of salt and then assayed for their chemotactic response to NaCl . Since we observed strong expression of ELPC-1 and ELPC-3 in ASE neurons ( Figure 3B and 3C , data not shown ) , we tested elpc-1 and elpc-3 mutants in a salt learning assay . At 20°C , the mutants behaved as wild type ( Figure 5A ) . At 25°C , wild-type worms exposed to 100 mM NaCl in the absence of nutrients moved away from NaCl , whereas elpc-1 or elpc-3 mutants treated in the same way continued to chemotax towards the NaCl ( Figure 5C ) . In the elpc-1 mutant , this defect was partially rescued by the elpc-1::gfp construct ( Figure S6 ) . Thus , C . elegans elpc-1 and elpc-3 are required for an experience-dependent change in behavior . In contrast , in tuc-1 mutant worms no statistically significant changes were observed ( Figure 5 ) . Ablation of the ASE neurons leads to an inability to chemotax towards certain water-soluble compounds including Na+ , Cl− , lysine and biotin [44] . elpc-1 and elpc-3 mutants were able to chemotax both to water soluble and volatile compounds at all temperatures tested ( Figure S7 ) . When elpc-1 or elpc-3 mutants were grown at 20°C to the time at which the chemosensory neurons have developed and then shifted to 25°C , salt learning was abnormal ( Figure 5B ) . Together , these observations suggest that the salt learning defect seen in elpc-1 and elpc-3 mutants is not caused by a defect in the development of the ASE chemosensory neurons or in their ability to detect salt . Since neuronal function in metazoans is known to be dependent upon the ability to synthesize and secrete neurotransmitters and neuropeptides , we tested whether these processes were abnormal in C . elegans elpc-1 and elpc-3 mutants . One established assay for examining the synthesis and secretion of neuropeptides involves a heterologous fusion protein , ANF::GFP . The prodomain of a preproANF–GFP fusion protein can be used as a reliable fluorescent reporter of dense-core vesicle transport and exocytosis in rat PC12 cells , as well as in D . melanogaster and C . elegans neurons [45]–[47] . In C . elegans , ANF::GFP is secreted by neurons into the pseudocoelomic space from where it is rapidly cleared by three pairs of coelomocytes [47] . In elpc-1 and elpc-3 mutants , we observed a reduced accumulation of ANF::GFP in coelomocytes ( Figure 6C and 6D ) , which could be caused by either less synthesis or reduced secretion of ANF::GFP from neurons . In both wild-type and elpc mutant worms carrying the ANF::GFP transgene , the fusion protein was visible in neurons , but the GFP signal was weaker in elpc mutants that was also reflected by western blot ( Figure 6A and 6B ) . As there was no significant reduction of ANF::GFP mRNA in elpc mutants , the lower production of ANF::GFP was at the posttranscriptional level ( Figure 6A and 6B ) . To investigate whether elpc-1 and elpc-3 also affected extracellular levels of a neurotransmitter , we examined whether the mutants showed increased resistance to aldicarb , an inhibitor of acetylcholinesterase present in the synaptic cleft . Wild-type worms exposed to aldicarb immediately hypercontract and then die after a few hours because they are unable to reduce synaptic levels of acetylcholine secreted by neurons [48] . Mutants with reduced acetylcholine-mediated signaling are partially or completely resistant to the drug . Aldicarb-resistant mutants fall into two classes , those that have pre-synaptic defects resulting in reduced synthesis or secretion of acetylcholine and those in which the fault lies in the post-synaptic neurons [49] . elpc-1 and elpc-3 mutants showed greater resistance to aldicarb than that displayed by aex-6 ( sa24 ) ( Figure 6E ) , which has been described previously as being partially resistant to the drug [50] . elpc-1 ( tm2149 ) mutant harboring the elpc-1::gfp transgene on an array behaved as wild type in the aldicarb assay ( Figure S8 ) . elpc-1 and elpc-3 mutant worms showed normal response to levamisole ( Figure 6F ) , which activates the post-synaptic acetylcholine receptor [49] , suggesting a defect in the pre-synaptic compartment . These results suggest that either less acetylcholine is produced in the neurons or less acetylcholine is released from the neurons in the elpc-1 and elpc-3 mutants . Recently it was shown that mouse ELP3 protein can acetylate α-tubulin in vitro [27] . Thus one possibility is that the neural defects seen in mice with reduced Elongator activity is caused by aberrant α-tubulin function . Acetylation of α-tubulins in a wide variety of species occurs on a conserved lysine residue at position 40 . In C . elegans , there is a single α-tubulin with a lysine at this position , MEC-12 [51] . To investigate whether Elongator in C . elegans promotes acetylation of α-tubulin , we examined acetylation in elpc-1 or elpc-3 mutants . As previously reported , an antibody that recognizes lysine 40-acetylated α-tubulin in various species , 6-11B-1 , could detect the residue in extracts from wild-type worms but not those from the mec-12 ( e1607 ) mutant . However , we observed no reduction in the levels of acetylated MEC-12 in elpc-1 or elpc-3 mutants ( Figure S9A ) . Furthermore , unlike elpc-1 or elpc-3 mutants , mec-12 ( e1607 ) is not aldicarb resistant ( Figure S9B ) . In humans , IKBKAP/hELP1 expression is not confined to the nervous system but is also seen in many other cell types [29] , [52] , [53] . In C . elegans , we also observed ELPC-1::GFP expression in several non-neuronal tissues ( Figure 3 ) . However , in an otherwise wild-type genetic background , although they grow slower than wild-type and had reduced fertility at 25°C ( Table 2 ) , the development of elpc-1 or elpc-3 mutants is not grossly disturbed . In yeast , elp1 and elp3 deletion strains are also viable . However , yeast cells lacking both ELP3 and TUC1 , which therefore lack both mcm5 and s2 groups of tRNAs having the nucleoside mcm5s2U34 , are not viable [15] . In the course of analyzing elpc-1; tuc-1 double mutant worms , we observed that the strain could be propagated at 15°C but not at 25°C . The elpc-1 ( tm2149 ) ; tuc-1 ( tm1297 ) double mutant hermaphrodites raised at 15°C for different periods of time were shifted to 25°C and then examined both for their own development and also for their ability to give rise to viable progeny . When 4th larval stage ( L4 ) hermaphrodites were shifted to 25°C , they continued to develop and became fertile adults . However , the eggs they laid arrested development during embryogenesis ( Figure 7A–7D ) . The arrest did not occur at one specific embryonic stage but rather at different stages in different embryos . Some embryos were arrested either prior to enclosure with relatively few cells ( <100 cells ) ( Figure 7A ) ; or at the 3-fold stage ( Figure 7D ) . However , the majority were arrested during or immediately after morphogenesis ( Figure 7B and 7C ) . Similar defects were seen in elpc-3; tuc-1 double mutants ( Figure S10 ) . Thus ELPC-1 , ELPC-3 and TUC-1 likely function at multiple times during embryogenesis . No synthetic defects were seen in elpc-1; elpc-3 double mutants , suggesting that Elongator function is abolished in both elpc-1 and elpc-3 single mutants . Temperature shift experiments with 1st or 2nd stage larvae ( L1 or L2 ) also indicated a role for ELPC-1 , ELPC-3 and TUC-1 in development of the vulva and for generation of germ cells . When L1 or L2 larval hermaphrodites containing both the elpc-1 ( tm2149 ) and tuc-1 ( tm1297 ) mutations were raised at 15°C and shifted to 25°C , they developed to become small sterile adults . Inspection of the shifted animals at high magnification indicated that vulval development was invariably abnormal ( Figure 7I and 7J , Figure S10I , S10J ) . In wild-type worms , the three progenitors of the vulva , P5 . p , P6 . p and P7 . p are induced to adopt vulval fates: they divide to give rise to 22 cells that together form a tube through which the eggs are laid in adult hermaphrodites . In the temperature-shifted elpc-1; tuc-1 and elpc-3; tuc-1 double mutants , the divisions of P5 . p , P6 . p and P7 . p failed to occur properly and significantly fewer vulval cells were formed ( Figure 7I and 7J , Figure S10I , S10J ) . At the L3 stage , when the vulval developmental fates are induced , expression of the elpc-1::gfp reporter was upregulated in P5 . p , P6 . p and P7 . p as well as in their immediate descendants ( Figure S11 ) , suggesting that one or more of the signaling pathways mediating vulval cell fate specification controls elpc-1 expression . Inspection of the gonads of the temperature shifted double mutants revealed that the overall organization of the germline was relatively normal ( data not shown ) . However , the oocytes completely failed to mature ( Figure 7E and 7F , Figure S10E , S10F ) ; the sperm were highly vacuolated and grossly abnormal ( Figure 7G and 7H , Figure S10G , S10H ) . These observations imply that elpc-1 and elpc-3 also function in development of non-neuronal tissues . The developmental defects in the elpc-1; tuc-1 double mutant were rescued by extrachromsomal arrays harboring the elpc-1::gfp transgene . When elpc-1 ( tm2149 ) ; tuc-1 ( tm1297 ) double mutant hermaphrodites raised at 15° were allowed to lay eggs at this temperature for two hours and the eggs subsequently shifted to 25° , the progeny invariably arrested either during embryogenesis or during early larval stages ( n = 65 ) . However , 60% ( n = 40 ) of elpc-1 ( tm2149 ) ; tuc-1 ( tm1297 ) ; svEx808[elpc-1::gfp Punc-122::gfp] embryos raised grew to become adults with normal vulval development . 15% of these adults gave rise to some live larval progeny indicating partial rescue of both the germline defect and the requirement during early embryogenesis . A second array , svEx806[elpc-1::gfp Punc-122::gfp] also rescued although not quite as efficiently: 40% of embryos grew to become adults .
Although a requirement for the Elongator complex for the modification of wobble uridines in yeast tRNA is well documented [12] , studies on the role of Elongator in this process in metazoans have not been previously reported . Our results demonstrating that ELPC-1 and ELPC-3 are required for the addition of mcm5 and ncm5 side chains to uridine residues in C . elegans tRNA imply that Elp1p and Elp3p function has been conserved in evolution . Our results also confirm , however , that differences exist in tRNA modification in eukaryotes . In yeast there are 13 tRNA species with a uridine at the wobble position . Of these , eleven contain the nucleoside ncm5U , ncm5Um , mcm5U or mcm5s2U [11]–[14] . In our analysis of C . elegans wild-type tRNAs , we found ncm5U and mcm5s2U but not mcm5U . This observation is consistent with an earlier investigation showing that mcm5U is not present in tRNAs isolated from calf liver [54] . For example , nucleoside 34 in from yeast has mcm5U [55] , whereas that from calf liver has mcm5s2U [56] . These findings suggest that mcm5U might be absent from tRNAs in metazoans . In yeast , Elongator was suggested to participate in three distinct cellular processes: transcriptional elongation , polarized exocytosis and formation of modified wobble uridines in tRNA [12] , [21] , [22] . Strong genetic evidence was provided that the pleiotropic phenotypes seen in yeast , including those in transcription and exocytosis , were caused by a translational dysfunction due to lack of mcm5 and ncm5 side chains at wobble uridines [16] . This suggests that the physiological relevant role of Elongator complex in this organism is in the formation of modified nucleosides in tRNA , i . e . wobble uridine tRNA modification is crucial for the translation of mRNAs that encode proteins important for transcriptional elongation and polarized exocytosis . Cellular localization studies primarily placed Elongator subunits in the cytosol in yeast , mouse and human cells [22] , [23] , [27] , [57]–[60] . As modifications in the anticodon region normally take place in the cytosol [61] , such a localization is consistant with a role in wobble uridine modification . In C . elegans , we did not observe any fluorescence of ELPC-1::GFP in the nucleus suggesting that Elongator in this organism functions in the cytosol . In elpc-1 and elpc-3 mutants , we observe reduced expression of an ANF::GFP neuropeptide reporter . Given that ANF::GFP mRNA levels are normal in the mutants , the reduction in ANF::GFP accumulation could in principle be explained either by increased degradation of the protein or by decreased translation . Since tRNAs are intimately involved in protein synthesis , we believe it more likely that ELPC-1 and ELPC-3 affect ANF::GFP levels by promoting translation . Further evidence indicating a role for Elongator in translation is that the recovery of GFP signals after photobleaching in strains with gcy-5::gfp , mec-4::gfp and myo-3::gfp reporter genes is slower in Elongator mutants than in wild type . The effect of Elongator on translation is also consistent with the synthetic effects we observe in elpc-1; tuc-1 and elpc-3; tuc-1 double mutants . The reduction in accumulation of β-galactosidase activity in elpc-1 or elpc-3 single mutants ( in which the mcm5 side chain of mcm5s2U containing tRNAs is absent ) is similar to that seen in tuc-1 single mutants ( in which the s2 side chain is absent ) . In the double mutants ( in which both the 2′and 5′modifications are lost ) the efficiency of translation is further reduced . An explanation for the reduced efficiency of translation in C . elegans worms lacking elpc-1 or elpc-3 activity is that the modifications of uridine residues at the wobble position aid codon-anticodon interactions [7]–[10] . Experiments in vivo with S . cereverisiae , suggest that the primary function of the mcm5U , ncm5U and mcm5s2U nucleosides is to improve binding to A- and G- ending codons , decoded by tRNAs containing these modified nucleosides [14]–[16] . For tRNAs normally modified at both the 2′and 5′positions , the absence of either modification ( or both ) did not lead to any obvious misreading of U- or C-ending codons [15] , [16] . Thus , presence of modifications at wobble uridines in tRNAs appears to promote the rate of elongation during translation rather than its fidelity . There are examples of tRNA modification mutants that show temperature sensitive ( ts ) phenotypes , suggesting a reduced functionality of the hypomodified tRNA at the elevated temperature [16] , [62] , [63] . In yeast , elp and tuc1 mutations result in hypomodification of and [12] , [15] . In the anticodon loop , both tRNAs are rich in uridines that have a low stacking potential , and in , mcm5 and s2 of U34 are required for a canonical anticodon loop structure [64] . Therefore , we believe that the temperature sensitive phenotype observed in elpc and tuc-1 single mutants and enhanced in elpc-1; tuc-1 and elpc-3; tuc-1 double mutants is caused by destabilization of anticodons in hypomodified tRNAs , resulting in further weakening of codon-anticodon interactions . The higher levels of expression of the elpc::gfp reporters within the nervous system of C . elegans is consistent with the finding that the most severe defects of elpc-1 or elpc-3 single mutants are observed in nervous system . It is interesting to note that a strong expression of Elongator subunits was also observed in the nervous system of mice [27] . A possible explanation for the greater requirement for Elongator in neurons is that neurons have markedly higher rates of protein synthesis than most other cell types [41] , [65] , [66] . It is also striking that in both C . elegans and mice , expression within the nervous system is not uniform . Perhaps different neurons have different rates of translation . In C . elegans and other metazoans , neuronal function is dependent upon the ability to synthesize and secrete neurotransmitters and neuropeptides . In elpc-1 and elpc-3 mutants , the production of ANF::GFP neuropeptide is reduced at the posttranscriptional level . Thus Elongator mutations might cause the neurological defects by impairing the translation of neuropeptides . In addition , our findings that elpc-1 and elpc-3 mutants appear to have reduced levels of acetylcholine in the synaptic cleft suggest that Elongator is required for the production or secretion of neurotransmitter . Since elpc-1::gfp and elpc-3::gfp are expressed in a set of chemosensory neurons , the salt chemotaxis learning defect displayed by Elongator mutant worms is likely to be a consequence of inefficient communication among various neurons due to low production of neurotransmitters or neuropeptides . It is interesting to note that mutations in the human ELP1 gene , also called IKBKAP , cause the neurodegenerative disease , Familial Dysautonomia ( FD ) [28] , [29] . Furthermore , association studies in humans have revealed that variants at the ELP3 locus confer increased risk for the neurodegenerative disorder Amyotrophic Lateral Sclerosis ( ALS ) [67] . Neuronal defects are also observed in Drosophila , Zebrafish and mouse with reduced function of ELP3 [27] , [67] . Conflicting reports exist concerning the origin of the defects caused in human cells by a reduction in hELP1/IKAP levels [25] , [26] , [52] , [68] , [69] . Recently , in mice Elongator was suggested to catalyze α-tubulin acetylation [27] . However , our observations that acetylation is not obviously abnormal in C . elegans elpc-1 or elpc-3 mutants suggest that the neuronal defects observed in Elongator mutants in the worm are not caused by a failure to acetylate α-tubulin . In contrast to the elpc-1 and elpc-3 mutants , tuc-1 mutants do not display defects in either the salt learning assay or in secretion of ANF::GFP . In yeast , the growth defects of Elongator mutants are more pronounced than those of the tuc mutants [15] , [16] . One possible explanation for these differences might be that the absence of the s2 modification has less effect on codon-anticodon interactions than the absence of ncm5- and mcm5-groups . Alternatively , the effects on salt learning might be caused by reduced expression of a protein encoded by an mRNA rich in codons decoded by tRNAs harboring solely the 5′modification . Previous studies on ELP1 and ELP3 function in vertebrates have focused on their roles in neurons . While we have shown that ELPC-1 and ELPC-3 are important for nervous system function in the worm , our results clearly demonstrate that they also act in non-neural tissues . Although their expression is far from ubiquitous , the expression of ELPC::GFP reporters is clearly not restricted to neurons . More importantly , the defects in temperature-shifted elpc-1; tuc-1 and elpc-3; tuc-1 double mutants indicate that Elongator is also involved in embryogenesis and vulval development . The phenotypes observed suggest that tRNA modification is a mechanism by which the efficiency of translation is modulated during metazoan development . Our observations suggest that Elongator acts in neurological and developmental processes in C . elegans by modulating translation . An important task in the future is to identify the mRNAs whose translation is dependent on Elongator activity . Identification of these mRNAs might help in the understanding of the molecular mechanisms of Elongator-related human diseases .
The names of genes in the text have been given according to existing nomenclature rules for S . cerevisiae , C . elegans and humans . The yeast ELP1 gene encodes a protein , Elp1p; the equivalent gene in C . elegans , elpc-1 encodes ELPC-1; in humans , IKBKAP/hELP1 encodes IKAP/hELP1 . The respective mutant alleles are elp1 ( S . cerevisiae ) and elpc-1 ( tm2149 ) ( C . elegans ) . Neurons in C . elegans have three-letter names e . g . , ASE . These names are not acronyms or abbreviations . C . elegans worms were cultured and handled as described previously [70] . All strains were maintained at 20°C unless likewise indicated . All are derived from the wild-type strain , Bristol N2 [70] . For routine propagation , worms were maintained on nematode growth medium ( NGM ) plates [70] . The following mutations were used in this study . Linkage group ( LG ) I , tom-1 ( ok285 ) [71]–[73] , aex-6 ( sa24 ) [50] , lev-11 ( x11 ) [74]; elpc-1 ( tm2149 ) , LG III , mec-12 ( e1605 ) , mec-12 ( e1607 ) [75] , LG IV , tuc-1 ( tm1297 ) [39]; LG V , elpc-3 ( tm3120 ) . The transgenes used were ubIn5[hsp16::lacZ] [76] , oxIs180[Paex-3::ANF::gfp] [47] , svEx557[Pelpc-1::elpc-1::gfp] , zdIs5 I[mec-4::gfp lin-15 ( + ) ] [77] , svEx666[lin-25::HA myo-3::gfp] , svEx806[elpc-1::gfp Punc-122::gfp] , svEx808[elpc-1::gfp Punc-122::gfp] , adEx1262[gcy-5::gfp lin-15 ( + ) ] [78] . The elpc-1 , elpc-3 and tuc-1 deletion mutants were backcrossed eight times with wild-type N2 before use . To generate the elpc-1::gfp fusion , the entire elpc-1 coding region together with 435 base pairs of DNA upstream of the start ATG was amplified by using primers 5′-AAAAGCATGCTCCGGTACGGTATGTGGC-3′ and 5′-AAAACTGCAGTGGGAAAACTGAAG CAAATGAA-3′ . The PCR product was subcloned into pPD95 . 77 GFP expression vector between PstI and SphI sites . A Leica DMRB microscope equipped with both Nomarski differential interference contrast and epifluorescence optics was used to view worms at high magnification . Images were captured with a Deltpix CCD camera and software ( Deltapix , Copenhagen ) . Confocal microscopy was performed on a Leica TCS SP2 confocal microscope . Confocal images were captured using Leica confocal software . Techniques described by Gaur et al . ( 2007 ) were used with minor modifications to isolate and analyze tRNA from C . elegans worms . For each strain , worms from twenty 9 cm culture plates containing mixed-stage populations of worms were used . After extensive washing with M9 buffer , the worm pellets were frozen in the liquid nitrogen and then thawed in the presence of 0 . 5 volumes of TRIzol ( Invitrogen ) . A tissue-grinder ( Kontes ) was used to break open the worms . After extraction of the lysate with chloroform , followed by addition of isopropanol , total RNA was sedimented by centrifugation . tRNA was separated from other types of RNA by using methods described previously [79] . Purified tRNA was digested with Nuclease P1 for 16 h at 37°C and then treated with bacterial alkaline phosphatase for 2 h at 37° . The hydrolysate was analyzed by high pressure liquid chromatography with a Develosil C-30 reverse-phase column as described [79] . ncm5U , mcm5U , mcm5s2U and mcm5Um have all been found on wobble uridines in S . cerevisiae tRNA . We did not examine C . elegans tRNA for the presence of ncm5Um because P1 and BAP cannot digest the dinucleotide ncm5UmpX to nucleosides [80] . Total RNA was extracted with the aid of a BIO-RAD Aurum total RNA mini kit according to the instruction manual . Real-time PCR was carried out in 25 µl reaction mixes . iScript one-step RT-PCR kit with SYBR green ( BIO-RAD ) and the iCycler iQ Real-Time PCR Detection System ( BIO-RAD ) were used . The data were normalized to tbb-2 and ubc-2 mRNA levels . Six independent assays were performed for each strain analyzed . For each strain analyzed , one 6-cm plate containing a population of well-fed worms was subjected to a 2 h heat shock at 33° . The worms were washed from the plate with M9 salt solution , sedimented , washed once in M9 and then once in breaking buffer ( 100 mM Tris-HCl , 1 mM DTT , 20% glycerol ) . After resuspension in 250 µl of breaking buffer containing Roche protease inhibitor cocktail , the worms were broken open by sonication . Five 2 sec pulses at maximum effect were used . The extracts were transferred to microcentrifuge tubes and worm debris was sedimented by centrifugation at 13 , 000 rpm for 15 min . β-galactosidase activity in the cleared extracts was measured using standard protocols [81] . The assay was performed as described in detail by Kourtis and Tavernarakis [41] . Worms carrying the gcy-5::gfp , mec-4::gfp or myo-3::gfp reporters were mounted on the agarose pad in the presence of levamisol and photobleached with light from an HBO 103W/2 mercury lamp ( OSRAM ) . A 63× objective was used for photobleaching gcy-5::gfp and mec-4::gfp strains , a 20× objective for myo-3::gfp strains . Salt chemotaxis assays were performed as described by Ward [82] and Bargmann and Horvitz [44] . All the assays were carried out at room temperature ( ca . 21 . 5°C ) on 9 cm agar plates containing 5 mM KH2PO4 pH 6 . 0 , 1 mM CaCl2 , 1 mM MgSO4 and 2% agar . N2 , elpc-1 ( tm2149 ) , elpc-3 ( tm3120 ) and tuc-1 ( 1297 ) strains were maintained at 25°C for at least three generations prior to being assayed . The salt gradient with a peak 0 . 5 cm from one edge of the plate was formed overnight by placing a block of agar measuring approximately 5 mm in each dimension and containing 100 mM NaCl , 5 mM KH2PO4 pH 6 . 0 , 1 mM CaCl2 , 1 mM MgSO4 and 2% agar . In each single test , 80–100 young adult worms were washed three times in 5 mM KH2PO4 pH 6 . 0 , 1 mM CaCl2 , 1 mM MgSO4 and then placed in the center of the assay plates . Before the worms were placed on the assay plate , 1 µl of 0 . 5 M sodium azide was spotted both at the salt gradient peak and at the opposite side of the plate to capture the worms moving to those areas . The numbers of worms at different positions on the plate were counted every 10 min after the start of the assay . The formula was used to calculate the chemotaxis index . In this equation , A was the number of worms at the attractant area , C the number of worms at the control spot , and N the total number of worms placed on the plates . Each experiment was repeated at least 4 times . For chemotaxis assays with isoamyl alcohol , the odorant was dropped on the assay plate immediately prior to the addition of worms to the plate . The assay was performed as described [43] , with minor modifications . For each assay , adult worms were washed off the culture plates with chemotaxis washing buffer ( 5 mM KH2PO4 pH 6 . 0 , 1 mM CaCl2 , 1 mM MgSO4 ) and then washed three times in the same buffer . For the naive condition , worms were washed and then assayed immediately without further incubation . The other worms were conditioned respectively on nematode growth medium ( NGM ) plates containing 100 mM NaCl , or on NaCl-free NGM plates for 4 hours . After conditioning , worms were collected again and placed on the assay plates . After 30 min , the numbers of worms in the NaCl spot ( A ) and the control region ( C ) were counted . The index was calculated using the formula , . ANF::GFP levels were measured by western blotting using an anti-GFP antibody ( Clontech , JL-8 ) . 50 L4 larvae of each genotype were collected , boiled in SDS sample buffer for 5 min and loaded onto a 10% SDS PAGE . Quantification of imaging pixel intensity was performed by NIH image J . To measure acetylated α-tubulin levels by western blot , protein was extracted from young adult worms . To avoid protein degradation , worms were suspended in ice-cold extraction buffer containing proteinase inhibitors and rapidly frozen in liquid nitrogen . The frozen pellets were ground to a powder in a mortar . 20 µg protein was loaded on the gel in each lane . The dilution of anti-lys40-acetylated-α-tubulin antibody ( abcam , 6-11B-1 ) was 1∶1000 , and of anti-α-tubulin antibody ( Sigma , B-5-1-2 ) was 1∶2000 . The assays were performed as described by Mahoney et al . [49] . 25–30 worms were used for each genotype . The assay was performed blind in triplicate at room temperature ( ca . 21 . 5°C ) . The worms were cultivated at 25°C prior to being assayed . The assay was performed as described by Speese et al . [47] . Fluorescence confocal micrographs were made of coelomocytes . The intensity of GFP fluorescence in captured images in grey scale was measured with the aid of the NIH ImageJ software .
|
The efficiency of protein synthesis can be modulated by alterations of various components of the translation machinery . In translation , transfer RNAs act as adapter molecules that decode mRNA into protein and thereby play a central role in gene expression . In the tRNA maturation process , a subset of the normal nucleosides undergoes modifications . Modified nucleosides in the tRNA anticodon region are important for efficient translation . We found that , in the worm C . elegans , components of the Elongator complex are required for the formation of a certain set of tRNA modifications in the anticodon region . We observed a reduced efficiency of translation as well as a lower production of neurotransmitters in Elongator mutant worms . Elongator is conserved in eukaryotes , and mutations in a subunit of human Elongator cause a severe neurodegenerative disease , familial dysautonomia ( FD ) . It is unclear in humans whether Elongator acts on the translational level through tRNA modification to regulate neuronal processes . Our observations in C . elegans , together with the role of yeast Elongator in translation , show that the function of Elongator in tRNA modification is conserved . Inactivation of Elongator may cause neuronal defects by affecting translation .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"neuroscience/behavioral",
"neuroscience",
"genetics",
"and",
"genomics/animal",
"genetics",
"biochemistry/rna",
"structure",
"genetics",
"and",
"genomics/gene",
"expression",
"developmental",
"biology/cell",
"differentiation",
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/gene",
"function",
"developmental",
"biology/neurodevelopment",
"biochemistry/transcription",
"and",
"translation",
"neuroscience/neurobiology",
"of",
"disease",
"and",
"regeneration",
"developmental",
"biology/developmental",
"molecular",
"mechanisms"
] |
2009
|
Defects in tRNA Modification Associated with Neurological and Developmental Dysfunctions in Caenorhabditis elegans Elongator Mutants
|
Infections with schistosomes and soil-transmitted helminths exert a considerable yet underappreciated economic and public health burden on afflicted populations . Accurate diagnosis is crucial for patient management , drug efficacy evaluations , and monitoring of large-scale community-based control programs . The diagnostic accuracy of four copromicroscopic techniques ( i . e . , Kato-Katz , Koga agar plate , ether-concentration , and FLOTAC ) for the detection of Schistosoma mansoni and soil-transmitted helminth eggs was compared using stool samples from 112 school children in Côte d'Ivoire . Combined results of all four methods served as a diagnostic ‘gold’ standard and revealed prevalences of S . mansoni , hookworm , Trichuris trichiura , Strongyloides stercoralis and Ascaris lumbricoides of 83 . 0% , 55 . 4% , 40 . 2% , 33 . 9% and 28 . 6% , respectively . A single FLOTAC from stool samples preserved in sodium acetate-acetic acid-formalin for 30 or 83 days showed a higher sensitivity for S . mansoni diagnosis ( 91 . 4% ) than the ether-concentration method on stool samples preserved for 40 days ( 85 . 0% ) or triplicate Kato-Katz using fresh stool samples ( 77 . 4% ) . Moreover , a single FLOTAC detected hookworm , A . lumbricoides and T . trichiura infections with a higher sensitivity than any of the other methods used , but resulted in lower egg counts . The Koga agar plate method was the most accurate diagnostic assay for S . stercoralis . We have shown that the FLOTAC method holds promise for the diagnosis of S . mansoni . Moreover , our study confirms that FLOTAC is a sensitive technique for detection of common soil-transmitted helminths . For the diagnosis of S . stercoralis , the Koga agar plate method remains the method of choice .
Schistosomiasis and soil-transmitted helminthiasis are widespread in many parts of the developing world where they negatively impact on human health and wellbeing , and thus exacerbate poverty [1]–[4] . Schistosomiasis is caused by blood flukes of the genus Schistosoma; in Africa both S . mansoni and S . haematobium are endemic [2] , [5]–[7] . Soil-transmitted helminthiasis is caused by intestinal nematodes; hookworms ( Ancylostoma duodenale and Necator americanus ) , the roundworm ( Ascaris lumbricoides ) , the whipworm ( Trichuris trichiura ) , and the threadworm ( Strongyloides stercoralis ) [1] , [3] , [8] , [9] . A detailed understanding of the epidemiology of these parasitic worm infections is important for the design , implementation , monitoring , and evaluation of helminth control programs [10] , [11] . Commonly used diagnostic methods for these parasites rely on the detection of helminth eggs or larvae in human stool . These copromicroscopic approaches have drawbacks , such as low sensitivity for the detection of light-intensity infections [12] , [13] . At present , the Kato-Katz technique is the most widely used copromicroscopic method in epidemiological surveys pertaining to human intestinal helminth infections because of its simplicity [14] , low cost , and the established system to stratify infection intensity into different classes based on cut-offs of egg-counts [15] . The small amount of stool analyzed ( usually 41 . 7 mg ) explains why the Kato-Katz technique has a low sensitivity to detect eggs whenever they are present at low frequency or appear highly clustered ( theoretical analytic sensitivity = 24 eggs per gram of stool ( EPG ) ) [16] , [17] . The sensitivity can be increased by examining multiple Kato-Katz thick smears prepared from the same stool sample or , better yet , from multiple stool samples [12] , [18]–[22] . This point highlights the appeal of parasitological diagnostic methods capable of screening a larger amount of stool , e . g . , 0 . 5 g or even 1 g in the case of the ether-concentration method [23] or the FLOTAC technique [24] . The ether-concentration method is often used for the diagnosis of helminth infections , particularly in specialized laboratories [23] , [25]–[27] . Importantly , it allows the concurrent diagnosis of intestinal protozoa , and is sometimes used in combination with the Kato-Katz method to enhance diagnostic sensitivity for helminths , and hence to deepen our understanding of polyparasitism [28]–[30] . An important feature of the ether-concentration method is that it uses preserved stool samples , fixed in either sodium acetate-acetic acid-formalin ( SAF ) [23] , or diluted formalin [24] , thus allowing sample storage and analysis at later time points . However , considerable inter-laboratory discrepancies have been noted for helminth and particularly intestinal protozoa diagnosis [26] , . The Koga agar plate technique allows direct observation of S . stercoralis and hookworm larvae hatched from fresh stool samples incubated on nutrient agar in a humid chamber , usually after 48 hours [31] . Recent studies suggest that the FLOTAC technique [24] holds promise for the diagnosis of soil-transmitted helminth infections in humans [32]–[34] . Its potential for the diagnosis of S . mansoni has yet to be investigated . The FLOTAC technique takes advantage of the fact that during flotation , parasitic elements such as helminth eggs gather in the apical portion of the flotation column and can be readily translated , i . e . , cut transversally for subsequent viewing under a microscope . Thus , parasitic elements are separated from fecal debris , facilitating their identification and enumeration . Protocols have been developed for the FLOTAC basic technique ( theoretical analytic sensitivity = 1 EPG ) , the FLOTAC dual technique , the FLOTAC double technique , and the FLOTAC pellet technique ( all: theoretical analytic sensitivity = 2 EPG ) [24] . Importantly , in a single FLOTAC examination usually ∼1 g of stool is analyzed , and hence a single FLOTAC allows a 24-fold higher amount of stool to be examined than a single Kato-Katz thick smear , which is an important factor explaining the higher sensitivity of the FLOTAC technique . The aim of this study was to compare the diagnostic accuracy of different techniques and sampling efforts , i . e . , single and multiple Kato-Katz thick smears , ether-concentration and the FLOTAC method , for the detection and quantification of helminth eggs . The performance of the Koga agar plate technique for the detection of helminth larvae was also assessed . Particular emphasis was placed on the diagnosis of S . mansoni .
The study was cleared by the institutional research commissions of the Swiss Tropical and Public Health Institute ( Basel , Switzerland ) and the Centre Suisse de Recherches Scientifiques ( CSRS; Abidjan , Côte d'Ivoire ) , and was approved by local and national health authorities of Côte d'Ivoire . The school director and teachers were informed about the objectives and procedures of the study . Parents , legal guardians , and children were informed about the study and sufficient time was given to ask questions . Written informed consent was obtained from the parents or legal guardians of all participating children . Participation was voluntary and children could withdraw from the study at any time without further obligations . At the end of the study , all children attending the school of Azaguié-IRFA ( “Institut de Recherches sur les Fruits et Agrumes” ) were treated free of charge with praziquantel ( single 40 mg/kg oral dose using a ‘dose-pole’ ) and mebendazole ( single 500 mg oral dose ) according to WHO recommendations [35] . Additionally , children infected with S . stercoralis were treated with ivermectin ( single 200 µg/kg oral dose ) . The study was carried out in June 2008 in the primary school of Azaguié-IRFA , located in a rural setting of Azaguié in the region of Agboville , south Côte d'Ivoire ( geographical coordinates: 05°36′10 . 5″ N latitude , 04°00′58 . 5″ W longitude ) . The village is located 56 km north of Abidjan , the economic capital of Côte d'Ivoire . In the school year 2007/2008 , 200 children attended grades 1–6 . We aimed for a sample size of 120 school children , similar to our preceding FLOTAC research carried out elsewhere in Côte d'Ivoire [32] . Allowing for a drop out of ∼10% , we randomly selected 133 school children from all grades and invited them to submit a single fresh morning stool . School children were given plastic containers ( 125 ml ) for stool collection . Upon submission , each container was labeled with a unique identification number and transferred within 1–2 hours to the CSRS in Abidjan . Stool samples were processed at the day of collection , usually within 2–3 hours after reaching the laboratory . Stool specimens were collected over six consecutive days ( one sample from each of 12 children on day 1 and one sample from each of 20 children on the following 5 days ) . As depicted in Figure 1 , each fresh stool sample was subjected to the Kato-Katz method ( 3 thick smears ) , the Koga agar plate method ( 1 examination ) , and the FLOTAC dual technique ( 1 examination , fresh stool sample homogenized in SAF ) . Additionally , ∼5 g of the fresh stool were preserved in 25 ml SAF . After 9 days of preservation , these samples were filtered in order to remove large fibers ( wire mesh aperture: 250 µm ) and split into 5 sub-samples , each weighing ∼1 g . Among them , 3 sub-samples were examined with the FLOTAC dual technique 10 , 30 , and 83 days post-stool collection , 1 sub-sample was subjected to the ether-concentration method 40 days post-stool collection , and the fifth sub-sample was used as back-up and was finally discarded . Processing of stool samples was as follows . First , triplicate Kato-Katz thick smears ( 41 . 7 mg each ) were prepared . Slides were read twice; first within 30–60 min for detection of hookworm eggs and again after ∼2 hours for diagnosis of S . mansoni , A . lumbricoides , T . trichiura , and other helminths . For each helminth species , the number of eggs was counted under a microscope by one of four experienced laboratory technicians and recorded separately for both readings . Second , a hazelnut-sized stool sample ( average weight = 2 . 29 g ) was placed in the middle of an agar plate . The plates were incubated in a humid chamber at 28°C for 48 hours . Plates were inspected under a microscope for the characteristic traces of hookworm and S . stercoralis larvae . Subsequently , the plates were rinsed with SAF , the solutions centrifuged and the sediment examined for larvae under a microscope by one experienced laboratory technician . The technician was blinded to the preceding Kato-Katz test results . Third , the samples for FLOTAC were prepared as follows: the filtered stool suspensions were filled up to 10 ml with SAF and equally split into two 15-ml Falcon tubes . Three ml of ether were added to each tube to facilitate the removal of fatty compounds that could interfere with egg detection under a microscope . The Falcon tubes were shaken rigorously for at least 1 min and then centrifuged for 2 min at 170 g . The supernatant was discarded and each pellet suspended in 6 ml of either flotation solution FS4 ( sodium nitrate: 315 g NaNO3 suspended in 685 ml H2O; specific gravity ( s . g . ) = 1 . 20 ) or FS7 ( zinc sulphate: 685 g ZnSO4 H2O suspended in 685 ml H2O; s . g . = 1 . 35 ) [24] . In the present study , the FLOTAC dual technique was employed ( i . e . , one of two different FS in each of the two chambers of the same FLOTAC apparatus ) . From a panel of 14 different FS [36] , we selected FS4 because it had proved to be useful for soil-transmitted helminth diagnosis in previous studies carried out in Côte d'Ivoire and Zanzibar [32] , [33] . FS7 was chosen because it was particularly suitable for the detection of S . mansoni eggs in fecal samples obtained from infected mice and hamster ( J . Keiser , L . Rinaldi , and J . Utzinger; unpublished data ) . Each suspension was transferred into one of the two 5-ml chambers of the FLOTAC apparatus . The apparatus was centrifuged ( using a Hettich Universal 320 centrifuge; Tuttlingen , Germany ) at 160 g for 5 min . Subsequently , the upper portion in the FLOTAC apparatus was translated , and the observation grid examined under a microscope at 100× magnification by one experienced laboratory technician . The technician was blinded to the preceding Kato-Katz and ether-concentration test results . Helminth eggs were counted and reported for each species in each chamber separately . Fourth , the ether-concentration method was performed as described in detail elsewhere [27] . Data were double-entered in Microsoft Office Access 2007 , cross-checked using EpiInfo ( TM 3 . 4 . 1 ) , and analyzed with STATA version 9 . 1 ( StataCorp LP; College Station , TX ) . The different diagnostic methods were compared using 2-way contingency tables of frequencies , and agreement between 2 diagnostic techniques was determined ( e . g . , Kato-Katz versus ether-concentration , and Kato-Katz versus FLOTAC for S . mansoni diagnosis ) . Kappa ( ĸ ) statistic was employed to determine the strength of agreement using the following cut-offs: ( i ) ĸ<0 . 00 indicating no agreement; ( ii ) ĸ = 0 . 00–0 . 20 indicating poor agreement; ( iii ) ĸ = 0 . 21–0 . 40 indicating fair agreement; ( iv ) ĸ = 0 . 41–0 . 60 indicating moderate agreement; ĸ = 0 . 61–0 . 80 indicating substantial agreement; and ( v ) ĸ = 0 . 81–1 . 00 indicating almost perfect agreement [37] , [38] . It is important to note that ĸ values depend on the marginal distributions of contingency tables . We employed a test of marginal homogeneity and , since significant values were obtained , raked ĸ values were calculated throughout [39] , [40] . The sensitivity and negative predictive value ( NPV ) were calculated for each method , considering the combined results from all different methods , FS , and at all time points investigated , as diagnostic ‘gold’ standard . Hence , any positive test result , regardless of the technique employed , was considered a true-positive result . Following this approach specificity was set to be 100% , justified by the direct observation of parasite eggs [21] , [26] , [32] , [33] . Helminth-specific egg counts were expressed as EPGs for three among the four techniques . For the Kato-Katz method , the number of helminth eggs in each of the three thick smears prepared from a single fresh stool sample collected from each participant were added and then multiplied by a factor 8 to obtain EPGs . For the ether-concentration method , EPGs were estimated by dividing the number of helminth eggs with the amount of stool examined after 40 days of preservation ( ∼1 g; see Figure 1 ) , that is one-fifth of the total amount of weighed stool that was preserved and used for four different tests with the remaining fifth part of the preserved stool sample finally discarded . For the examination of stool samples with the FLOTAC dual technique , FS-specific EPGs were estimated as follows . For the fresh stool sample ( ∼1 g; see Figure 1 ) , the number of counted helminth eggs was divided by the exact amount of stool and multiplied with a factor 2 ( stool sample was equally aliquoted to one of the two FLOTAC chambers ) and , additionally , with a factor 1 . 2 ( stool sample was solved in 6 ml FS , but only 5 ml was filled into the FLOTAC apparatus ) . For the SAF-preserved stool , the number of counted helminth eggs was divided by one-fifth of the total amount of weighed stool ( ∼1 g; see Figure 1 ) , and then multiplied by factors 2 and 1 . 2 as described above . Since EPGs were not normally distributed ( as assessed by quantile normal plots ) , the geometric mean ( GM ) , including 95% confidence intervals ( CI ) , was calculated and graphically displayed . Our assumption was that non-overlapping 95% CIs indicate statistical significance ( p<0 . 05 ) . A post-calibration was performed 4 months post-stool collection to determine whether FS4 and FS7 are indeed among the most suitable FS for the FLOTAC technique for differential helminth diagnosis . For this purpose , we employed a composite human stool sample ( ∼100 g ) , pooling stool from 8 selected children with high-intensity helminth infections who provided a second stool sample just before administering anthelmintic drugs . This composite stool sample was preserved in SAF and transferred to Naples , Italy . From the panel of 14 different available FS [36] , a total of 9 were selected , including FS4 and FS7 . At least 3 replicates were examined for each of the 9 FS with and without a prior ether washing step , to allow an appraisal of the effect of ether on helminth diagnosis .
We obtained one sufficiently large stool sample to perform triplicate Kato-Katz , multiple FLOTAC , a single ether-concentration , and a single Koga agar plate test from 112 school children . Our study cohort comprised 61 ( 54 . 5% ) boys . The median age was 10 years ( range: 6–15 years ) . The combined results from the different copromicroscopic techniques were considered as diagnostic ‘gold’ standard . As shown in Figure 2 , eggs of S . mansoni were detected in 93 children ( 83 . 0% ) . The overall prevalences of hookworm , T . trichiura and A . lumbricoides were 55 . 4% , 40 . 2% and 28 . 6% , respectively . S . stercoralis larvae were detected in the stools of 38 children ( 33 . 9% ) . Table 1 summarizes the characteristics of 3 different techniques for S . mansoni diagnosis . Whilst a single Kato-Katz thick smear revealed S . mansoni at a prevalence of 56 . 3% , the cumulative prevalence after examination of 3 Kato-Katz thick smears was 64 . 3%; an increase of 14 . 2% . The observed S . mansoni prevalence based on a single ether-concentration test was 70 . 5% . Subjecting a fresh stool sample homogenized in SAF to the FLOTAC dual technique , but considering only results from FS7 , revealed a S . mansoni prevalence of 53 . 6% . Preservation of stool samples in SAF for 10 , 30 , and 83 days , and examination with FLOTAC FS7 revealed point prevalence estimates of 72 . 3–75 . 9% . The highest sensitivity for S . mansoni diagnosis ( 91 . 4% ) was found for the FLOTAC dual technique after 30 and 83 days of preservation in SAF . The sensitivity of a single ether-concentration test was 85 . 0% , whereas triplicate or only a single Kato-Katz revealed sensitivities of 77 . 4% and 67 . 7% , respectively . Table 2 shows 2-way contingency tables comparing the results of a single FLOTAC ( fresh stool , SAF preservation for 10 , 30 , or 83 days ) and the single ether-concentration test ( SAF preservation for 40 days ) with the combined results of the triplicate Kato-Katz thick smear readings . Using fresh stool samples for FLOTAC ( homogenized in SAF ) and Kato-Katz , 51 S . mansoni infections were concurrently detected by both techniques , whereas 21 additional infections were diagnosed by triplicate Kato-Katz only , and 9 additional infections were detected by FLOTAC only . The agreement between these 2 methods was moderate ( raked ĸ = 0 . 49 ) . Comparing the results of FLOTAC performed with stool samples preserved in SAF for 10 , 30 , or 83 days with triplicate Kato-Katz from fresh stool , both methods concurrently detected between 69 and 71 S . mansoni infections , whereas 12–16 additional infections were only found by the FLOTAC technique and 1–3 infections were only detected by triplicate Kato-Katz thick smears . Substantial agreement was found for the stool samples preserved in SAF for 10 days ( raked ĸ = 0 . 76 ) or 83 days ( raked ĸ = 0 . 71 ) , and an almost perfect agreement after 30 days of SAF preservation ( raked ĸ = 0 . 84 ) . The ether-concentration method and triplicate Kato-Katz diagnosed 69 S . mansoni infections concurrently , whereas 10 infections were only found by the ether-concentration method and 3 were detected by triplicate Kato-Katz thick smears only , owing to a substantial agreement ( raked ĸ = 0 . 79 ) . Triplicate Kato-Katz thick smears revealed a mean infection intensity of 121 . 2 EPG ( 95% CI: 86 . 8–169 . 2 EPG ) . The mean infection intensity based on a single ether-concentration examination was 110 . 7 EPG ( 95% CI: 76 . 0–161 . 1 EPG ) . These mean egg counts were significantly higher than those obtained with the FLOTAC dual technique , regardless of whether FS4 or FS7 were employed ( Figure 3A ) . There was one exception: a single FLOTAC performed on stool samples preserved in SAF for 83 days and using FS7 revealed similar EPGs as triplicate Kato-Katz thick smears and a single ether-concentration test . Using FS4 with stool preserved in SAF for 10 days resulted in the detection of only 10 ( 8 . 9% ) , and after 83 days of preservation of no S . mansoni-positive individuals . The use of FS7 , on the other hand , revealed 81 ( 72 . 3% ) and 85 ( 75 . 9% ) infections after 10 and 83 days of preservation , respectively . While the observed S . mansoni fecal egg counts in FS4 decreased over time to zero , an increase was observed in FS7 from a mean of 32 . 3 EPG ( 95% CI: 24 . 7–42 . 3 EPG ) at day 10 to 57 . 7 EPG ( 95% CI: 41 . 5–80 . 2 EPG ) at day 83 . Figure 4 shows that the shape of S . mansoni eggs was somewhat altered when using the FLOTAC dual technique and employing FS7 . However , the characteristic lateral spine remained , and hence unambiguous diagnosis was ascertained . For hookworm , the observed prevalence increased from 12 . 5% after a single to 21 . 4% after triplicate Kato-Katz examination , an increase of 71 . 2% . For T . trichiura there was an increase from 8 . 0% to 12 . 5% ( +56 . 3% ) . No increase was observed for A . lumbricoides; a prevalence of 19 . 6% was obtained already after the first Kato-Katz thick smear reading ( Figure 2B–D and Table 1 ) . Analyses with the Koga agar plate method revealed 28 ( 25 . 0% ) hookworm infections . The highest observed prevalence of hookworm infection was obtained with the FLOTAC dual technique using fresh stool samples homogenized in SAF ( 43 . 8% ) . The observed prevalence at days 10 and 83 post-conservation decreased to 31 . 3% and 17 . 0% , respectively . A single Kato-Katz revealed a hookworm infection intensity of 165 . 7 EPG ( 95% CI: 97 . 0–282 . 4 EPG ) , whereas triplicate Kato-Katz thick smears suggested an intensity of less than half of this value ( 64 . 6 EPG; 95% CI: 32 . 8–127 . 2 EPG ) . The difference in fecal egg counts determined by the Kato-Katz and the FLOTAC techniques in both FS ( fresh stool: FS4 = 18 . 1 EPG; 95% CI: 11 . 3–29 . 2 EPG , and FS7 = 18 . 7 EPG; 95% CI: 11 . 3–30 . 9 EPG ) was significant . In the preserved stool samples , the observed EPGs for hookworm decreased significantly in FS4 from day 10 ( 13 . 3 EPG; 95% CI: 7 . 4–23 . 8 EPG ) to day 83 ( 2 . 7 EPG; 95% CI: 1 . 2–5 . 9 EPG ) ( Figure 3B ) . In FS7 , the fecal egg counts also decreased , but the difference showed no statistical significance after 10 and 83 days of SAF conservation ( from 18 . 1 EPG ( 95% CI: 11 . 5–28 . 6 EPG ) to 11 . 1 EPG ( 95% CI: 6 . 1–20 . 2 EPG ) ) . The highest prevalence of A . lumbricoides was estimated by a single FLOTAC from fresh stool homogenized in SAF ( 20 . 5% ) , but triplicate Kato-Katz showed only a marginally lower prevalence ( 19 . 6% ) . The examination of stool samples preserved in SAF for 83 days using FLOTAC or an ether-concentration test at day 40 after stool collection resulted in observed prevalences of 10 . 7% and 12 . 5% , respectively . The A . lumbricoides fecal egg counts determined with Kato-Katz were significantly higher than those obtained with FLOTAC , except for the results generated after 83 days of stool preservation . There was a significant increase for A . lumbricoides egg counts both with FS4 ( from 148 . 4 EPG ( 95% CI: 39 . 9–551 . 5 EPG ) to 1159 . 3 EPG ( 95% CI: 596 . 6–2252 . 7 EPG ) ; a 7 . 8-fold increase ) and with FS7 ( from 181 . 2 EPG ( 95% CI: 60 . 0–546 . 9 EPG ) to 1688 . 5 EPG ( 95% CI: 889 . 4–3205 . 8 EPG ) ; a 9 . 3–fold increase ) when comparing the 10 and 83 post-stool collection preservation time points ( Figure 3C ) . The highest observed T . trichiura prevalence ( 32 . 1% ) was obtained with a single FLOTAC examined after 30 days of stool preservation in SAF . Considerably lower prevalences were obtained with a single ether-concentration test ( 20 . 5% ) and triplicate Kato-Katz thick smears ( 12 . 5% ) . The T . trichiura fecal egg count for a single Kato-Katz thick smear was 56 . 3 EPG ( 95% CI: 22 . 4–141 . 4 EPG ) . For triplicate Kato-Katz , the respective egg count was 24 . 0 EPG ( 95% CI: 11 . 4–50 . 6 EPG ) . Consistently more T . trichiura eggs were found in FS4 than in FS7 , but the difference was only significant at day 30 post-preservation ( FS4 = 35 . 1 EPG , 95% CI: 24 . 4–50 . 4 EPG; FS7 = 15 . 8 EPG , 95% CI: 10 . 2–24 . 4 EPG ) . There was an apparent increase of egg counts over the course of stool preservation in SAF , i . e . , in FS4 from 17 . 5 EPG ( 95% CI: 11 . 2–27 . 2 EPG ) to 39 . 6 EPG ( 95% CI: 25 . 8–60 . 8 EPG ) , a 2 . 3-fold increase , and in FS7 from 11 . 8 EPG ( 95% CI: 7 . 9–17 . 5 EPG ) to 22 . 2 EPG ( 95% CI: 14 . 3–34 . 7 EPG ) , a 1 . 9-fold increase ( Figure 3D ) . S . stercoralis larvae were found on Koga agar plates prepared with stool samples from 38 children but only 1 of these 38 infection was diagnosed after triplicate Kato-Katz examinations and 2 of these 38 infections were detected with FLOTAC in the fresh stool samples processed in SAF . No S . stercoralis larvae were found in the preserved samples , neither by FLOTAC nor by ether-concentration . The use of FS7 resulted in the highest S . mansoni fecal egg count ( average: 38 . 3 EPG , SD: 2 . 7 EPG; 6 replications ) , but stool samples had to be washed with ether as otherwise , due to the darkening effect of organic debris , accurate reading was not feasible . For hookworm diagnosis , FS4 produced the highest fecal egg count ( average: 103 . 2 EPG , SD: 23 . 6 EPG; 6 replications ) . Of note , hookworm eggs were readily detected only in the absence of ether for sample preparation . With regard to A . lumbricoides and T . trichiura , the results from the post-calibration were less clear-cut than those for S . mansoni and hookworm , as all tested FS resulted in relatively high egg count averages for A . lumbricoides ( 306 . 0–515 . 0 EPG , SD: 22 . 2–176 . 8 EPG; 3–4 replications ) and T . trichiura ( 6 . 0–26 . 0 EPG , SD: 2 . 8–15 . 0 EPG; 3–4 replications ) , whenever an ether washing step was included .
Accurate diagnosis is key for adequate patient management and for guiding the design , implementation , and monitoring of community-based infectious disease control programs [13] , [41] . We compared the diagnostic accuracy of two widely used techniques for detection and quantification of helminth eggs in fecal samples – the Kato-Katz thick smear using fresh stool , and the ether-concentration method using SAF-preserved samples – with the recently developed FLOTAC technique . Particular emphasis was placed on the diagnosis of S . mansoni because of the public health importance of intestinal schistosomiasis [2] , [5]–[7] , and because the FLOTAC method had not previously been investigated for this parasite . Additionally , the Koga agar plate method was employed , mainly for the diagnosis of S . stercoralis , but also for the detection of hookworm larvae . Best results , i . e . , high sensitivities ( 87 . 1–91 . 4% ) for detecting S . mansoni eggs ( prevalence: 72 . 3–75 . 9% ) , were achieved after stool samples were homogenized , preserved in SAF , and examined after 10–83 days with the FLOTAC dual technique . Almost as sensitive was a single ether-concentration ( 85 . 0% ) , revealing a S . mansoni prevalence of 70 . 5% . Triplicate Kato-Katz examinations resulted in a prevalence estimate of 64 . 3% . A single Kato-Katz and FLOTAC using fresh stool revealed considerably lower S . mansoni point prevalences of 56 . 3% and 53 . 6% , respectively . It should be noted , however that neither FLOTAC , nor the ether-concentration test , nor multiple Kato-Katz readings detected ‘all’ S . mansoni or ‘all’ soil-transmitted helminth infections . There was moderate to almost perfect agreement between the FLOTAC or ether-concentration method and triplicate Kato-Katz thick smears according to raked ĸ values . With regard to soil-transmitted helminth diagnosis , our results confirm that a single FLOTAC is more sensitive than multiple Kato-Katz thick smears for the detection of hookworm , A . lumbricoides , and T . trichiura eggs in fecal samples [32]–[34] . A single FLOTAC using fresh stool processed with SAF was also more sensitive than a single ether-concentration test for the diagnosis of hookworm , A . lumbricoides , and T . trichiura infections . Finally , for hookworm diagnosis , a single FLOTAC using fresh stool was more sensitive than a single Koga agar plate test . The design of our study allowed investigating the effect of the duration of stool preservation on helminth species-specific diagnosis . The duration of stool fixation had a considerable effect on the diagnostic performance of copromicroscopic techniques . While the number of S . mansoni infections detected after 10 , 30 , and 83 days of stool preservation in SAF remained constant ( 81–85 infections ) , there was an apparent increase in fecal egg counts from day 10 to day 83 , from 32 . 3 EPG to 57 . 7 EPG ( considering only FS7 ) . The observed prevalence of hookworm and A . lumbricoides decreased with increasing duration of SAF conservation before FLOTAC analysis . For example , while the point prevalence of hookworm was 43 . 8% for FLOTAC using fresh stool , it decreased to 17 . 0% after stool samples had been preserved in SAF for 83 days . On the other hand , there was no apparent decline in the prevalence of T . trichiura over the 83-day SAF preservation period . Higher fecal egg counts were observed for A . lumbricoides and T . trichiura as a function of stool preservation duration . Regarding hookworm diagnosis , there was a sharp decrease in fecal egg counts as a function of preservation time using SAF , suggesting a negative impact of this preservation medium on hookworm eggs . However , during the post-calibration investigation and subsequent studies , it was found that the introduction of an ether washing step resulted in lower hookworm egg counts . Indeed , considerably higher hookworm egg counts were revealed in SAF-preserved stool samples in the absence of ether , whereas destroyed hookworm eggs could be observed after exposure of the sample to ether . These observations indicate that the prolonged preservation of stool in SAF in combination with ether used for sample preparation might destroy the fragile hookworm eggs . To test this hypothesis , it will be interesting to investigate the exact influence of the preservation media alone , i . e . , we still lack data on the influence of SAF and/or ether on helminth egg counts . The underlying mechanisms resulting in increasing S . mansoni , A . lumbricoides , and T . trichiura fecal egg counts estimates with time of preservation , and for the higher diagnostic sensitivity , yet lower egg counts when using FLOTAC as opposed to the Kato-Katz thick smear method , remain elusive and are the subject of ongoing deliberations and studies . We speculate that the helminth larvae are able to further develop and gain weight in these environmentally resistant eggs . This might lead to a change in density , and hence altered floating behavior of the eggs . However , these apparent fluctuations give rise to fears regarding the consistency of the diagnostic performance of FLOTAC when performed after non-standardized preservation time , on different populations , and by different laboratories . The somewhat higher fecal egg counts of S . mansoni , A . lumbricoides , and T . trichiura using FLOTAC at later time points of stool conservation , and the consistently higher fecal egg counts using Kato-Katz might be explained by the following additional reasons . First , helminth eggs should not be considered “inert elements” . Instead , interactions occur between the different compartments within a floating fecal suspension ( e . g . , FS components , parasitic elements , fixative , ether , and residues of the host alimentation ) , and these might be complex [24] . New research is therefore needed to elucidate potential interactions between these compartments . Second , the high fecal egg counts derived from the Kato-Katz thick smear readings shown in Fig . 3 might be misleading . EPG values obtained from Kato-Katz thick smear readings are not continuous due to the multiplication factor used ( i . e . , a factor 24 for a single , and a factor 8 for triplicate Kato-Katz thick smear readings ) . Hence , the minimum positive value for a single measurement is 24 EPG . Third , there are additional reasons why the Kato-Katz technique might overestimate fecal egg counts . For example , when scraping the plastic spatula of the Kato-Katz kit across the upper surface of the fine-meshed screen placed on top of the stool sample , the feces is sieved , and helminth eggs are concentrated [42] , [43] , an issue we are currently investigating . The available results pose considerable challenges for articulating recommendations regarding the optimal deployment of diagnostic tools for patient diagnosis , drug efficacy evaluations , and surveillance in areas where soil-transmitted helminths and S . mansoni are co-endemic . While the results pertaining to S . mansoni clearly argue for SAF-conservation of stool samples and their analysis with FS7 at a later time point , the data regarding common soil-transmitted helminth infections suggest that immediate diagnosis with FS4 should be pursued . Important trade-offs between time and fecal egg counts also exist for different soil-transmitted helminth species . Sound conclusions can probably only be drawn once more results and experience from the field are available , but it is clear that a method which needs different times and solutions to reliably diagnose distinct helminth species infections in a poly-parasitized patient is not ideal . The current study is part of a broad attempt at validating the FLOTAC technique for human helminth diagnosis . Although we knew from previous investigations that FS4 is particularly suitable for the detection of soil-transmitted helminth eggs [32] , , and prior investigations with fecal pellets obtained from S . mansoni-infected mice and hamster revealed that FS7 is suitable for detection of S . mansoni eggs , this issue has never been addressed in a systematic manner , using a single pool of stool of uniform characteristics . In a post-calibration approach , we employed a composite human stool sample of ∼100 g , pooling stool from a few selected children with high-intensity helminth infections . The results of the post-calibration underscore that the ether washing step potentially destroys hookworm eggs after a certain conservation period in SAF . No S . mansoni eggs were found when using FS4 , and only few hookworm eggs were found in FS7 . With regard to A . lumbricoides and T . trichiura , the results from the post-calibration were less clear-cut than those for S . mansoni and hookworm , as a number of FS resulted in high fecal egg counts for A . lumbricoides and T . trichiura . Regarding parasitological S . stercoralis diagnosis , it is conventionally either done with the Koga agar plate method – as in the present study – or the Baermann method [12] , [21] , [31] , [44]–[46] . Our study offered an opportunity to also obtain preliminary results with FLOTAC for S . stercoralis diagnosis . Only 2 individuals were found positive for S . stercoralis when fresh stool samples were subjected to FLOTAC , whereas the Koga agar plate method revealed 38 infections . One of the 2 samples determined as S . stercoralis-positive using the FLOTAC method contained so many larvae that they were even observed in the Kato-Katz thick smears . Of note , the tegument of the S . stercoralis larvae detected by FLOTAC showed signs of degeneration , which might be due to the SAF preservation , the prior washing step with ether , the FS , or a combination of these chemicals . More research is needed to determine whether the FLOTAC technique might be further adapted to allow S . stercoralis diagnosis . In conclusion , the high sensitivity of a single FLOTAC examination for diagnosing common soil-transmitted helminth infections has been confirmed , but fecal egg counts are consistently lower when compared to the Kato-Katz method . This is an important issue and warrants additional studies . Importantly , we have shown that the FLOTAC method holds promise for the detection of S . mansoni eggs , particularly in well-homogenized stool samples after preservation in SAF for at least 10 days . Further validation of the FLOTAC technique is under way in different parts of the world , as this technique might become an indispensable tool for patient management and rigorous monitoring of anthelmintic drug efficacy studies and community-based helminth control programs .
|
Infections with parasitic worms ( e . g . , Schistosoma mansoni , hookworm , roundworm , whipworm , and threadworm ) are still widespread in the developing world . Accurate diagnosis is important for better patient management and for monitoring of deworming programs . Unfortunately , methods to detect parasite eggs or larvae in stool samples lack sensitivity , particularly when infection intensities are low . The most widely used method for the diagnosis of S . mansoni , hookworm , roundworm and whipworm in epidemiological surveys is the Kato-Katz technique . Recently , the FLOTAC technique has shown a higher sensitivity than the Kato-Katz method for the diagnosis of hookworm , roundworm and whipworm , but no data are available for S . mansoni . We compared the diagnostic accuracy of the FLOTAC with the Kato-Katz , ether-concentration and Koga agar plate techniques for S . mansoni and other parasitic worm infections using stool samples from 112 school children from Côte d'Ivoire . FLOTAC showed the highest sensitivity for S . mansoni diagnosis . Egg counts , however , were lower when using FLOTAC , an issue which needs further investigations . The FLOTAC , Kato-Katz and ether-concentration techniques failed to accurately detect threadworm larvae , and hence , the Koga agar plate remains the method of choice for this neglected parasite .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"public",
"health",
"and",
"epidemiology"
] |
2010
|
Comparing Diagnostic Accuracy of Kato-Katz, Koga Agar Plate, Ether-Concentration, and FLOTAC for Schistosoma mansoni and Soil-Transmitted Helminths
|
Conventional patent-based drug development incentives work badly for the developing world , where commercial markets are usually small to non-existent . For this reason , the past decade has seen extensive experimentation with alternative R&D institutions ranging from private–public partnerships to development prizes . Despite extensive discussion , however , one of the most promising avenues—open source drug discovery—has remained elusive . We argue that the stumbling block has been the absence of a critical mass of preexisting work that volunteers can improve through a series of granular contributions . Historically , open source software collaborations have almost never succeeded without such “kernels” . Here , we use a computational pipeline for: ( i ) comparative structure modeling of target proteins , ( ii ) predicting the localization of ligand binding sites on their surfaces , and ( iii ) assessing the similarity of the predicted ligands to known drugs . Our kernel currently contains 143 and 297 protein targets from ten pathogen genomes that are predicted to bind a known drug or a molecule similar to a known drug , respectively . The kernel provides a source of potential drug targets and drug candidates around which an online open source community can nucleate . Using NMR spectroscopy , we have experimentally tested our predictions for two of these targets , confirming one and invalidating the other . The TDI kernel , which is being offered under the Creative Commons attribution share-alike license for free and unrestricted use , can be accessed on the World Wide Web at http://www . tropicaldisease . org . We hope that the kernel will facilitate collaborative efforts towards the discovery of new drugs against parasites that cause tropical diseases .
There is a lack of high-quality protein drug targets and drug leads for neglected diseases [1] , [2] . Fortunately , many genomes of organisms that cause tropical diseases have already been sequenced and published . Therefore , we are now in a position to leverage this information by identifying potential protein targets for drug discovery . Atomic-resolution structures can facilitate this task . In the absence of an experimentally determined structure , comparative modeling can provide useful models for sequences that are detectably related to known protein structures [3] , [4] . Approximately half of known protein sequences contain domains that can be currently predicted by comparative modeling [5] , [6] . This coverage will increase as the number of experimentally determined structures grows and modeling software improves . A protein model can facilitate at least four important tasks in the early stages of drug discovery [7]: prioritizing protein targets for drug discovery [8] , identifying binding sites for small molecules [9] , [10] , suggesting drug leads [11] , [12] , and optimizing these leads [13]–[15] . Here , we address the first three tasks by assembling our computer programs into a software pipeline that automatically and on large-scale predicts protein structures , their ligand binding sites , and known drugs that interact with them . As a proof of principle , we applied the pipeline to the genomes of ten organisms that cause tropical diseases ( “target genomes” ) . We also experimentally tested two predicted drug-target interactions using Nuclear Magnetic Resonance ( NMR ) spectroscopy . By virtue of pairing specific proteins with already known drugs , our pipeline has the potential of increasing the efficiency of target identification , target validation , lead discovery , lead optimization , and clinical trials . The current project is part of our efforts within the Tropical Disease Initiative ( TDI , http://www . tropicaldisease . org ) [16] . TDI was conceived as a decentralized and web-based open source drug discovery effort in which academic and corporate scientists volunteer to work together on discovering drugs for neglected diseases . TDI's open source approach complements many new initiatives that have been proposed over the last decade [1] , [8] , [16]–[25] . However , relatively few volunteers have so far truly engaged in these efforts and their impact is still difficult to assess [26] . Based on our experience with The Synaptic Leap ( TSL ) online discussion forum of TDI ( http://www . thesynapticleap . org ) , we suggest that a major stumbling block for open source drug discovery has been the absence of a critical mass of preexisting work that volunteers can build on incrementally . Here , we address this bottleneck by introducing a “kernel” to facilitate drug discovery for tropical diseases . This kernel ( v1 . 0 ) includes 297 potential drug targets from the target genomes and is freely available via web 2 . 0 dissemination tools on the TDI web site . We begin by describing our computational pipeline as well as the experimental procedures for testing two selected targets ( Methods ) . Next , we describe the modeling of proteins in ten pathogen genomes , prediction of binding of known drugs to the modeled proteins , and experimental testing of these predictions for two select protein targets ( Results ) . Finally , we discuss how we expect a full-scale TDI open source project to use the kernel and its potential impact on open source drug discovery ( Discussion ) .
We have assembled a computational pipeline that relies on several databases and programs , taking as input protein sequences and producing an output containing protein models as well as predicted locations of binding sites for small molecules on their surfaces and predicted types of molecules they bind . The pipeline , which relies on the MODPIPE package [27] and the AnnoLyze program [9] , has been applied to genomes of ten pathogens that cause tropical diseases . The output of the pipeline has been stored in a relational database for easy searching and dissemination over the web . We selected the following ten target genomes based on both disease burden and the completeness of published sequences: Cryptosporidium hominis ( CyrptoDB [28] ) , Cryptosporidium parvum ( CryptoDB [28] ) , Leishmania major ( GeneDB [29] ) , Mycobacterium leprae ( OrthoMCL-DB [30] ) , Mycobacterium tuberculosis ( TubercuList [31] ) , Plasmodium falciparum ( PlasmoDB [32] ) , Plasmodium vivax ( PlasmoDB [32] ) , Trypanosoma brucei ( GeneDB [29] ) , Trypanosoma cruzi ( GeneDB [29] ) , and Toxoplasma gondii ( ToxoDB [33] ) . We then mapped the transcript sequences onto UniProt ids [34] . Functional annotation for predicted binding sites in our models relied on the following databases: ( i ) UniProt [34] , which contains 385 , 721 sequences from the SwissProt database and 5 , 814 , 087 sequences from the TrEMBL database , was used to annotate the transcripts from the target genomes; ( ii ) MODBASE [6] , which contains 6 , 805 , 385 comparative models calculated by MODPIPE for domains in 1 , 810 , 521 proteins , was used to store all comparative models; ( iii ) DBAli [35] , which contains 1 . 7 billion pairwise alignments generated by an all-against-all comparison of known protein structures , was used to identify structure relationships between our modeling templates and other known protein structures; ( iv ) LigBase [36] , which contains 232 , 852 structurally defined ligand-binding sites in PDB , was used as a resource for AnnoLyze to predict ligand binding sites on pathogen protein models; ( v ) MSDChem [37] , which contains 8 , 287 small ligands , was used as an annotated repository of small molecules in the PDB database; and ( vi ) DrugBank [38] , which contains 4 , 765 drug-like compounds ( including 1 , 485 FDA-approved small molecule drugs , 128 FDA-approved biotech drugs , 71 nutraceuticals , and 3 , 243 experimental drugs ) , was used to identify small molecules in the MSDChem database that have similar chemical composition to known drugs . Models for all sequences from the ten target genomes were calculated using MODPIPE , our automated software pipeline for comparative modeling [27] , [39] . It relies primarily on the various modules of MODELLER [40] for its functionality and is adapted for large-scale operation on a cluster of PCs using scripts written in PERL and Python . Sequence-structure matches are established using a variety of fold-assignment methods , including sequence-sequence [41] , profile-sequence [42] , [43] , and profile-profile alignment [43] , [44] . Odds of finding a template structure are increased by using an E-value threshold of 1 . 0 . By default , ten models are calculated for each of the alignments [40] . A representative model for each alignment is then chosen by ranking based on the atomic distance-dependent statistical potential DOPE [45] . Finally , the fold of each model is evaluated using a composite model quality criterion that includes the coverage of the modeled sequence , sequence identity implied by the sequence-structure alignment , the fraction of gaps in the alignment , the compactness of the model , and various statistical potential Z-scores [45]–[47] . We only used the models that were predicted to have a “correct” fold ( i . e . , a MODPIPE quality score higher than 1 . 0 ) ; based on our benchmarking studies , we expect the true positives rate of 93% and the false positives rate of 5% . The AnnoLyze program [9] was used to predict binding sites for small molecules on all well-assessed models . Briefly , AnnoLyze predicts ligand-binding sites on the surface of a model by transferring known ligands in the LigBase database [36] via the target-template alignment . Such predictions are made in a two step process ( Figure 1 ) : ( i ) transfer of a binding site between known structures ( i . e . a ligand co-crystallized with a protein structure is transferred to another known structure if at least 75% of the LigBase-defined binding site residues are within 4 Å of the template residues in a global superposition of the two structures and if at least 75% of the binding site residue types are invariant ) ; and ( ii ) transfer of a binding site to a comparative model using as a reference the alignment to its template ( i . e . a ligand predicted in the previous step to bind the template or a ligand co-crystallized with the template is transferred to the comparative model if the binding sites are conserved at the same level as in the previous step ) . Using these cutoffs , approximately 30% of the selected models had at least one predicted binding site for small molecules ( Table 1 ) , which were then mapped to MSDChem entries . The jcsearch program from the JChem package [48] was used with default parameters to match related compounds in MSDChem and DrugBank . Four types of matches were collected: ( i ) exact matches ( i . e . their SMILES strings [49] matched with a Tanimoto score [50] equal to 1 . 0 ) ; ( ii ) supra-structure matches in which a matched DrugBank query molecule is a part of an MSDChem molecule; ( iii ) sub-structure matches in which an identified MSDChem molecule is a part of a DrugBank query molecule; and ( iv ) similar matches with a Tanimoto score between MSDChem and DrugBank molecules of at least 0 . 9 . The tested proteins ( i . e . , a putative thymidylate kinase from P . falciparum and a nucleoside diphosphate kinase from M . leprae ) were produced by cloning the full length annotated ORFs into pET47b plasmids ( Novagen ) . The resulting plasmids were purchased from GeneArt ( http://www . geneart . com , Regensburg , Germany ) and sequenced using conventional methods to confirm the intended constructs were obtained . The proteins were then over-expressed as fusion proteins using BL21 ( DE3 ) Codon plus cells ( Strategene ) . Purification of the proteins was facilitated by a hexa-His tag at the N-terminus and an engineered cleavage site for the TEV protease . Purification to homogeneity was carried out using metal-affinity chromatography ( Talon , Clontech ) , followed by TEV cleavage . All spectra were recorded at 300 K with a Bruker Ultrashield Plus 600 MHz NMR spectrometer equipped with a 5 mm TCI cryogenically cooled probe . A typical NMR sample contained a concentration of 5 µM of protein , 100 µM of ligand , 100 µM of glucose as a negative control , 100 mM NaCl , and 25 mM phosphate buffer at pH 7 . 0 . The concentration of ligand for the Saturation Transfer Difference ( STD ) experiments was 500 µM . For each sample , a 1D 1H reference , a Water-LOGSY [51] and a STD [52] experiment were recorded . 8 K points were used for a sweep width of 9 , 600 Hz and a total of 1 K and 512 scans were accumulated for the Water-LOGSY and STD experiments , respectively . The entire kernel , including all predicted models and binding sites , is freely available over the web ( http://www . tropicaldisease . org/kernel ) . The server uses the WordPress package ( http://www . wordpress . org ) , a widely used platform that facilitates easy creation , storage , and dissemination of each target entry in our database . WordPress supports numerous “plugins” , including a rating system that allows TDI web site users to rate targets for “druggability . ” The package also supports bookmarking by most web-based social networks . In particular , each of the TDI kernel's target pages includes a “blog it” button that allows registered users of The Synaptic Leap ( TSL , http://www . thesynapticleap . org ) to post TDI entries directly into the TSL discussion panels . TSL is our web-based “collaboratory” portal that is designed to host open source drug discovery projects in much the same way SourceForge hosts software collaborations . The TDI kernel is fully searchable and downloadable through our Web site ( http://www . tropicaldisease . org/kernel/ ) . Options include direct downloads of individually requested targets , pre-defined sets for each of our ten target genomes , and user-defined batch downloads . Additionally , all our predictions are available as supporting information files to this article ( Datasets S1 , S2 , S3 , S4 ) . Users receive the data with no restriction in accordance with the Science Commons protocol for implementing open access data [53] that was designed to embody normal academic attribution norms and facilitate tracking of work based on the kernel . While our predictions are in the public domain , some of the drugs used in our predictions might be subject to patents .
The accuracy of our comparative protein structure models built using MODPIPE was predicted by a variety of criteria , including target-template sequence identity , coverage of the target sequence , fraction of gaps in the alignment , and statistical potential scores . One third of the total models ( 21 , 031 ) were assessed to have sufficient accuracy for predicting the location and type of their binding sites for small compounds ( i . e . , at least 50% of their Cα atoms are predicted to be within 3 . 5 Å of their correct positions , corresponding to the correct fold and at least an approximately correct alignment with the template structure ) . These models covered 11 , 714 protein targets , corresponding to 17% of all proteins in the ten target genomes ( Table 1 and Figure 2 ) . There are an average of ∼2 . 5 models per protein target , each model potentially based on a different template structure and/or covering a different domain of the modeled sequence . Different genomes presented different levels of difficulty to our modeling procedure: 75% of the models for M . leprae proteins met our accuracy standards , while only approximately 10% of T . gondii models did . These coverage correspond to accurate predictions for 3 , 070 targets in Trypanosoma cruzi ( 15 . 7% of the genome ) and 300 targets ( 3 . 9% of the genome ) for T . gondii ( Table 1 ) . We applied our AnnoLyze program to predict the binding sites for small molecules in the MSDChem database on 11 , 714 well-modeled targets . A total of 3 , 499 ( ∼30% ) of these targets had a predicted binding site from their comparative modeling template or a known binding site transferred from a structurally similar protein . Once again , the T . cruzi genome had the largest number of predicted binding sites located in 769 targets , while T . gondii contained only 138 targets with a predicted small-molecule binding site ( Table 1 and Figure 2 ) . In general , there was an almost linear relationship between the genome size and the number of targets with predicted binding sites . The M . leprae genome provided a notable exception , with accurate models covering domains in 55 . 6% of the proteins and predicted binding sites for a small molecule in only 310 of these targets . The coverages of comparative modeling and ligand binding site prediction vary from one genome to another ( Table 1 ) . For example , T . gondii has poor structure coverage of its 7 , 793 genes predicted in ToxoDB ( 3 . 85% ) . This poor structural coverage may be partly a result of a relatively inaccurate current assignment of genes , as suggested by differences between four methods for predicting genes from a genome [54]; these annotations agreed in only 12% of the genes . Moreover , 3 , 837 genes in ToxoDB are poorly annotated with keywords such as “hypothetical” , “putative” , and “predicted” . In contrast , M . leprae , which is a minimal mycobacterial genome [55] , resulted in the highest coverage of all target genomes ( 55 . 64% ) . This high coverage is a consequence of a larger proportion of its sequences having homologs whose complexes with small molecules have been defined structurally . Finally , there is an artificially large number of predictions for T . cruzi . The T . cruzi genome was sequenced from a hybrid strain from two divergent parental lines [56] , which resulted in a large number of its genes with duplicated entries in the GeneDB database . Given that our computational pipeline relies on homology for predicting the structure and binding sites of a query sequence , we analyzed the predictions across ortholog sequences from the ten target genomes . A total of 236 of the 297 selected targets group into 46 ortholog groups as defined by the OrthoMCL-DB database [30] . Our predictions agreed for 38 of the 46 ortholog groups ( i . e . , the same ligands were predicted to bind all the orthologs within the cluster ) . Only 4 of the 46 ortholog groups resulted in a complete disagreement ( i . e . , all orthologs resulted in different predicted binding ligands ) . Finally , the remaining 4 ortholog groups had intermediate results ( i . e . , some but not all of the orthologs in the cluster were predicted to bind the same ligand ) . To link small molecules from MSDChem to chemical compounds in DrugBank , we used JChem to perform an all-against-all comparison of the SMILES strings from both databases ( Table 1 ) . This linking allowed us to predict 297 proteins that are likely to bind a known drug from DrugBank or a compound similar to it ( i . e . , with a Tanimoto score of at least 0 . 9 ) ; 143 of these targets were predicted to have a binding site for a known drug ( i . e . , a Tanimoto score of 1 . 0 ) . Next , we outline two predictions that make sense in the light of the known antiprotozoal activity of the corresponding drugs . Our pipeline correctly predicted that the known antiprotozoal drug Trimethoprim ( DrugBank identifier DB00440 ) interacts with a dihydrofolate reductase ( UniProt identifier A1QV37 ) in Mycobacterium tuberculosis . Trimethoprim is a pyrimidine-like inhibitor of dihydrofolate reductases that acts as an antibacterial agent and has weak antimalaria activity [57] . Moreover , our predictions suggest that Trimethoprim might also inhibit a dihydrofolate reductase from M . leprae ( UniProt identifier Q9CBW1 ) , given that its binding site is 93 . 3% identical in sequence to that of dihydrofolate reductase from M . tuberculosis ( Figure 3A ) . In a second example , our predictions shed light on the molecular mechanism of aroyl-pyrrolyl-hydroxyamides , a class of histone deacetylase inhibitors , which have previously been reported to have antileishmanial activity [58] , [59] . Although the structure of Leishmania major's histone deacetylase is unknown ( UniProt identifier Q4QCE7 ) , it can be modeled using the structure of the human histone deacetylase as a template ( sequence identity is 36 . 0% ) . Using the ligand binding site prediction protocol of AnnoLyze , we predict a binding site for SSH ( octanedioic acid hydroxyamide phanylamide ) in the human histone deacetylase ( PDB identifier 1t64A ) , as found in the Aquifex aeolicus histone deacetylase ( PDB identifier 1c3sA ) . The coverage and sequence identity of the binding site for SHH , which is an exact match to the drug Vorinostat ( DrugBank identifier DB02546 ) , was 100 . 0% and 90 . 9% , respectively . Thus , our predictions suggest molecular details of Vorinostat's mechanism of action as an inhibitor of L . major histone deacetylase ( Figure 3B ) . Two additional predicted drug targets were used to test our computational methods using NMR spectroscopy: ( i ) a putative thymidylate kinase from Plasmodium falciparum ( UniProt identifier Q8I4S1 ) predicted to bind Zidovudine ( a nucleoside reverse transcriptase inhibitor ) and ( ii ) a nucleoside diphosphate kinase from M . leprae ( UniProt identifier Q9CBZ0 ) predicted to bind Fludarabine ( a DNA polymerase alpha , ribonucleotide reductase and DNA primase inhibitor ) . Both targets were selected based on the feasibility of NMR experiments ( i . e . , protein shorter than 250 amino acid residues in length ) , non-trivial modeling ( i . e . , the target and the template were globally aligned with less than 75% sequence identity ) , and non-trivial prediction of the ligand ( i . e . , using only similarity matches ) . Thymidylate kinases ( TMPK ) catalyze the reversible phosphorylation of deoxythymidine monophosphate ( dTMP ) to deoxythymidine diphosphate ( dTDP ) and are essential for the survival of the organism . In particular , the TMPK from P . falciparum was recently expressed and biochemically characterized in terms of its molecular affinity to several substrates and appears to be a good target for drug discovery , especially for binding to purine-based inhibitors [60] . We predicted that TMPK from P . falciparum binds ATM ( 3′-azido-3′-deoxythymidine-5′-monophosphate ) . ATM is highly similar to Zidovudine , which lacks only the 5′ monophosphate of ATM . Zidovudine is a dideoxynucleoside that prevents the formation of phosphodiester linkages needed for the completion of nucleic acid chains . It has been used as a potent inhibitor of HIV replication , acting as a chain-terminator of viral DNA during reverse transcription . An experimental structure of P . falciparum TMPK is not available , but can be predicted by comparative modeling based on 41% sequence identity to a known structure of the yeast TMPK ( PDB identifier 3tmkA ) . 3tmkA also has a predicted binding site for ATM , which was transferred from another crystallized structure of yeast TMPK ( PDB identifier 2tmkA ) . Using NMR Water-LOGSY and STD experiments , we have tested the binding capacity of both ATM and Zidovudine to the surface of P . falciparum TMPK . In the Water-LOGSY experiments , the large bulk water magnetization is partially transferred via the protein-ligand complex to the free ligand in a selective manner . As a consequence , the resonances of the ligand have a sign opposite to that of non-interacting compounds; their signal also appears stronger . To test the applicability of the Water-LOGSY experiment to P . falciparum TMPK , we tested glucose as a negative control ( i . e . , non-interacting ligand ) and dTMP as a positive control ( i . e . , a known ligand for TMPK ) , resulting in the expected negative and positive interacting signals , respectively ( Figure 4A ) . With this validation in hand , similar experiments were performed with ATM and Zidovudine . Both ATM and Zidovudine result in positive Water-LOGSY signals , confirming their predicted interaction with P . falciparum TMPK . The results were further validated by the positive signals in the STD spectra that are better suited for detecting interactions between strong binders and proteins . Nucleoside diphosphate kinases ( NDK ) have major roles in the synthesis of nucleoside triphosphates other than ATP . In particular , the NDK from M . leprae was predicted to bind cAMP ( adenosine-3′ , 5′-cyclic-monophosphate ) . cAMP has a similar structure to the known drug Fludarabine , which inhibits DNA synthesis and has been used in chemotherapy for the treatment of hematological malignancies . We built a comparative model of M . leprae NDK based on 58% sequence identity to the NDK form Thermus thermophilus ( PDB identifier 1wkjA ) . 1wkjA has a predicted binding site for cAMP , based on its similarity to Myxococcus xanthus NDK ( PDB identifier 1nhkR ) , which is known to bind cAMP . As for TMPK , we used Water-LOGSY and STD experiments to determine whether or not cAMP and Fludarabine bind to the surface of M . leprae NDK . For this target , glucose and GDP , a known NDK ligand , were used as negative and positive controls , respectively ( Figure 4B ) . The Water-LOGSY experiments showed an almost undetectable interaction , between cAMP and NDK . This finding was confirmed by the STD experiment . However , neither of the experiments resulted in positive signs in the NMR spectra of the interaction between Fludarabine and NDK , invalidating our prediction .
Identifying targets and lead compounds that have good odds for surviving clinical trials is one of the most challenging tasks facing the pharmaceutical industry . This challenge is particularly urgent in the neglected disease context where the upstream end of the development pipeline is in danger of drying up [1] . Here , we have introduced a new computational pipeline that generates comparative models of input protein sequences , the location of small molecule binding sites on these models , and the types of compounds that bind to them . We have applied this pipeline to ten complete genomes of pathogens causing neglected diseases and the set of compounds in the DrugBank database , which contains both known drugs and related molecules . Using NMR spectroscopy , we have also experimentally tested two predictions , validating one of them . The high efficiency and coverage of our computational methods is particularly important for tropical disease research , where commercial markets are too small to support conventional patent-based research models . Identifying new protein targets and previously developed drugs that interact with them have the potential of greatly simplifying experimental validation of these new targets , lead optimization , and clinical trials . Moreover , our approach can lead to characterizations of the mechanism of action of already known drugs . Because tropical diseases affect millions of people , the stakes could not be higher . A total of 68 , 877 protein sequences encoded by ten genomes were input into MODPIPE , resulting in models for 11 , 714 ( 17% ) target sequences that were estimated to be sufficiently accurate for predicting the location and type of binding sites on their surfaces . With these models in hand , AnnoLyze , our binding site prediction program , was able to predict a binding site for a small molecule on 3 , 499 potential targets , of which 297 were predicted to bind a molecule similar to a known drug , including 143 predicted to bind a known drug . These protein targets , available through the TDI's kernel web site ( http://www . tropicaldisease . org/kernel/ ) , can be regarded as “low hanging fruits” for drug discovery in tropical diseases . Using NMR spectroscopy , we have experimentally tested whether or not two of these targets actually bind their predicted drug ligands . While our experiments have not tested for either binding site localization or binding affinity , they do confirm that the drug Zidovudine indeed interacts with a P . falciparum thymidylate kinase . In contrast , the prediction of the binding of Fludarabine to M . leprae nucleoside diphosphate kinase was invalidated . This prediction was based on the relatively low conservation of the predicted binding site ( 75% sequence identity between the binding site residues in the template and target ) , indicating that such predictions should be treated with caution . The key contribution of this work results from the structural analysis of putative binding sites in the surface of protein structure models of genes from ten organisms that cause tropical diseases . However , it is not clear how to assess the false positives and false negatives rates for our computational method based on the existing experimental information . Our understanding of errors in comparative modeling [9] and in similarity-based transfer of functional sites between homologs [4] , combined with the limited experimental validation reported here , suggests that a useful fraction of predictions are correct . We urge other investigators to donate their expertise and facilities to validate our many predictions , within the open source context . The main goal of our exercise was to narrow down the number of targets and identify their putative ligands for experimental follow-up , so that the overall process is faster , more thorough , and less expensive . We see the TDI kernel only as a beginning . For example , our methods not only predict plausible ligands for a target , but also localize the binding site on the surface of the protein , a necessary step for further leveraging our results for optimizing the lead compounds by a combination of computational and experimental methods , such as computational docking , site-directed mutagenesis , and synthetic chemistry . We also recognize that the kernel's list of “hits” does not even remotely exhaust the ten target genomes . Researchers who want TDI to investigate additional candidates ( whether or not previously published ) should contact us or engage in online discussions ( http://www . thesynapticleap . org ) . Moreover , our TDI and TSL web sites provide a full suite of Web 2 . 0 tools for disseminating the kernel for further annotation . It would be counterproductive for TDI to patent or otherwise seek intellectual property rights in these discoveries . Of course , there is no guarantee that others do not claim such rights . For example , some of the drugs in DrugBank may be the subjects of patents . Nevertheless , the existence of unpatented targets and at least some unpatented compounds will give sponsors bargaining power in negotiations with patent owners , if they demanded excessive royalties . The net result will be to reduce the royalties that patent owners can charge and sponsors must pay . Many open source licenses contain “viral” terms , which limit users' ability to seek intellectual property of their own . In the case of drug discovery , however , such strategies are likely to be expensive and , in some cases , legally dubious [61] , [62] . Nevertheless , these obstacles are not fatal and one can imagine schemes in which discoveries are embargoed for months or years , so that access is limited to those who promised not to seek patents of their own [63] . We have decided against trying to impose a viral condition on subsequent researchers . First and foremost , open source requires as many workers , volunteer and commercial , as possible , implying minimal restrictions on the data , including viral terms . Second , at least some of the organisms included in the kernel ( e . g . , M . tuberculosis ) have potential commercial markets large enough to offset a fraction of sponsors' R&D costs . Nevertheless , it is still possible that an unscrupulous corporation , for example , could try to patent trivial improvements to the kernel . This , however , seems unlikely in the impoverished world of neglected disease research , at least for the immediate future . In the meantime , we prefer to leave the question open until open source collaboration has been firmly established . That will put the final responsibility where it belongs – with the volunteers whose labor and insights we are depending on to turn TDI's kernel into safe , effective , and affordable cures .
|
Open source drug discovery , a promising alternative avenue to conventional patent-based drug development , has so far remained elusive with few exceptions . A major stumbling block has been the absence of a critical mass of preexisting work that volunteers can improve through a series of granular contributions . This paper introduces the results from a newly assembled computational pipeline for identifying protein targets for drug discovery in ten organisms that cause tropical diseases . We have also experimentally tested two promising targets for their binding to commercially available drugs , validating one and invalidating the other . The resulting kernel provides a base of drug targets and lead candidates around which an open source community can nucleate . We invite readers to donate their judgment and in silico and in vitro experiments to develop these targets to the point where drug optimization can begin .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"computational",
"biology/protein",
"structure",
"prediction",
"computational",
"biology/macromolecular",
"structure",
"analysis",
"pharmacology/drug",
"development",
"biotechnology/small",
"molecule",
"chemistry"
] |
2009
|
A Kernel for Open Source Drug Discovery in Tropical Diseases
|
Kaposi’s sarcoma herpesvirus ( KSHV ) is the etiologic agent of Kaposi’s sarcoma ( KS ) as well as two lymphoproliferative diseases , primary effusion lymphoma and multicentric Castleman’s disease . KSHV encodes viral proteins , such as K1 , that alter signaling pathways involved in cell survival . Expression of K1 has been reported to transform rodent fibroblasts , and K1 transgenic mice develop multiple tumors , suggesting that K1 has an important role in KSHV pathogenesis . We found that cells infected with a KSHV virus containing a WT K1 gene had a survival advantage under conditions of nutrient deprivation compared to cells infected with KSHV K1 mutant viruses . 5’ adenosine monophosphate-activated protein kinase ( AMPK ) responds to nutrient deprivation by maintaining energy homeostasis , and AMPK signaling has been shown to promote cell survival in various types of cancers . Under conditions of AMPK inhibition , we also observed that cells infected with KSHV containing a WT K1 gene had a survival advantage compared to KSHV K1 mutant virus infected cells . To explore the underpinnings of this phenotype , we identified K1-associated cellular proteins by tandem affinity purification and mass spectrometry . We found that the KSHV K1 protein associates with the gamma subunit of AMPK ( AMPKγ1 ) . We corroborated this finding by independently confirming that K1 co-immunoprecipitates with AMPKγ1 . Co-immunoprecipitations of wild-type K1 ( K1WT ) or K1 domain mutants and AMPKγ1 , revealed that the K1 N-terminus is important for the association between K1 and AMPKγ1 . We propose that the KSHV K1 protein promotes cell survival via its association with AMPKγ1 following exposure to stress .
Kaposi’s sarcoma-associated herpesvirus ( KSHV ) is the causative agent of the endothelial cancer , Kaposi’s sarcoma ( KS ) , and two B-cell lymphomas including primary effusion lymphoma ( PEL ) and multicentric Castleman’s disease ( MCD ) [1–3] . KSHV-related malignancies primarily arise in immune-suppressed individuals including HIV-positive individuals and organ transplant recipients , although these cancers can also occur in the absence of immunosuppression . KS is a common cancer in some sub-Saharan African countries [4 , 5] . KSHV is a double-strand DNA gammaherpesvirus that is 165 to 170 kb long [6] . The KSHV genome contains multiple open reading frames that are conserved among other herpesviruses , and genes K1-K15 that are unique to KSHV [7] . Similar to other herpesviruses , KSHV has latent and lytic phases . Upon entering the host cell , KSHV typically establishes latency and expresses a limited number of viral proteins . Upon reactivation , which can be induced in vitro with various compounds such as 12-O-tetradecanoylphorbol-13-acetate ( TPA ) , histone deacetylase ( HDAC ) inhibitors , and TLR 7/8 ligands , KSHV enters the viral lytic cycle resulting in the production of infectious virions [7 , 8] . Both latent and lytic phases appear to be important for KSHV pathology . Expression of latent genes generally promotes the survival of the infected cell and persistence of infection during cell division . Lytic gene expression results in the production of inflammatory cytokines , pro-angiogenic factors and viral proteins that subvert the host immune system and promote virion production . KSHV K1 is primarily expressed during the lytic phase although recent studies indicate that K1 is also expressed at low levels during latency [9–11] . K1 is a 46-kDa transmembrane glycoprotein that contains a C-terminal immunoreceptor tyrosine-based activation motif ( ITAM ) analogous to the signaling molecules in the B-cell receptor ( BCR ) signaling complex [12] . The K1 ITAM has been found to interact with various SH2 containing signaling molecules , including among others , the p85 regulatory unit of phosphoinositide-3-kinase ( PI3K ) [13] . K1 has been shown to initiate a signaling cascade leading to intracellular calcium mobilization , upregulation of NFAT and AP-1 transcription factors , and production of inflammatory cytokines [12 , 13] . It is thought that K1 is maintained in an activated state by oligomerization of the K1 ectodomain and subsequent phosphorylation of the ITAM tyrosines by Src family kinases [14] . K1 has a role in KSHV-induced tumor development . K1 expression immortalizes primary endothelial cells , transforms rodent fibroblasts , and K1 transgenic mice develop spindle cell sarcomatoid tumors and plasmablastic lymphoma , suggesting that the K1 protein is important for KSHV-induced tumor development [15–17] . These cancerous phenotypes may be due to K1’s modulation of cellular proteins in signaling pathways that are important for cell survival . We and others have previously shown that K1 activates the PI3K/Akt/mTOR pathway and protects against Fas-mediated apoptosis [18–20] . In our current studies , we observed that cells infected with KSHV viruses containing a wild-type K1 gene ( KSHV-K1WT and KSHV-K1REV ) displayed a survival advantage under conditions of nutrient deprivation compared to viruses containing mutant K1 genes ( KSHV-K15XSTOP and KSHVΔK1 ) . To understand the underpinnings of this phenotype , we performed tandem affinity purification and mass spectrometry to identify K1 binding proteins . We found that KSHV K1 associates with the gamma subunit of 5’adenosine monophosphate-activated protein kinase ( AMPKγ1 ) . AMPK is a heterotrimeric serine/threonine kinase composed of an alpha catalytic subunit and two regulatory subunits , beta and gamma [21] . Each subunit is part of a larger isoform family including the following subunit isoforms: α1 , α2 , β1 , β2 , γ1 , γ2 and γ3 [22–25] . The isoforms of each subunit are found in different compartments within the cell . AMPKα1 and AMPKα2 localize to the cytoplasm . AMPKα2 also localizes to the nucleus in rat pancreatic and HeLa cells [26] . AMPKα1 and AMPKβ1 are in the perinuclear region in HEK-293 cells [27] . Mammalian AMPKα2 , AMPKβ1 , and AMPKγ1 are in the nuclei of neurons [28] . The subunit isoforms can come together in various combinations to make different heterotrimers . The differences in function of each heterotrimer are still under investigation . The presence of the three subunits is necessary for full activation of AMPK and the regulatory subunits stabilize expression of the catalytic α subunit [29] . AMPK responds to stresses that reduce ATP levels by inhibiting anabolic and activating catabolic pathways to maintain energy homeostasis [30] . Binding of adenosine monophosphate ( AMP ) to the gamma subunit allosterically activates AMPK and promotes phosphorylation of AMPKα at Thr172 by upstream kinases [31–33] . AMPK also responds to environmental stress factors that reduce cellular ATP levels such as hypoxia [34–37] . The role of AMPK as a tumor promoter is actively being explored [38 , 39] . Some studies suggest that AMPK promotes tumor cell survival in vitro and in vivo . Inhibition of AMPK results in reduced prostate cell survival and increased apoptosis under normal and stressed conditions [40 , 41] . AMPK has also been shown to promote survival in multiple myeloma , colorectal and glioma cancer cell lines [42–44] . In vivo , AMPK signaling was found to be elevated in developing tumors in a glioblastoma rat model [45] . Moreover , there is reduced in vivo tumor growth of xenografts prepared from transformed AMPKα1/α2-null MEFs compared to wild-type ( WT ) MEFs [37] . Thus , there is accumulating evidence suggesting that AMPK may promote cancer cell survival and tumor development . Here we report that K1 binds AMPKγ1 and that this interaction is important for K1’s ability to enhance cell survival .
BAC16 recombinant viruses containing WT K1 ( KSHV-K1WT and KSHV-K1REV ) were made as previously described [46] . Immortalized human umbilical vein endothelial cells ( HUVEC ) [17] or iSLK cells were infected with BAC16 recombinant viruses containing WT K1 ( KSHV-K1WT and KSHV-K1REV ) or mutant K1 ( KSHV-K15XSTOP and KSHVΔK1 ) genes [46] . These recombinant BAC16 viruses contain a GFP marker to monitor cell infectivity [46] . Both HUVEC and iSLK cells were stably selected until 100% of cells were green indicating that all cells were infected with these viruses . Each HUVEC cell line was then subjected to stress by withdrawing serum and growth factors . KSHV-K1WT and KSHV-K1REV ( revertant ) harbor a wild type K1 gene while the KSHV-K15XSTOP and KSHVΔK1 lack K1 ( Fig 1A ) . We evaluated cell viability at various time-points by MTS ( [3- ( 4 , 5-dimethylthiazol-2-yl ) -5- ( 3-carboxymethoxyphenyl ) -2- ( 4-sulfophenyl ) -2H-tetrazolium , inner salt ) , which is a measure of metabolic activity , and trypan blue exclusion assay . At 24 , 48 , and 72 hours following nutrient deprivation , HUVEC infected with KSHV-K1WT and KSHV-K1REV had more viable cells compared to the KSHV-K15XSTOP and KSHVΔK1 HUVEC as observed using the MTS assay ( Fig 1B ) . The differences were greatest at 72 hours post-nutrient deprivation . To further substantiate these results , we also performed trypan blue exclusion assays and observed that the KSHV-K1WT and KSHV-K1REV infected cells were more viable compared to the cells infected with KSHV-K15XSTOP and KSHVΔK1 infected cells at 48 and 72 hours ( Fig 1C ) . Similar to the MTS assay results , we observed reductions in cell viability in KSHV-K15XSTOP and KSHVΔK1 infected cells compared to KSHV-K1WT and KSHV-K1REV infected cells at 24 , 48 , and 72 hours post-starvation ( Fig 1C ) . When we added serum and growth factors back to the media of KSHV-K1 WT and K1 mutant infected cells that had been starved for 72 hours , we noticed that there was still a reduction in viable KSHV-K1 mutant infected cells compared to the number of viable KSHV-K1 WT infected cells by trypan blue exclusion assay ( S1B Fig ) . There is an increase in viable KSHV-K1 mutant cells from 72 hours of starvation to 72 hours of nutrient replenishment , just not to the same levels observed in KSHV-K1 WT ( S1A and S1B Fig ) . Furthermore , iSLK cell lines stably infected with these same viruses ( KSHV-K1WT , KSHV-K1REV , KSHV-K15XSTOP and KSHVΔK1 ) were also evaluated following serum withdrawal . Every two days for a total of 12 days , we evaluated cell viability by trypan blue exclusion . At all time-points following serum withdrawal , we observed that KSHV-KTWT and KSHV-K1REV infected iSLK cells had increased viability compared to KSHV-K15XSTOP and KSHVΔK1 iSLK cells ( Fig 1D ) . These findings suggest that K1 promotes KSHV-infected cell survival in the context of the whole genome and under conditions of nutrient deprivation . A major regulator of metabolic stress is AMP-activated protein kinase ( AMPK ) . AMPK responds to metabolic stress by activating catabolic pathways and inhibiting anabolic cell signaling pathways to maintain energy homeostasis [47] . Because we observed that cells infected with KSHV expressing a WT K1 gene had a survival advantage compared to cells infected with KSHV K1 mutant viruses following exposure to metabolic stress , and AMPK is involved in maintaining metabolic homeostasis , we hypothesized that K1 might modulate AMPK function . To explore this possibility , we compared cell survival in HUVEC infected with KSHV-K1WT and KSHV-K1REV wild-type viruses to KSHV-K15XSTOP and KSHVΔK1 mutant viruses following treatment with the AMPK inhibitor , compound C . Compound C is a reversible and competitive inhibitor of ATP [48] . In the presence of 5μM ATP and absence of AMP , compound C has a Ki of 109 ± 16 nM [48] . Compound C significantly prevents AMPK activation in vitro at 20 μM , and at 40 μM in cells treated with the AMPK activators metformin or AICAR [48] . According to Zhou et al . , compound C has minimal impact on structurally related kinases such as ZAPK , SYK , PKCθ , PKA and JAK3 [48] . Inhibition of AMPK by compound C has been shown to induce cell death in various cell lines [41 , 49] . We observed increased cell viability in KSHV-K1WT and KSHV-K1REV infected cells compared to KSHV-K15XSTOP infected cells by MTS assay ( Fig 2A ) . Corroborating these results , we also observed increased cell viability in KSHV-K1WT and KSHV-K1REV compared to KSHVΔK1 infected cells , suggesting that cells infected with WT K1 virus are less sensitive to the stress induced by AMPK inhibition than cells infected with the KSHV K1 mutants ( Fig 2B ) . As expected , cell viability between KSHV-K1WT and KSHV-K1REV infected cells was essentially the same ( Fig 2A and 2B ) . To determine whether KSHV-K1WT survival could be impacted by knock down of AMPK , we treated HUVEC stably infected with KSHV-K1WT with AMPKα1 and AMPKα2 siRNA . We observed reduced cell viability in HUVEC infected with KSHV-K1WT that had been treated with AMPKα1 and AMPKα2 siRNA compared to cells treated with NS siRNA at 48 and 72 hours ( S2A Fig ) . We also evaluated AMPKα1 and AMPKα2 expression by immunoblot to confirm knock down of AMPKα1and AMPKα2 ( S2B Fig ) . This data suggests that when AMPK is depleted , KSHV-K1WT cells are susceptible to cell death . To evaluate the impact of K1 expression by itself on cell survival following treatment with the AMPK inhibitor , compound C , we created FLAG epitope-tagged K1 ( K1 ) or empty vector ( EV ) HEK-293 stable cell lines . The EV or K1 HEK-293 cells were treated with increasing concentrations of compound C , and cell viability was evaluated by trypan blue exclusion assay . We observed an increased number of viable cells in the K1 expressing cells compared to EV expressing cells using two different concentrations ( 10 μM and 20 μM ) of compound C ( Fig 3A ) . We observed this difference at both 8 and 24 hours following incubation of the cells with 10 μM compound C ( Fig 3B ) . In order to determine whether K1 protected from apoptosis induced by AMPK inhibition , we also performed an assay to detect active caspase-3 , which is an indicator of apoptosis . Active caspase-3 levels were remarkably elevated in EV cells compared to K1 expressing cells , suggesting that K1 protects cells from apoptosis when AMPK is inhibited ( Fig 3C ) . K1 expression in these cells was confirmed by a Western blot ( Fig 3D ) . This data suggests that K1 expression can keep cells alive by diminishing the effects of AMPK inhibition . To investigate how K1 may be promoting cell survival following exposure to metabolic stress , we wanted to determine cellular proteins associated with K1 . We identified K1-associated cellular proteins by performing tandem affinity purification of K1 from HEK-293 cells and subjecting cellular proteins bound to K1 to mass spectrometry . Stable cell lines expressing a FLAG and HA double epitope-tagged version of K1 and EV HEK-293 cells were generated as previously described [50] . For tandem affinity purification , FLAG-HA-K1 or EV HEK-293 expressing cells were lysed with NP40 buffer . The clarified lysates were incubated with anti-FLAG M2 affinity gel , washed with NP40 buffer , and then eluted with 3X FLAG peptide . The eluates were subsequently incubated with an anti-HA resin , washed with NP40 buffer and eluted . The eluates were resolved by sulfate polyacrylamide gel electrophoresis ( SDS-PAGE ) and Coomassie stained . Only bands that were present in the K1 lane and absent in the EV lane were isolated and submitted for MALDI/TOF/TOF mass spectrometry ( S3A Fig ) . A Western blot of the affinity-purified eluate was also performed to confirm successful pull-down of FLAG-HA-K1 ( S3B Fig ) . Using mass spectrometry , we found AMPKγ1 associated with K1 ( Fig 4A ) . We also observed an association between K1 and heat-shock protein 90 ( HSP90 ) ( Fig 4A ) , which confirmed our previous report on the association of K1 with HSP90 [50] . Next , we constructed a V5 epitope-tagged AMPKγ1 ( AMPKγ1 ) in pcDNA3 vector . To confirm the association between K1 and AMPKγ1 as determined by mass spectrometry , we transiently expressed AMPKγ1 in HEK-293 cells stably expressing EV or FLAG epitope-tagged K1 ( K1 ) . We performed a co-immunoprecipitation by incubating EV- or K1-expressing HEK 293 clarified lysates containing equal amounts of protein with anti-V5 antibody to pull down the V5 epitope-tagged AMPKγ1 . We detected the multimer and monomer forms of K1 co-immunoprecipitating with AMPKγ1 in K1-expressing cells , but not from EV control cells ( Fig 4B ) . To further substantiate the association between K1 and AMPKγ1 , we performed the reverse immunoprecipitation and immunoprecipitated K1 . We observed that AMPKγ1 co-immunoprecipitated with K1 from K1-expressing cells but not from EV control cells ( Fig 4C ) . In the cell , AMPKγ1 complexes with AMPKα1 and AMPKβ1 . In addition to AMPKγ1 , we next wanted to determine whether the other AMPK subunits were part of the protein complex associated with K1 . We transfected V5-AMPKγ1 in HEK-293 cells stably expressing empty vector ( EV ) or FLAG-K1 ( K1 ) , immunoprecipitated V5-AMPKγ1 , and probed for endogenous AMPKα1 and AMPKβ1 . In addition to detecting K1 as we previously observed , we also observed the expression of AMPKα1 and AMPKβ1 ( Fig 5A ) , suggesting that there is an association between K1 and the three subunits of AMPK . Next , we determined whether we could detect association of K1 and endogenous AMPK . As the commercially available AMPKγ1 antibody is not appropriate for immunoprecipitation , we immunoprecipitated endogenous AMPKβ1 and probed for K1 ( Fig 5B ) . Upon AMPKβ1 immunoprecipitation , we detected K1 , which further confirmed an association between K1 and the endogenous AMPK complex ( Fig 5B ) . K1 has an immunoglobulin-like N-terminus , a transmembrane region , and a cytoplasmic tail containing an ITAM . To identify the region of K1 that associates with AMPKγ1 , we performed co-immunoprecipitations of various FLAG-tagged K1 domain deletion mutants and V5-tagged AMPKγ1 . We transiently expressed V5-AMPKγ1 ( AMPKγ1 ) along with one of the following in HEK-293 cells: K1WT , K1 lacking the C-terminus ( K1ΔCT ) , or lacking the N-terminus ( K1ΔNT ) . We also transfected an equivalent amount of EV ( pcDNA3 ) and K1WT as a control . The construction of the FLAG-tagged K1 mutants has previously been described [50] . We immunoprecipitated K1WT , K1ΔCT , or K1ΔNT and probed for V5-tagged AMPKγ1 ( Fig 5C ) . As previously observed , we detected co-immunoprecipitation of AMPKγ1 and K1WT ( Fig 5C , lane 1 ) . We also detected co-immunoprecipitation of AMPKγ1 and K1ΔCT indicating that the K1 C-terminus is not important for K1 and AMPKγ1 association ( Fig 5C , lane 2 ) . We did not observe co-immunoprecipitation of AMPKγ1 and K1ΔNT suggesting that AMPKγ1 associates with K1 via the K1 N-terminus ( Fig 5C , lane 3 ) . About 10–20% of K1 is localized to the plasma membrane with the major fraction of K1 being found in the endoplasmic reticulum [51] . K1 can also be internalized and internalization is associated with K1 signaling [52] . Additionally , all three subunits of AMPK have been shown to localize to the cellular membrane fraction [27] . We wanted to evaluate the localization of the endogenous AMPK subunits in EV- and K1-expressing stable HEK-293 cells . We lysed equal numbers of EV- and K1-expressing HEK-293 cells and separated the cellular fractions . The fractions were then resolved by SDS-PAGE and Western blot . The blots were probed using antibodies specific for each AMPK subunit and isoform . In addition to being localized to the cytoplasm , we observed that AMPKα , AMPKβ1 , and AMPKγ1 were detected in the cellular membrane fraction ( Fig 6A ) . Along with the AMPK subunits , K1 was also found in the membrane fraction ( Fig 6A ) . We evaluated the purity of the cytoplasmic , membrane , and nuclear fractions by probing for MEK1/2 , K1 , and histone H3 respectively ( Fig 6A ) . These proteins are restricted to each of these fractions . Based on these findings , we conclude that K1 and multiple AMPK subunits and isoforms ( without over expression ) are localized to the cellular membrane fraction , suggesting that K1 associates with AMPK in the cellular membrane . We subsequently evaluated localization of FLAG-K1 ( K1 ) and endogenous AMPK by immunofluorescence staining . Because there is no available AMPKγ1 antibody that was appropriate for immunofluorescence staining , and we had determined that AMPKβ1 co-localized with K1 ( Fig 5B ) , we stained for K1 and endogenous AMPKβ1/2 in EV- or K1- stably expressing HUVEC . We fixed the cells with formaldehyde , washed and then permeabilized the cells with Triton-X-100 . We stained cells with FLAG-FITC to detect FLAG-K1 and AMPKβ1/2 antibodies followed by an anti-rabbit Alexa Fluor 647 secondary antibody . By confocal microscopy , we acquired z-stacks on fully stained EV and FLAG-tagged K1 HUVEC ( Fig 6B ) . We also completed z-stacks on controls containing only the secondary anti-rabbit AF647 in order to demonstrate that the staining for AMPKβ1/2 is specific and not due to non-specific secondary staining . We observed co-localization of K1 and endogenous AMPKβ1/2 in the perinuclear area ( Fig 6B ) as determined by a Mander’s overlap coefficient of 0 . 83 , which was determined using ImageJ software . The EV transfected cells stained positive for AMPKβ1/2 but not K1 , as expected . As described above , we observed ( Fig 5C ) that K1ΔNT does not associate with AMPKγ1 by co-immunoprecipitation . This lack of association may be due to the fact that the two proteins do not interact or because K1ΔNT is mislocalized in the cell . In order to confirm that K1ΔNT is expressed in the membrane but still does not interact with AMPK , we transiently expressed K1ΔNT , and fractionated the cell lysates into cytoplasmic , membrane , and nuclear components . We then evaluated K1ΔNT expression in the membrane fraction by immunoblotting for K1ΔNT . We observed that K1ΔNT is expressed in the membrane fraction ( Fig 6C ) . Next , we transiently transfected the K1 domain deletion mutants and AMPKγ1 . We separated the lysates into membrane , cytoplasmic , and nuclear fractions . We removed detergent from the membrane fraction by spin column removal and completed a bicinchoninic acid assay ( BCA assay ) to determine the protein concentrations for each sample . Using equal amounts of protein , we immunoprecipitated EV ( pcDNA3 ) , K1WT , K1ΔCT , or K1ΔNT and probed for V5-tagged AMPKγ1 in the membrane fraction . We detected K1WT and AMPKγ1 association , but we did not detect K1ΔNT and AMPKγ1 association in the membrane fraction ( Fig 6D ) corroborating our previous findings . Thus far , we have observed that KSHV K1 promotes survival in stressed cells , and K1 associates with AMPK via the K1 N-terminus . We next wanted to determine whether the association of K1 and AMPK is important for the survival advantage observed in stressed cells . We generated lentivirally transduced HEK-293 cells stably expressing FLAG epitope-tagged K1WT , K1ΔCT , K1ΔNT and empty vector ( EV ) . We treated these cells with the AMPK inhibitor , compound C , and evaluated cell viability using the MTS assay . We observed an increased percentage of viable K1WT expressing cells when AMPK was inhibited , compared to cells expressing K1ΔNT . This result suggests that the association between K1 and AMPK is important for survival in stressed cells ( Fig 7A ) . Surprisingly , we also observed that K1ΔCT expressing cells appear sensitive to AMPK inhibition , indicating that the K1-C terminus is also important for survival in stressed cells ( Fig 7A ) . We confirmed expression of the various K1 constructs by completing a Western blot using an anti-FLAG antibody or an anti-K1 antibody ( Fig 7B ) . When we immunoblotted with an anti-FLAG antibody , we observed low levels of K1ΔNT . Thus , we re-probed this blot using an anti-K1 antibody , and saw expression of K1ΔNT but no K1ΔCT expression since the K1 antibody is directed towards an epitope on the K1 C-terminus ( Fig 7C ) . We found that K1 associates with AMPK and this association is important for the survival advantage in stressed cells . We next wanted to determine the status of AMPK activity in stressed EV and K1 expressing cells . We exposed HUVEC stably expressing empty vector ( EV ) or FLAG-tagged K1 ( K1 ) to media without serum and growth factors containing either compound C or DMSO control for 24 hours . We then performed an AMPK-specific kinase activity assay . We incubated lysate from EV or K1 expressing HUVEC with or without an AMPK substrate , a synthetic peptide called SAMS peptide ( HMRSAMSGLHLVKRR ) , AMP , and radiolabeled γ-32P-ATP [53] . We next evaluated the incorporation of radiolabeled phosphate from γ-32P-ATP into SAMS peptide . Compound C treatment resulted in an overall reduction in AMPK activity in both EV and K1 expressing cells compared to untreated cells . However , in the presence of compound C , we observed a higher degree of AMPK activity in K1 expressing cells compared to EV expressing cells ( Fig 8A ) . This data suggests that K1 expression promotes AMPK activity in stressed cells . We also confirmed that K1 expression is not altered by nutrient deprivation and compound C treatment by Western blot analysis ( Fig 8B ) .
Cell survival during KSHV infection is paramount to the establishment of life-long infection . Upon infection , KSHV primarily enters a latent state . During latency , KSHV expresses a limited number of proteins and microRNAs that enable it to successfully persist in the cell by avoiding the immune response and by promoting cell survival [54] . When KSHV reactivates and enters the lytic stage , the cell must remain viable throughout viral replication and virion assembly so that infectious virions are generated . Cell death prior to completion of the lytic program would result in defective viral replication . We observed that KSHV infected cells containing a WT K1 gene had a survival advantage compared to cells infected with KSHV K1 mutants following exposure to stress . To explore the underpinnings of this phenotype , we completed tandem affinity purification and mass spectrometry to identify K1-associating proteins . We identified AMPKγ1 as a K1-associating protein . We corroborated this finding independently and found that K1 co-immunoprecipitated with AMPKγ1 . By performing co-immunoprecipitations of AMPKγ1 and K1 domain mutants , we found that the K1 N-terminus is important for K1 and AMPKγ1 association . The association between K1 and AMPK is important for the survival advantage in stressed cells because we observed reduced cell viability in cells expressing K1ΔNT compared to K1WT . Our studies indicate that KSHV K1 promotes survival via its association with AMPK , and KSHV K1 facilitates AMPK activity in stressed cells . To explore alternative possibilities for not observing the association between K1ΔNT and AMPKγ1 , we confirmed that K1ΔNT is not mislocalized in the cell and is expressed in the membrane by performing a Western blot for K1ΔNT in the membrane fraction . Another explanation for not observing association between AMPKγ1 and K1ΔNT may be that K1ΔNT may not fold correctly due to the lack of the N-terminus . Under normal culture conditions , our lab and others have shown that K1 activates the PI3K/Akt/mTOR pathway [13 , 18] . It has been reported that cells having overly active Akt and consequently a high glycolytic rate are more sensitive to cell death following starvation compared to control cells [55] . Moreover , when starved-cells are treated with an activator of AMPK , cells with active Akt are protected from cell death [55] . KSHV-infected cells also have an active PI3K/Akt/mTOR pathway and a high glycolytic rate [56–58] . Furthermore , simultaneous activation of AMPK , Akt and mTOR , has been observed in liver cancer cells following nutrient starvation [59] . Thus , low levels of AMPK activation may promote metabolic adaptation and consequently , increase KSHV-infected cell survival . We propose that K1 promotes AMPK activity during metabolic stress and in this way enhances KSHV-infected cell survival . We observed a modest reduction in the number of viable KSHV-K1WT cells following knock down of the catalytic subunits , AMPKα1 and AMPKα2 , compared to KSHV-K1WT cells treated with non-specific siRNA ( S2A Fig ) . This data suggests that in addition to promoting cell survival via AMPK , K1 likely contributes to the survival advantage observed in KSHV-K1WT cells by activating other pro-survival pathways such as the PI3K/Akt pathway as has been previously described [18] [19 , 20] . Thus , K1 promotes cell survival by multiple methods , including its association with AMPKγ1 . One of the K1-interacting proteins previously identified is Syk , a Src homology 2 ( SH2 ) -containing protein tyrosine kinase [13] . Activation of Syk in B cell lymphomas is associated with cell survival since inhibition of Syk is clinically efficacious in treating lymphoma [60] . Moreover , Syk activity is critical for the normal functioning of endothelial cells and vascular integrity in vivo . Syk deficient mice exhibit petechiae in utero , as well as , a reduced number of endothelial cells that are morphologically defective , indicating that Syk activity is critical for endothelial cell survival and function [61] . In vitro , Syk activity is important for HUVEC proliferation and migration [62] . Huang et al . found that AMPK associates with Syk and induced its activation [63] . Although the association of AMPK and Syk has not been evaluated in HUVEC , we may speculate that the association of K1 , AMPK , and Syk may facilitate downstream signaling cascades that promote cell survival . The role of AMPK during herpesvirus infection is complicated , and whether it promotes viral replication or inhibits it may depend on a variety of factors . During human cytomegalovirus ( HCMV ) infection , AMPK has been found to promote a metabolic environment that is conducive to viral replication [64 , 65] . HCMV infection augments glycolysis and AMPK inhibition blocks increased glycolysis that is induced by HCMV infection and AMPK inhibition also hinders viral DNA synthesis [65] . During human herpes simplex virus-1 ( HSV-1 ) infection , AMPK activity facilitates neuron survival and reduces viral production [66] . Thus , AMPK appears to impact viral production differently during HCMV and HSV-1 infection . Recently , Cheng et al . observed that endogenous AMPKα1 inhibits KSHV replication following primary infection [67] . AMPK does not seem to affect KSHV infectivity nor trafficking to the nucleus , but it does have an inhibitory effect on KSHV lytic gene expression since knockdown of AMPK results in increased expression of some KSHV lytic genes and corresponding proteins [67] . Thus , AMPK appeared to inhibit the KSHV lytic cycle but not the establishment of latency . Our studies examine the role of AMPK in a different context i . e . in latently infected cells expressing WT or mutant K1 under conditions of metabolic stress . We report that AMPK activity contributes to survival of latent KSHV-infected cells . Because AMPK can be activated by a variety of cellular stresses and can impact multiple cell signaling pathways , it is highly plausible that AMPK can differentially impact the infected-cell depending on the life-cycle of the virus . There are also multiple isoforms of AMPK , and various combinations of these isoforms can form different heterotrimeric complexes . We do not yet understand how these different AMPK combinations modulate AMPK function , and this is an additional complexity that we need to understand in the future in order to evaluate AMPK’s function during the KSHV lifecycle .
HEK-293 ( ATCC , CRL-1573 ) and iSLK cells [46] were maintained in Dulbecco’s Modified Eagle Medium ( DMEM ) and human telomerase-reverse transcriptase-immortalized human umbilical vein endothelial cells ( hTERT-HUVEC ) [17] were cultured in endothelial growth basal medium ( EBM-2 ) from Lonza and supplemented with an endothelial cell growth medium ( EGM-2 ) bullet kit without heparin and ascorbic acid supplements . All cell lines were supplemented with 10% heat-inactivated fetal bovine serum ( HI-FBS ) , 1% penicillin-streptomycin ( PS ) , and 1% L-glutamine and maintained at 37°C and 5% CO2 . Additionally , EV ( pcDNA3 ) or FLAG-K1 stably expressing HEK-293 cells were maintained in 1 mg/mL G418 . For transfection of HEK-293 cells , cells were transfected with 10 μg pcDNA3-V5-AMPKγ1 or vector per 10 cm plate using XtremeGENE HP reagent according to the manufacturer’s instructions . The AMPK inhibitor , compound C , was purchased from Calbiochem and suspended in dimethyl sulfoxide ( DMSO ) . HEK-293 cells stably expressing EV or FLAG-HA-K1 were made as previously described [50] . Briefly , FLAG-HA was cloned following the signal peptide sequence on the N-terminus of K1 ( Accession# AAG01599 . 1 ) . The FLAG-HA-K1 was then cloned into the pcDNA3 vector . A V5-epitope tag was added to the N-terminus of PRKAG1 ( NP_002724 ) by PCR and then cloned into the pcDNA3 vector . The pcDNA3-K1 WT that had previously been made [68] was used as a template for preparing the K1 domain deleted mutants , which have been previously described [50] . To construct pcDNA3-FLAG-K1ΔCT , pcDNA3-K1 WT ( amino acids 1–303 ) and the following primers , for-5’-CGCCCGAAGCTTATGGCCCTGCCCGTGACCGCCCTG-3’ and rev-5’-CGCCACAAGGTTTCAGTACCAATCCACTGGTTG-3’ were combined for amplification by PCR . The PCR product was then cut with HIND III and cloned into pcDNA3 . pcDNA3-FLAG-K1ΔCT lacks the amino acids 266–303 . To construct pcDNA3-K1ΔNT-FLAG , pcDNA3-K1 WT was combined with primers , for-5’-CATCTTGCATCCAGTATTTATGACAC-3’ and rev-5’-CGCCGCTCTAGATTCCACTGGTTGCG-3’ . The PCR product was then cut with Bam HI and XbaI , and cloned into pcDNA3 . pcDNA3-K1ΔNT−FLAG lacks amino acids 1–241 . The pcDNA3-K1 WT construct was used as a template for making the lentiviral K1 WT and mutant constructs . pLenti-FLAG-K1 domain mutants were generated using Q5 Site-Directed Mutagenesis kit ( Q5 ) by New England Biolabs Inc and appropriate primer sets were designed according to the Q5 kit specifications . FLAG-K1 , FLAG-K1ΔCT , and FLAG-K1ΔNT were cloned into the lentiviral vector , pLenti CMV Puro DEST . Empty vector is pLenti CMV Puro . DEST . pcDNA3-FLAG-HA-K1 or pcDNA3 empty vector were transfected into HEK-293 cells and selected in media containing 1 mg/mL G418 . All lentiviruses were prepared using the Virapower lentiviral expression system as per the manufacturer’s instructions ( Invitrogen ) . hTERT-HUVEC or HEK-293 cells were cultured to approximately 80% confluency in a 6-well dish . At the time of lentiviral transduction , cells were rinsed with PBS and 2 mLs of lentivirus ( unconcentrated ) was added in the presence of 10 μg/mL polybrene . The cells were centrifuged for 90 minutes at 3000 RCF at 30°C . The cells were then incubated overnight at 37°C and 5% CO2 . The media containing lentivirus was replaced with the appropriate fresh media the following day . Forty-eight hours following lentiviral transduction , HUVEC cells underwent selection with 0 . 5 μg/mL puromycin for 1 week . Transduced HEK-293 cells underwent selection with 1 μg/mL puromycin for 1–2 weeks . The FLAG HA Tandem Affinity Purification Kit by Sigma was used . FLAG-HA-K1 or EV HEK-293 expressing cells were lysed and the lysates were incubated with anti-FLAG M2 affinity gel , washed and eluted with 3X FLAG peptide . The eluates were subsequently incubated with an anti-HA resin , washed with NP40 buffer and eluted . The eluates were resolved by sulfate polyacrylamide gel electrophoresis ( SDS-PAGE ) and Coomassie stained . Bands present in the K1 sample but absent in the EV were submitted to the Harvard Mass Spectrometry core for MALD/TOF/TOF mass spectroscopy analysis . EV or FLAG-HA-K1 stably expressing HEK-293 cells were plated at 650 , 000 cells/well in a 6-well dish or at 60 , 000/well in a 24-well dish . The next day , the media was removed and replaced with complete media containing the relevant concentration of compound C . Six to eight hours later the media from each well was collected; the cells were gently washed with PBS and trypsinized . The cells were pelleted and resuspended in 0 . 6–1 mL complete media . An aliquot was removed for the trypan blue exclusion assay . The cells were then pelleted , the supernatant was discarded and the pellets were immediately frozen and maintained at -80°C until utilized for the caspase-3 assay . HUVEC infected with KSHV containing WT K1 or mutant K1 were plated at 20 , 000 cells per well of a 24-well plate in EBM-2 without serum and growth factors . Cells were not previously washed . iSLK cells containing WT K1 or mutant K1 were plated in a similar manner , but in complete media . The next day , the iSLK media was replaced with serum-free DMEM . At the time of counting , cells were trypsinized and cell suspension was then combined with trypan blue ( 0 . 4% Sigma Aldrich ) at a 1:1 dilution . Each sample was counted in duplicate or triplicate using a hemacytometer . The previously frozen cell pellets were thawed on ice . The caspase-3 assay was then completed based on the manufacturer’s instructions ( ApoAlert Caspase-3 Fluorescent Assay by Clontech Laboratories ) . Briefly , the pellets were lysed and maintained on ice followed by centrifugation . The clarified supernatant was then assayed for active caspase-3 and fluorescence was determined using the CLARIOstar plate reader ( BMG Labtech ) . Active caspase-3 concentrations were determined using a standard curve , and active caspase-3 values were then further normalized to cell number . Five thousand cells per 100 μL were counted and resuspended in EBM-2 containing 30 μg/mL hygromycin but lacking all other supplements . Cells were plated in triplicate and incubated for 24 , 48 and 72 hours . Cell proliferation was determined using the Cell Titer 96 Aqueous One Solution Cell Proliferation Assay ( Promega ) according to the manufacturer’s instructions . Stored aliquots of previously frozen MTS were thawed in a water bath at 37°C . Twenty microliters of MTS reagent was then dispensed into each well using a multichannel pipet . The plate was then gently shaken for 30 seconds and placed in an incubator at 37°C for 2–3 hours . At the end of incubation , the plate was gently tapped to mix the formazan product . The absorbance was then immediately measured at 490 nm using a CLARIOstar plate reader ( BMG Labtech ) . Wells that contain only media were subtracted as background from all OD values . Elevated absorbance values are indicative of metabolically active cells . For compound C treatments , we normalized to vehicle treated cells . For the starvation of HUVEC , the same number of cells was plated . In these experiments OD values are not normalized but show relative optical density values . HEK-293 stably expressing EV ( pcDNA3 ) or FLAG-HA-K1 ( pcDNA3 ) were washed , trypsinized , centrifuged and counted . Five million cells were prepared using a cell fractionation kit according to the manufacturer’s instructions ( Cell Signaling Technology ) . Equal volumes of each lysate from each fraction for EV and K1 were loaded and resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-PAGE ) and then transferred to a nitrocellulose membrane . Cells were harvested , washed twice with PBS , and then lysed in buffer containing 0 . 5% NP40 , 150 mM NaCl , 50 mM Tris-HCL pH 8 . 0 , and a cocktail of proteinase ( Roche ) and phosphatase ( Roche ) inhibitors . The lysates used to evaluate K1 protein expression were frozen and thawed two times . Protein concentrations were determined by Bradford assay . Equal amounts of protein ( 15–25 μg ) were loaded per lane and resolved by SDS-PAGE and then transferred to a nitrocellulose membrane . The following antibodies from Cell Signaling Technology were used: AMPKα #2603 , AMPKα1 #2795 , AMPKα2 #2757 , AMPKβ1 #12063 , AMPKβ1/2 #4150 , AMPKγ1#4187 , HRP-linked anti-rabbit IgG #7074 and HRP-linked anti-mouse IgG #7076 . In some experiments K1 expression was confirmed by immunoblotting with an HRP-conjugated mouse monoclonal anti-FLAG M2 antibody from Sigma #F1804 . The K1 monoclonal antibody was made by immunization with the peptide , KQRDSNKTVP , protein ID#AAB71616 ( gene accession #U86667 ) . Lysates containing equal amounts of protein as determined by the Bradford assay or bicinchoninic acid assay ( BCA assay ) were combined with EZview Red anti-FLAG M2 affinity gel ( Sigma , F2426 ) as per the manufacturer’s instructions . For V5-AMPKγ1 immunoprecipitation , protein A/G agarose ( Santa Cruz , sc-2003 ) was combined with monoclonal anti-V5 antibody ( Sigma , V8012 ) , which was used at 1ug of antibody/1 mg of protein , and rotated overnight at 4°C . The supernatant was removed and the affinity gel or agarose pellets were washed by adding 1mL of 0 . 1% NP40 lysis buffer followed by 5 minutes rotation at 4°C for a total of 4 times . For immunoprecipitation using the membrane fraction , affinity gel pellets were washed with 1mL of 0 . 1% NP40 lysis buffer , followed by 5 minutes rotation at 4°C for 3 times and with 1mL 0 . 5% NP40 lysis buffer followed by 5 minutes rotation at 4°C for one time . Detergent was then removed from the membrane fraction using Pierce Detergent spin columns . FLAG-K1 and/or FLAG-K1 domain deleted mutants were eluted using 3X FLAG peptide ( Sigma , F4799 ) as per the manufacturer’s instructions . Laemmli buffer ( 2X ) was added to the FLAG-K1eluate or directly to the V5-AMPKγ1/agarose samples ( 1:1 ) and all samples were heated at 100°C for 6 minutes . Proteins were resolved by SDS-PAGE followed by Western blot . HUVEC stably expressing EV or FLAG-K1 were incubated in EBM-2 without serum and growth factors for 24 hours . Either 5 uM compound C or DMSO ( 0 . 05% ) control was added at the start of starvation . Cells were washed with cold PBS and then lysed with a buffer containing 50mM Tris-HCL pH 7 . 4 , 1 mM EDTA , 1 mM EGTA , 250 mM mannitol , 1% Triton X-100 and proteinase ( Roche ) and phosphatase inhibitors ( Roche ) . Lysates were then clarified by centrifugation . For the AMPK activity assay , reagents were purchased from SignalChem and the manufacturer’s protocol followed . Briefly , 10 μLs of cell lysate was incubated with 5 μL of 1 mg/mL SAMS or peptide substrate solution , 5 μLs 0 . 5 mM AMP solution and 5 μLs γ-32P-ATP assay cocktail . Gamma-32P-ATP was purchased from Perkin Elmer . The mixture was incubated at room temperature for 30 minutes and then 20 μLs was added to phosphocellulose paper and washed 2 times in 1% phosphoric acid solution . Counts per minute ( cpm ) were acquired using a PerkinElmer liquid scintillation analyzer . Approximately 120 , 000 HUVEC cells stably expressing EV or FLAG-K1 were plated in MatTek 35 mm glass-bottom dishes . Cells were washed with PBS and fixed by 15-minute incubation in 3 . 7% formaldehyde at room temperature . Cells were washed 3X with PBS and then permeabilized by 15-minute incubation at room temperature in 0 . 2% Triton-X 100/PBS . Cells were then washed again 3X with PBS and then blocked in 10% bovine serum albumin ( BSA ) PBS for 30 minutes . Cells were stained 1:200 with a directly conjugated FITC-ECS ( DDDDK ) polyclonal antibody ( Bethyl laboratories ) and anti-AMPKβ1/2 ( 1:50 , Cell Signaling ) in 10% BSA for 1 hour at room temperature . Cells were washed 2X quick followed by 3X 5-minute washes . Samples were then incubated with anti-rabbit Alexa Fluor 647 ( 1:600 ) secondary antibody in 10% BSA at room temperature for 1 hour . All samples were then stained with DAPI for 1 minute and washed . Fluorescent images were acquired by taking z-stacks using a 63X oil objective on a Zeiss 700 confocal microscope . The overlap coefficient according to Manders ( R ) was determined using Image J . Confocal images from 5 cells in their entirety were evaluated for co-localization by calculating the overlap coefficient . The construction of the KSHV WT and mutant recombinant viruses has previously been described in detail [46] . Briefly , the BAC 16 was kindly provided by Dr . Jae U . Jung . pcDNA3 WT and mutant K1 constructs were used as templates for construction of recombinant viruses . pcDNA3 WT K1FLAG , which has a FLAG tag at the N-terminus , was constructed as previously described [68] . The pcDNA3-K15XSTOP construct has 3 stop codons following the start codon of WT K1 FLAG and two TGA stop codons replacing ATG start codons at positions 481 and 763 . K15XSTOP was inserted into BamHI and EcoRI sites of pcDNA3 WT K1FLAG . The original K1 gene is located within the BAC16 genome at position 105 to 959 . KSHV-K1REV was made by replacing the mutant K1 gene from KSHV-K15XSTOP with a RpsL-Neo cassette and then replaced the RpsL-Neo cassette with WT K1FLAG . The recombinant viruses containing KSHV-K1WT , KSHV-K15XSTOP , KSHVΔK1 and KSHV-K1REV were made utilizing the Red/ET recombination system ( Gene Bridges Inc ) as per the manufacturer’s instructions . KSHVΔK1 was constructed by replacing the K1 gene with the linear RpsL-neo cassette that is flanked by homologous arms [46] . Five × 105 cells of WT or recombinant virus infected iSLK cells were plated in one well of a 6- well plate overnight after which cells were reactivated with 3 μg/ml doxycycline and 1 mM sodium butyrate for 3 days . Supernatant was collected and cleared by centrifugation at 950g for 10 min and filtered through a 0 . 45 μm filter . iSLK cells were infected as previously described [46] and maintained in the presence of 1 μg/ml puromycin , 250 μg/mL G418 , and 1 . 2 mg/mL hygromycin [46] . In order to infect HUVEC , the filtered viral supernatants from reactivated iSLK cells were incubated with the immortalized HUVEC cells in the presence of 8 μg/ml of polybrene and centrifuged for 2 hours at 3000 RCF at 30°C . The cells were then placed in an incubator with 5% CO2 at 37°C . At 48 hours post-infection , 30 μg/ml hygromycin was added to the media to select for HUVEC stably infected with KSHV-K1WT , KSHV-K15XSTOP , KSHVΔK1 or KSHV-K1REV ( revertant ) recombinant viruses . The infected HUVEC cells were also maintained in the presence of 30 μg/ml hygromycin .
|
Infectious agents such as Kaposi’s sarcoma associated herpesvirus ( KSHV ) are etiologic agents of human cancer . KSHV-infected cells must survive various environmental stresses . Cells infected with KSHV express viral proteins that alter normal cellular processes to promote cell survival and viral persistence . We found that the KSHV K1 protein promotes survival under conditions of cellular stress , and that this survival advantage is at least partially dependent on the association of K1 and the cellular protein AMP-activated protein kinase ( AMPK ) . We also observed increased AMPK activity in K1-expressing cells compared to EV following exposure to metabolic stress . Several reports suggest that AMPK signaling may contribute to tumor development by promoting cell survival . Our results suggest that KSHV K1 modulates cellular AMPK function to enhance the survival of KSHV-infected cells in order to promote viral persistence .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"physiology",
"colorimetric",
"assays",
"medicine",
"and",
"health",
"sciences",
"cell",
"viability",
"testing",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"cell",
"metabolism",
"biochemical",
"analysis",
"viruses",
"enzyme",
"assays",
"immunoprecipitation",
"dna",
"viruses",
"bioassays",
"and",
"physiological",
"analysis",
"co-immunoprecipitation",
"cell",
"nucleus",
"cellular",
"structures",
"and",
"organelles",
"herpesviruses",
"research",
"and",
"analysis",
"methods",
"nuclear",
"membrane",
"cell",
"analysis",
"medical",
"microbiology",
"microbial",
"pathogens",
"kaposi's",
"sarcoma-associated",
"herpesvirus",
"mts",
"assay",
"cell",
"membranes",
"precipitation",
"techniques",
"cell",
"biology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2016
|
The KSHV K1 Protein Modulates AMPK Function to Enhance Cell Survival
|
Human cytomegalovirus ( HCMV ) , a β-herpesvirus , has evolved many strategies to subvert both innate and adaptive host immunity in order to ensure its survival and propagation within the host . Induction of IL-8 is particularly important during HCMV infection as neutrophils , primarily attracted by IL-8 , play a key role in virus dissemination . Moreover , IL-8 has a positive effect in the replication of HCMV . This work has identified an HCMV gene ( UL76 ) , with the relevant property of inducing IL-8 expression at both transcriptional and protein levels . Up-regulation of IL-8 by UL76 results from activation of the NF-kB pathway as inhibition of both IKK-β activity or degradation of Ikβα abolishes the IL-8 induction and , concomitantly , expression of UL76 is associated with the translocation of p65 to the nucleus where it binds to the IL-8 promoter . Furthermore , the UL76-mediated induction of IL-8 requires ATM and is correlated with the phosphorylation of NEMO on serine 85 , indicating that UL76 activates NF-kB pathway by the DNA Damage response , similar to the impact of genotoxic drugs . More importantly , a UL76 deletion mutant virus was significantly less efficient in stimulating IL-8 production than the wild type virus . In addition , there was a significant reduction of IL-8 secretion when ATM -/- cells were infected with wild type HCMV , thus , indicating that ATM is also involved in the induction of IL-8 by HCMV . In conclusion , we demonstrate that expression of UL76 gene induces IL-8 expression as a result of the DNA damage response and that both UL76 and ATM have a role in the mechanism of IL-8 induction during HCMV infection . Hence , this work characterizes a new role of the activation of DNA Damage response in the context of host-pathogen interactions .
Human cytomegalovirus ( HCMV ) is a β-herpesvirus that infects healthy individuals , usually asymptomatically , but can cause severe or fatal disease in immunocompromised individuals . Primary HCMV infection , as with other herpesviruses , is followed by establishment of lifelong latency with periodic reactivation and , in order to successfully establish itself in the host , the virus has evolved a broad range of host evasion strategies , modulating not only innate and adaptive immunity , but also host cell biology , for example , the cell cycle and apoptosis [1] . The induction of the interleukin-8 ( IL-8 ) during HCMV infection is particularly important for viral replication and possibly contributes to the efficient dissemination of the virus by neutrophils [2] , [3] . Interleukin-8 is a pro-inflammatory chemokine that attracts primarily neutrophils , and also monocytes and cytotoxic T cells , by interacting with the CXC chemokine receptors CXCR1 and CXCR2 [4] . Although expression of IL-8 is low or absent under normal conditions , it is highly inducible by a wide range of extracellular stimuli , such as the pro-inflammatory cytokine IL-1 , the tumor necrosis factor alpha ( TNFα ) [5] , bacteria and viruses [6] , [7] . Besides its relevant role in inflammation , IL-8 is a key component in several viral infections , modulating viral dissemination and virus replication , in part due to inhibition of the impact of interferon-α [8] . On the other hand , excessive amounts of locally produced IL-8 can have deleterious effects , and so IL-8 gene expression is tightly controlled at both transcriptional and post-transcriptional levels . Activation of IL-8 expression in the majority of cell types is critically controlled by the NF-kB transcription factor . The AP-1 and NF-IL-6 transcription factors may also contribute to optimal IL-8 activation , depending on the stimulus or the cell type [4] . The NF-kB canonical pathway involves the activation of the IKK complex , consisting of two catalytic kinase subunits , IKKα and IKKβ , and a regulatory subunit , IKKγ/NEMO . In most unstimulated cells , NF-kB dimers ( mostly p65/p50 dimers ) are localized in the cytoplasm as a complex with the IkB proteins . Upon stimulation , IkB is phosphorylated by the IKK complex , ubiquitinated and targeted for degradation , thus releasing the NF-kB subunits that translocate to the nucleus and induce transcription of target genes [9] . Although most of the physiological inducers of NF-kB involve the canonical pathway , alternative mechanisms leading to NF-kB nuclear localization and DNA binding have been identified . One of these pathways is induced by activation of the DNA damage response and , in contrast to inflammatory stimuli such as TNFα or IL-1β , the signal originates in the nucleus [10] . Activation of NF-kB by genotoxic stress requires induction of two independent parallel pathways . The first one triggered upon DNA damage results in the phosphorylation and activation of ATM , a nuclear protein kinase which regulates cell cycle checkpoints in response to DNA double-strand breaks [11] . The second pathway leads to SUMOylation of NEMO through a mechanism dependent on PARP1 , PIASy and Ubc9 . Activation of both these pathways leads to phosphorylation and ubiquitination of sumoylated NEMO in an ATM-dependent way . Ubiquitinated NEMO associated with ATM is exported back to the cytoplasm , activating the IKK complex and subsequent NF-kB activation in a similar manner to the canonical pathway [10] . The HCMV UL76 protein is virion-associated and expressed with late kinetics [12] . The corresponding gene belongs to the UL24 gene family , conserved in all herpesviruses and the only core gene without an assigned function [13] . Bioinformatic analysis identified UL24 gene family as a putative novel PD- ( D/E ) XK endonuclease [14] . This superfamily of restriction endonuclease-like fold proteins includes several restriction endonucleases ( e . g . EcoRI , EcoRII , BamHI , BglI , Cfr10I , NaeI ) , DNA repair enzymes ( MutH and Vsr ) , Holliday junction resolvases ( Hjc and Hje ) and other nucleotide-cleaving enzymes [15] . However , no endonuclease activity has been demonstrated experimentally for any of the UL24 homologues . Global mutational analysis of the HCMV genome classified UL76 as an augmenting gene for viral replication [16] , recently demonstrated to be involved in the regulation of the UL77 gene expression . Since UL77 is essential for viral replication , its regulation by UL76 may be important for efficient HCMV replication [17] . Expression of UL76 also induces cell cycle arrest at G2/M phase by inhibition of the mitotic Cdc2-cyclin B complex . Interestingly , this effect on the cell cycle is conserved in the UL24 human homologues representatives of the alpha , beta and gamma-subfamilies and the murine homologue from MHV-68 ( ORF20 ) [18] , [19] . The precise mechanism of cell cycle arrest induced by UL24 homologues remain to be clarified , but a recent report showed that HCMV UL76 induces chromosomal aberrations and DNA damage [20] . Here we identify a new function of UL76 , the induction of IL-8 expression , mediated by the ATM kinase and activation of the NF-kB pathway . Thus , activation of NF-kB by UL76 results from induction of the DNA damage response , similar to genotoxic drugs . Importantly , induction of IL-8 by HCMV is significantly reduced in the absence of ATM or in normal fibroblasts infected with a HCMV UL76 deletion mutant . Thus , viral infection induces IL-8 in a similar manner to the UL76 gene alone and UL76 is essential for maximal activation of IL-8 by HCMV .
The effect of UL76 on the activation of IL-8 transcription was demonstrated using a luciferase reporter construct containing the IL-8 promoter sequence . Transfection of a UL76 expression plasmid significantly activated transcription of IL-8 promoter in a dose-dependent manner ( Fig . 1A ) . Furthermore , cells expressing UL76 were demonstrated to secrete significantly higher levels of IL-8 as compared to the control vector ( p<0 . 01 ) ( Fig . 1B ) . In conclusion , UL76 induces IL-8 expression at both the level of transcriptional activation and protein secretion . Expression of IL-8 is tightly regulated at the transcriptional level . The sequence of nucleotides -1 to -131 in the proximal promoter region of IL-8 gene is essential for its transcription regulation and contains binding sites for NF-kB , AP-1 and NF-IL-6 transcription factors ( Fig . 2A ) [4] . To determine the mechanism of induction of IL-8 by UL76 , the luciferase activity of wild type IL-8 luciferase reporter was compared with its mutant derivatives containing a mutation in each of the three transcription factor binding sites . There was no significant difference in luciferase activity in response to co-transfection with UL76 when AP-1 or NF-IL-6 binding sites were mutated in the luciferase reporter construct , whereas IL-8 transcriptional activation was drastically reduced in the absence of the NF-kB binding site , indicating a critical role for the NF-kB pathway in the UL76-mediated induction of IL-8 ( Fig . 2B ) . Consistent with the previous results , expression of UL76 significantly activated an NF-kB responsive promoter ( Fig . 2C ) . To further characterize the activation of the NF-kB pathway by UL76 , we used a catalytically inactive mutant IKKβ and a mutant IkBα ( IkBαS32/36A ) in which the two critical serine residues were mutated to alanine , thus no longer permitting its phosphorylation and degradation . Both constructs function as dominant negatives , inhibiting the activity of cellular wild type IKKβ and IkBα , respectively . Co-transfection of each dominant negative with IL-8 luciferase reporter and UL76 expression plasmid or control vector resulted in a reduced induction of IL-8 in cells expressing UL76 ( Fig . 3A ) . After IkBα degradation by the proteosome , dimers of NF-kB subunits translocate to the nucleus where they bind to the target gene promoter region and activate transcription [9] . Thus , we evaluated the effects of UL76 expression on the subcellular localization of the NF-kB p65 subunit by immunofluorescence . As shown in Figure 3B , p65 was localized in the nucleus of the UL76-transfected HFF cells , in contrast to its cytoplasmic localization in control cells . This effect was not due to an increase in the expression of p65 ( Fig . 3C ) . Furthermore , similarly immunoblotting of 293T nuclear extracts with anti-p65 antibody revealed an accumulation of p65 in the nucleus of cells expressing UL76 ( Fig . 3D ) . This accumulation was specific , as can be seen from the constant levels of the nucleolin expression in the loading control . Consistent with these results , chromatin immunoprecipitation ( ChIP ) analysis demonstrated that expression of UL76 leads to NF-kB p65 binding to the IL-8 promoter ( Fig . 3E ) . In summary , IL-8 induction by UL76 requires a functional IKKβ and the degradation of IkBα to promote translocation of p65 subunit to the nucleus where it activates IL-8 transcription . Expression of UL76 results in an increased number of double stranded DNA breaks and phosphorylation of γH2AX , indicating activation of the DNA Damage response [20] . Consistent with these results , expression of UL76 results in activation of ATM and consequent phosphorylation of p53 and H2AX proteins ( Fig . 4A ) . Recently , several studies have characterized an alternative pathway to NF-kB activation that results from DNA damage . Based on this , we hypothesized that the ability of UL76 to induce DNA damage would lead to activation of NF-kB pathway and result in the induction of IL-8 expression . A characteristic feature of NF-kB pathway activation by genotoxic stress is the accumulation of IKKγ/NEMO in the nucleus where a series of post-translational modifications occurs [21] . The nuclear post-translational modifications of NEMO that are critical for NF-κB activation following genotoxic stress include ATM-independent sumoylation and ATM-dependent phosphorylation at serine 85 followed by monoubiquitination . Consistent with this mechanism , immunostaining using an anti-NEMO antibody revealed increasing amounts of nuclear NEMO in cells expressing the UL76-HA tagged protein ( Fig . 4B ) . Furthermore , immunoblotting of similarly transfected cells with a specific antibody to NEMO ( S85 ) phosphorylation demonstrated that expression of UL76 induces phosphorylation of NEMO as previously observed after genotoxic stress ( Fig . 4C ) . To evaluate the impact of the ATM kinase in IL-8 induction by UL76 , we used two different approaches: a specific ATM inhibitor , KU55933 , and a human fibroblast cell line deficient in ATM . The amount of IL-8 secreted by 293T cells expressing UL76 , or the control plasmid , in the presence or absence of KU55933 was determined by ELISA . Inhibition of ATM by KU55933 blocked the UL76-induced IL-8 secretion ( Fig . 4D ) . Similarly , IL-8 concentration was determined in supernatants of ATM -/- cells expressing UL76 or the control plasmid . As shown in Figure 4E , UL76 or etoposide , a genotoxic drug , are unable to induce IL-8 in the absence of ATM . Although UL76 expression leads to higher levels of IL-8 than etoposide stimulation , there is no increase in IL-8 secretion when compared to control vector , so this basal induction is possibly due to transfection . Moreover , this result is not due to the incapacity of the cell line to produce IL-8 since stimulation with TNFα , a membrane receptor-triggering NF-kB canonical pathway independent of ATM , is still capable of inducing IL-8 secretion ( Fig . 4E ) . Expression of UL76 was not affected in the ATM -/- cell line or in 293T cells cultured in the presence of the ATM inhibitor as confirmed by western blot ( Fig . 4D , E ) . In summary , these results indicate that activation of NF-kB pathway and consequent IL-8 induction by UL76 are ATM-dependent and result from activation of DNA damage . There is clear evidence that UL76 activates the DNA damage response , however , the mechanism employed by UL76 for this activation is still unknown . The prediction that the UL24 gene family encodes a novel PD- ( D/E ) XK endonuclease is a possible explanation [14] . Comparison of UL24 gene family sequences identified the three conserved PD- ( D/E ) XK signature amino acids of the endonuclease motif which are conserved in all homologues ( Fig . 5A ) [14] . A UL76 gene with these three critical amino acids mutated was constructed and used to determine the impact of the putative endonuclease activity on IL-8 induction . Levels of IL-8 secreted by cells expressing the mutant UL76 gene were reduced when compared to wild type UL76 gene; however , they were still significantly higher than control vector-expressing cells ( Fig . 5B ) . These results indicate that the putative endonuclease activity is not essential for the induction of IL-8 . In order to evaluate the impact of UL76 on the up-regulation of IL-8 in the context of HCMV infection , we used a previously described UL76 transposon mutant HCMV [16] . Supernatants from human fibroblasts infected with wild type HCMV AD169 BAC or UL76 mutant virus ( TNUL76 ) were collected at the indicated time points and secreted IL-8 was determined by ELISA . Consistent with previous studies [3] , [22] , HCMV infection resulted in high levels of IL-8 secretion during the course of the experiment ( Fig . 6A ) . Induction of IL-8 in cells infected with the UL76 mutant virus , however , was significantly reduced . At each time point , the amount of IL-8 secreted by cells infected with UL76 mutant virus ( TNUL76 ) was reduced by 42–52% compared with wild type HCMV . Equal infection by both viruses was confirmed by levels of the HCMV immediate-early 1 ( IE1 ) protein . Thus , UL76 is essential for optimal induction of IL-8 by HCMV . There was , however , no difference in the phosphorylation of ATM , NEMO and IkB proteins in cells infected with HCMV wild type compared to UL76-deficient virus ( data not shown ) . As the mutation of UL76 significantly increases the level of UL77 protein expression [17] , the effect of UL77 in the expression of IL-8 was evaluated by ELISA . In contrast to UL76-transfected cells , there is no induction of IL-8 in cells expressing UL77 . Moreover , UL77 has no inhibitory effect in the induction of IL-8 by different stimuli ( Fig . S1 ) . Overall , these results demonstrate that UL77 is not able to modulate IL-8 expression and thus , the reduction of IL-8 levels in cells infected with the UL76 mutant virus ( TNUL76 ) is not due to the regulation of UL77 expression by UL76 . A previous deletion mutant analysis of the IL-8 promoter in monocytic cells has shown that AP-1 and NF-kB transcription factors were required for optimal induction of IL-8 by HCMV [22] . The precise mechanism used by HCMV to induce IL-8 expression , however , is still not clear . As UL76 is required for maximal IL-8 induction by HCMV ( Fig . 6 ) and HCMV infection activates ATM [23] , [24] , we hypothesized that ATM would also have a role in IL-8 up-regulation during viral infection . To test this hypothesis , a primary fibroblast ATM -/- cell line was infected with HCMV AD169 BAC virus . Supernatants were collected at the indicated time points and IL-8 concentration was determined by ELISA . When compared to normal human fibroblasts ( Fig . 6 ) , the amount of secreted IL-8 was significantly reduced in the HCMV infected ATM -/- cells ( 16 , 79 vs 3 , 63 fold induction ) ( Fig . 7 ) . Similar results were obtained with a transformed ATM -/- fibroblast cell line ( data not shown ) . Infection with HCMV was confirmed by the presence of the viral protein UL44 ( Fig . 7 , below ) . Similar to wild type HCMV , ATM -/- cells infected with the UL76 deficient virus ( TNUL76 ) secreted lower levels of IL-8 compared to HFF infected cells . The levels of IL-8 of cells infected with TNUL76 were , however , even lower than the observed in ATM -/- cells infected with wild type virus ( Fig . S2 ) . Collectively , these results indicate that during HCMV infection , UL76 , and possibly other gene ( s ) , induces IL-8 expression , at least in part , through activation of ATM .
The UL24 gene family is one of the approximately 40 core genes that are conserved in all three herpesviruses subfamilies and the only one which still has no assigned function [13] . Previously , functional assays demonstrated that all homologues of UL24 gene family induce cell cycle arrest [18] , [19] . This work identifies another , and at first sight , apparently unrelated function of the UL24 homologue from HCMV ( UL76 ) , the induction of the expression of IL-8 . Further exploration of the mechanism , however , suggests that both activities may result from viral activation of the DNA Damage response . Deletion mutant analysis of the IL-8 promoter demonstrated that UL76 up-regulates IL-8 expression through activation of NF-kB pathway requiring a functional IKKβ and degradation of the IkB protein . Moreover , expression of UL76 resulted in the translocation of the NF-kB p65 subunit to the nucleus and its binding to the IL-8 promoter . These events , characteristic of the canonical NF-kB pathway , typically occur in the cytoplasm and are usually activated by membrane-receptor stimulation [9] . Thus , the exact mechanism of how UL76 activates the NF-kB pathway is an interesting paradox as UL76 is a nuclear protein . In recent years an alternative mechanism of activation of NF-kB pathway triggered by genotoxic stress has been described . In contrast to inflammatory stimuli such as TNFα or IL-1β , the signal originates in the nucleus [10] . Since it has been shown that UL76 is able to induce double strand breaks , and consequently activate DNA damage [20] , we hypothesized that UL76 might induce IL-8 expression as result of the DNA damage response . Indeed , and similar to the effect of genotoxic drugs such as etoposide , expression of UL76 resulted in an accumulation of nuclear NEMO and its activation ( phosphorylation at serine 85 ) . These findings suggest that the ATM kinase might play a role in the IL-8 induction by UL76 and thus , the predicted role of ATM was demonstrated by two strategies , one using a specific ATM inhibitor and the other employing an ATM knockout cell line . Abrogation of UL76-mediated induction of IL-8 occurred with both approaches . These observations indicate that induction of IL-8 by UL76 originates in the nucleus as a result of the DNA Damage response . The exact mechanism of activation of the DNA Damage response induced by UL76 is still not clear . A promising clue that we pursued was the identification of conserved putative PD- ( D/E ) XK endonuclease motifs in the UL24 gene family [14] . Its conservation suggests a critical role in the function of this gene family , thus , the DNA damage activation by UL76 could be a direct effect of its endonuclease activity . Mutation of the three predicted endonuclease signature amino acids was , however , inconclusive as it resulted in a reduction rather than an abolition of IL-8 induction . It is possible that the observed reduction may be related with an unknown function of these conserved domains rather than loss of a putative endonuclease activity . In fact , the bioinformatically predicted endonuclease motif might not be related to a functional endonuclease activity , as no endonuclease activity has been demonstrated experimentally for any of the UL24 homologues . Importantly , UL76 has a critical role in the up-regulation of IL-8 during HCMV infection as demonstrated by the significant reduction of secreted IL-8 in cells infected with an UL76 deletion mutant virus . This result is particularly important as IL-8 enhances HCMV replication and contributes to the efficient viral dissemination by neutrophils [2] , [3] . The only HCMV gene that has been described as an activator of IL-8 is the Immediate Early 1 gene ( IE1 ) [22] . Thus , the incomplete inhibition of IL-8 secretion by HCMV observed in the absence of UL76 may be due to the effect of IE1 gene . The existence of other gene ( s ) that may also contribute for HCMV-induced IL-8 expression , in addition to IE1 and UL76 , is not excluded . Experiments to observe a similar impact of UL76 in the signaling pathway at the level of virus infected cells were negative , possibly due to alternative virus strategies activating these proteins . We emphasize , however , that the key observation is the diminished expression of IL-8 induced by the UL76 deficient virus , which clearly demonstrates a role for UL76 in the up-regulation of IL-8 in HCMV infected cells . The relevance of IL-8 in HCMV life cycle is emphasized by the fact that the HCMV UL146 gene encodes a homologue to CXC chemokines such as IL-8 ( vCXCL1 ) , which functions as a selective agonist for CXCR2 and , with lower affinity , for CXCR1 [25] , [26] . The IL-8 production observed in cells infected with wild type HCMV or UL76 mutant virus , however , is independent of the presence of the viral CXCL1 , as this gene is deleted from the HCMV AD169 strain used in this work [16] . Induction of IL-8 expression by HCMV requires activation of the NF-kB pathway [22] . Thus , one objective of this work was to elucidate the mechanism of NF-kB activation by HCMV that leads to IL-8 production . Here we demonstrate that ATM also has a critical role in the induction of IL-8 by HCMV as infection of ATM -/- cells with wild type HCMV resulted in considerably lower levels of secreted IL-8 compared to the similar infection of normal human fibroblasts . On the other hand , the incomplete inhibition of IL-8 expression in ATM-/- cells infected with HCMV suggests that other NF-kB pathways are involved . It is possible that these are not redundant effects , but activation of different NF-kB pathways , possibly through different viral proteins , may be necessary for the induction of optimal levels of IL-8 by infected cells that will be beneficial for HCMV replication as previously reported [2] . It may be significant that a major reduction in viral replication is observed when normal cells are infected with a UL76 deficient virus [16] or when ATM deficient cells are infected with HCMV wild type virus [27] . It is possible that this defect in viral replication is associated with the reduced IL-8 levels observed in cells infected with UL76 deficient HCMV or in ATM deficient cell infected with wild type HCMV . Supporting this hypothesis is the fact that IL-8 enhances HCMV replication [2] . In summary , the non-homologous UL76 gene of HCMV has not only evolved for manipulation of the host cell cycle , but also activates expression of the pro-inflammatory chemokine IL-8 . Both of these activities appear to depend on activation of pathways triggered as a result of the DNA Damage response and may favor propagation of the virus . The fact that , in recent years , several viruses have been demonstrated to activate the DNA Damage response raised new important questions . It is not known if this activation results from recognition of DNA damage or if it is due to the recruitment of DNA repair proteins observed during viral infections such as HCMV . Furthermore , it is not completely understood how the activation of DNA Damage pathway is beneficial for viral replication . Our present work establishes a new role of the induction of DNA Damage response in the context of viral infection that may help to elucidate some of these questions , as it demonstrates how viruses exploit the complex crosstalk that occur between different cell signaling pathways .
Human embryonic kidney 293T cells were cultured in 5% CO2 in Dulbecco's Modified Eagle's Medium ( Gibco ) supplemented with 10% fetal calf serum ( Gibco ) at 37°C . Human foreskin fibroblasts ( HFF ) ( obtained from European Collection of Cell Cultures ) , a transformed ( GM09607 ) and a primary ( GM01588 ) A-T human fibroblast cell lines ( obtained from the Coriell Institute for Medical Research ) were cultured in Minimum Essential Medium with Earle's salts supplemented with 10% fetal calf serum ( Gibco ) . The UL76 and UL77 gene from HCMV AD169 were cloned into pcDNA3 plasmid fused in frame with an amino-terminal influenza haemaglutinin peptide ( HA ) tag . The three putative endonuclease amino acids in the UL76 gene were mutated to glycine ( pcDNA3HA-E/K mut plasmid ) according to the Directed Mutagenesis kit protocol ( Stratagene ) . The luciferase reporter constructs containing human IL-8 promoter ( -131 ) or a mutation in the NF-kB , AP-1 or NF-IL-6 binding site were a gift from Dr Naofumi Mukaida and have been described before [22] . The reporter plasmid for NF-κB [p ( PRD2 ) 5tkΔ ( -39 ) lucter] was a gift from Dr Steve Goodbourn . Dominant negative mutants of IKKβ and IkBα ( S32/36A ) plasmids containing an HA tag , were obtained from Dr Michael Karin [28] and Dr Dean Ballard [29] , respectively . The pCMVβ plasmid contains a β-galactosidase gene under the control of human cytomegalovirus immediate early promoter . The HCMV laboratory strain AD169 bacterial artificial chromosome ( BAC ) DNA was obtained from Dr Ulrich Koszinowski . The UL76 mutant virus ( TNUL76 ) , a gift from Dr Thomas Shenk , was generated by site-directed transposon mutagenesis of HCMV AD169 BAC and has been previously described [16] . Wild type or UL76 mutant virus BAC DNA were transfected in HFF cells by electroporation . Supernatants of transfected cells were collected and used for virus stock production . To prepare virus stocks of wild type AD169 BAC virus and TNUL76 mutant virus , HFF cells were infected at a multiplicity of infection ( MOI ) of 0 . 01 . After virus adsorption for one hour , infected cells were cultured at 37°C and medium was collected every three days . Pre-cleared supernatants were centrifuged two hours at 12000 rpm at room temperature . Virus aliquots were stored at −80°C . Virus stock titers were determined by plaque assay . Briefly , HFF cells were cultured with 10-fold dilutions of virus suspension and allowed to absorb for 1 h . Cells were then cultured with complete medium containing 10% carboxymethylcellulose ( CMC ) for 10–15 days . Cellular monolayers were fixed in 4% paraformaldehyde and stained with 0 . 1% toluidine blue . Quantification of the viral plaques was performed using a dissecting microscope . 293T cells were co-transfected in triplicate with 100 ng of IL-8 luciferase reporter plasmid or luciferase reporter constructs containing mutations in the IL-8 promoter ( ΔNF-kB , ΔAP-1 and ΔNF-IL-6 ) , 25 ng of β-galactosidase internal control plasmid ( pCMVβ ) and 300 ng of pcDNA3 or pcDNA3HA-UL76 , according to the Lipofectamine 2000 ( Invitrogen ) protocol . A similar transfection protocol was performed using the NF-kB luciferase reporter plasmid . Cells were lysed 28 h–30 h post-transfection and the luciferase activity was measured using the luciferase assay system ( Promega ) according to the manufacturer's protocol . β-galactosidase activity was measured using the Galacton-Plus kit from Tropix ( Bedford , MA ) . The luciferase activity was normalized relative to the β-galactosidase activity of each sample as control of transfection efficiency . Supernatants of 293T cells or ATM -/- ( GM09607 ) transfected with pcDNA3 ( negative control ) , pcDNA3HA-UL76 or pcDNA3HA-UL77 plasmids were collected at 48 h post-transfection . As control , cells were stimulated with etoposide ( 10 µM ) ( Sigma ) , TNFα ( 10 ng/ml ) ( Peprotech ) or IL-1β ( 1 ng/ml ) ( Cell Signalling ) for 5 h . The concentration of IL-8 secreted was determined using an IL-8 ELISA kit ( BD Biosciences ) following the manufacturer's instructions . Similarly , supernatants of HFF or ATM -/- ( GM01588 ) cells infected with wild type HCMV or UL76 mutant HCMV ( TNUL76 ) at a MOI of 3 , or mock-infected , were harvested at the indicated time points and clarified by centrifugation before quantification of IL-8 by ELISA . For ATM inhibition experiments , ATM inhibitor KU55933 ( 10 µM ) ( Calbiochem ) was added to cells 1 h before transfection or infection with HCMV and was maintained in the medium during the experiment . Plates were analyzed at 450 nm using a BioRad ELISA Reader ( BioRad ) and levels of IL-8 were determined by comparison to a standard curve . The 293T or HFF cells were cultured on sterile glass coverslips and transfected with pcDNA3HA-UL76 or control pcDNA3 plasmid according to the Lipofectamine 2000 ( Invitrogen ) protocol . As positive control cells were stimulated with recombinant human TNFα ( 20 ng/ml ) ( Peprotech ) for 30 minutes . At the indicated times post-transfection , cells were washed with PBS and fixed with 4% paraformaldehyde for 20 minutes . Fixed cells were permeabilised with PBS-0 . 1% Triton X-100 for 20 minutes . After washing , the cells were blocked with PBS-0 . 05% Tween 20 containing 5% normal goat serum for one hour . The samples were incubated with a mouse monoclonal anti-p65 ( F-6 ) or anti-IKKγ/NEMO ( B-3 ) antibody ( Santa Cruz Biotechnology ) followed by incubation with goat anti-mouse Texas Red secondary antibody ( Molecular Probes ) and rat monoclonal anti-HA-FITC conjugated antibody ( Roche ) to visualize UL76 HA-tagged protein . After incubation with DAPI , the coverslips were mounted in “Slow Fade” ( Invitrogen ) and images were acquired with a DeltaVision microscope ( Applied Precision/Olympus ) . 293T cells were transfected with pcDNA3HA-UL76 or control pcDNA3 plasmid and nuclear extraction was performed using a Nuclear Extraction Kit according to the manufacturer's indications ( Active Motif ) . Briefly , at the indicated time points post-transfection , cells were collected in ice-cold PBS in the presence of phosphatase inhibitors . Cytoplasmic extracts were obtained by resuspending the cells in hypotonic buffer followed by addition of detergent . After centrifugation the pelleted nuclei were lysed and nuclear proteins were solubilized in the lysis buffer supplemented with a protease inhibitor cocktail . Protein concentrations were determined by Bradford assay ( Bio-Rad Laboratories ) . Total lysates from cells transfected with pcDNA3HA-UL76 or control pcDNA3 plasmid were prepared using lysis buffer supplemented with a mixture of protease and phosphatase inhibitors ( Calbiochem ) , for 30 minutes on ice . Protein concentrations were determined by Bradford assay ( Bio-Rad Laboratories ) . Proteins from total or nuclear lysates were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis ( SDS-PAGE ) and transferred to polyvinylidene difluoride ( PVDF ) membrane ( GE Healthcare ) . Membranes were blocked with 5% nonfat milk for one hour at room temperature . Primary antibodies used were: mouse monoclonal anti-p65 ( F-6 ) , mouse monoclonal anti-IKKγ/NEMO ( B-3 ) , rabbit anti-nucleolin/C23 ( H-250 ) , mouse monoclonal anti-p53 , rabbit anti-ATM , mouse anti-IE1 HCMV , mouse anti-pp52 ( UL44 ) ( Santa Cruz Biotechnology ) , rabbit anti-IKKγ/NEMO ( S85 ) ( Assay Biotech ) , rabbit anti-phospho-Histone H2AX ( Ser139 ) , mouse monoclonal anti-phospho-p53 ( Ser15 ) , mouse monoclonal anti-phospho-ATM ( Ser1981 ) ( Cell Signaling ) , mouse monoclonal anti-β-actin , anti-HA and anti-tubulin ( Sigma ) . Mouse monoclonal anti-β-actin and rat monoclonal anti-HA horseradish peroxidase-conjugated antibodies were purchased from Sigma . IRDye 800CW anti- mouse and anti-rabbit antibodies were purchased from Li-Cor Biosciences . Immunoblots were developed by enhanced chemiluminescence detection according to the manufacturer's instructions ( ECL , Thermo Scientific Pierce ) or using the Odyssey Infrared Imaging System ( Li-Cor; Lincoln , NE ) . Densiometry analysis was performed using ImageJ software or Image Studio Lite Analysis Software ( Li-Cor ) . 293T cells were transfected with pcDNA3HA-UL76 or control plasmid according to the Lipofectamine 2000 ( Invitrogen ) protocol . Thirty hours post-transfection , cells were cross-linked with 1% formaldehyde ( Calbiochem ) for 10 minutes at room temperature . After washing with PBS , cells were resuspended in SDS lysis buffer with protease inhibitor cocktail ( Sigma ) and chromatin was sheared by sonication . Immunoprecipitation was performed overnight at 4°C , using 2 µg of rabbit polyclonal anti-NF-kB p65 ( A ) or control IgG antibody ( Santa Cruz Biotechnology ) . After incubation with protein G magnetic beads ( Dynabeads , Invitrogen ) for one hour at 4°C , immunocomplexes were washed and eluted . The cross-linking was reversed by heating at 65°C for 4 h . Chromatin-associated proteins were digested with proteinase K and DNA was purified by QIAGEN PCR purification kit following manufacturer's protocol . Immunoprecipitated DNA was quantified by real-time quantitative PCR using SYBR Green Master Mix ( Applied Biosystems ) and primer pair spanning the human IL-8 promoter region from −121 to +61: sense 5′-GGGCCATCAGTTGCAAATC -3′ and antisense 5′-TTCCTTCCGGTGGTTTCTTC-3′ . Primers targeting the genomic region from −1042 to −826 of the IL-8 gene were used as negative control region: sense 5′-AACAGTGGCTGAACCAGAG-3′ and antisense 5′-AGGAGGGCTTCAATAGAGG-3′ . Data were shown as mean values with standard deviation ( SD ) . Differences between experimental groups were determined by a two-tailed Student t test using GraphPad Prism 5 software .
|
The importance of herpesviruses is evident by their prevalence in the human population and the diverse range of diseases that they provoke . Their ability to establish latency provides a compelling example of how herpesviruses successfully evade the immune system and manipulate cellular biology . One promising approach for the development of novel anti-viral strategies is to study viral proteins involved in these host-pathogen interactions . One example is the induction of the pro-inflammatory chemokine IL-8 by HCMV which enhances viral replication and dissemination of the virus by neutrophils . Here , we have identified HCMV UL76 gene , conserved in all herpesviruses , as an inducer of IL-8 , and thus with an important impact on HCMV pathogenesis . The induction of IL-8 by UL76 results from activation of the DNA Damage response , which may also explain why UL76 also induces cell cycle arrest . These findings are a clear example of how viruses manipulate intracellular signaling pathways with different outcomes that will be beneficial for viral infection . Finally , the fact that UL76 is a non-homologous gene substantiates the premise that many such pathogen genes without homology may indeed have evolved for host manipulation , and are a repository of potential useful tools for experimental manipulation in health and disease .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"signal",
"transduction",
"virology",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"molecular",
"cell",
"biology"
] |
2013
|
Human Cytomegalovirus Gene UL76 Induces IL-8 Expression through Activation of the DNA Damage Response
|
Many large-scale studies on intrinsically disordered proteins are implicitly based on the structural models deposited in the Protein Data Bank . Yet , the static nature of deposited models supplies little insight into variation of protein structure and function under diverse cellular and environmental conditions . While the computational predictability of disordered regions provides practical evidence that disorder is an intrinsic property of proteins , the robustness of disordered regions to changes in sequence or environmental conditions has not been systematically studied . We analyzed intrinsically disordered regions in the same or similar proteins crystallized independently and studied their sensitivity to changes in protein sequence and parameters of crystallographic experiments . The observed changes in the existence , position , and length of disordered regions indicate that their appearance in X-ray structures dramatically depends on changes in amino acid sequence and peculiarities of the crystallographic experiment . Our study also raises general questions regarding protein evolution and the regulation of protein structure , dynamics , and function via variations in cellular and environmental conditions .
In the past decade , significant progress has been achieved in our understanding of the ubiquity and function of intrinsically disordered proteins [1]–[8] . What once seemed to be a set of exceptions to the traditional structure-to-function paradigm , where every protein was believed to have unique and stable 3D structure to carry out specific function , turned into a field where computational and experimental approaches were developed and combined to accurately characterize disordered proteins [9] , understand their function [4] , [7] , [8] or mechanisms of binding [10]–[13] , and estimate their abundance in the protein universe [14]–[16] . Undoubtedly , bioinformatics analyses and methods played a significant role in this process , especially a set of predictors and statistical techniques [8] , [17] . However , despite previous success , questions can be raised about the generality of our view of disordered proteins in terms of sequence-to-structure determinants and influence of environmental conditions . Here , we attempt to address these questions by investigating the variability of observed disordered regions with changes in sequence and environmental conditions used for crystallization . Recent studies document the effects of varying environmental conditions on regions of intrinsic disorder in similar proteins . Zurdo et al . studied two yeast ribosomal stalk proteins , P1α and P2β , which have different functional roles despite high sequence similarity and suggested that their functional differences stem from different structures [18] . Although neither protein is compact in solution and possesses folded structure under physiological pH and temperature , P1α was found to be mostly disordered with low helical content , whereas P2β had significant residual structure . This residual structure disappeared at temperatures below 30°C , but was regained under low pH or in the presence of trifluoroethanol . Palaninathan et al . reported that conformational changes were observed in the tertiary and quaternary structures in the crystals of the native human transthyretin ( TTR ) [19] . At pH = 4 . 0 , TTR forms a tetramer and its crystal structure includes electron density for a functionally important EF helix-loop region . At pH = 3 . 5 , this region is completely disordered . Our search of the Protein Data Bank ( PDB ) resulted in additional examples where slight changes in experimental conditions strongly correlated with the presence or absence of disordered regions . One such case is cyclophilin 40 ( Cyp40 ) , shown in Figure 1 ( complete list of analyzed proteins can be found in Table S1 , Suppl . Data ) . Cyp40 is one of the principal members of a family of large immunophilins found in mammals . The exact biological function of large immunophilins is incompletely understood , though they are believed to be strongly associated with Hsp90 and play a crucial regulatory role in the upkeep of steroid receptor activity . In PDB , Cyp40 is stored as 1IIP-A ( tetragonal form ) and 1IHG-A ( monoclinic form ) . Both structures were obtained using the vapor diffusion , hanging drop method with recorded temperature of 277K , but 1IIP-A was crystallized at a pH of 8 . 0 , whereas 1IHG-A was crystallized at pH of 6 . 1 . The two proteins are identical , yet a rmsd of 14 . 2 Å was obtained from their structural alignment . Importantly , 1IHG-A contains an ordered region A299-Y365 that was absent from the structure of 1IIP-A ( Figure 1 ) . Neither protein was solved in the presence of natural ligands . In addition to experimental studies , computational analyses of redundant sets of experimentally determined structures for identical protein regions have provided evidence of the existence of numerous protein fragments observed in both ordered and disordered states [20] . The authors analyzed these ‘dual-personality’ fragments and showed that they are characterized by amino acid compositions different than those for either ordered or disordered proteins and that their main functional roles are regulatory . The examples discussed above demonstrate the strong influence experimental parameters can have on disordered residues in crystallized proteins . However , a hypothesis that variation in experimental conditions could potentially trigger structural changes affecting the existence , position or length of intrinsically disordered regions has not been systematically tested and quantified . In the following work , we provide evidence of significant variation of disordered regions , and protein structures in general , under the same or different experimental conditions that we believe can serve as a basic indicator of environmental regulation of protein structure and disordered regions in vivo .
Figure 2 shows the mean agreement of disordered residues obtained in pairs of identical proteins and proteins with sequence identity in the range [90 , 100 ) % . When all experimental conditions were similar , the agreement of disordered residues for identical sequences was 92% ( 95% for monomers only ) . For the same set of experimental conditions , however , and sequence identity in the range [90 , 100 ) % , the agreement of disordered regions decreased to 52% for the set of all protein chains ( P = 1 . 4⋅10−48; Wilcoxon test ) and 50% for monomers ( P = 5 . 5⋅10−10; Wilcoxon test ) . We also investigated the situation when at least one experimental condition was different ( e . g . temperature , salt concentration , and/or pH value ) . For both identical proteins and those in the [90 , 100 ) % range , the reduction of the mean agreement of residues designated as disordered was about 11 percentage points ( see Fig . 2 caption for P-values ) . In an attempt to estimate which of the experimental conditions had the largest influence on the variability of observed disordered regions , a count for each condition was incremented for each protein pair with inexact matches of disordered regions whenever this condition differed . We found that salt concentration had slightly larger impact ( 39% ) than temperature ( 31% ) and pH value ( 30% ) , as shown in Figure 2 ( inset ) . Furthermore , we found that , in general , an increase in temperature ( 6% ) and pH value ( 7% ) lead to an increase in the number of disordered residues in identical or similar protein chains . In contrast , an increase in salt concentration ( 11% ) leads to a decrease in the number of observed disordered residues . We also grouped all pairs of sequences with identity≥90% into those solved using at least one , two , or three similar experimental conditions and at least one , two , or three different experimental conditions . We estimate that , assuming unchanged experimental platforms for structure determination , the mean agreement of intrinsically disordered residues is 73% ( 79% , 83% ) if one ( two , three ) or more experimental conditions are similar ( Figure 3 , left ) . When different experimental conditions were considered , the agreement of disordered residues was consistently around 50% . In Table 1 we present complete results of the consistency measurements for both ordered and disordered regions for the pairs of chains with sequence identity≥90% . Ordered regions from such pairs of proteins appeared as highly overlapping ( >98% ) , which is due to the unbalanced number of ordered and disordered residues in the non-redundant data set ( 14∶1 ratio ) . Finally , we estimated the mean agreement of disordered residues using pairs of similar and identical protein sequences wherein experimental information at the time of pair generation was not considered . If identical protein pairs are considered , the mean overlap of disordered and ordered residues was 89% and 99% , respectively . When we considered disordered regions of length 30 or more , the mean overlap was 93% and 98% , respectively ( Figure 4 ) . Interestingly , all pairs from our analysis in which long disordered regions significantly differed belonged to dissimilar experimental classes thus strongly suggesting that the appearance of disordered regions is influenced by variations in experimental conditions ( e . g . 1COT-B and 1S6P-B ) . Consideration of similar sequences resulted in a significant reduction in the mean overlap: 31% for all disordered regions and 35% for long disordered regions only . Note that the slightly smaller overlap of disordered residues , compared to the one from Figure 2 , is due to the influence of completely ordered proteins for which we were unable to extract experimental conditions and therefore were excluded from the analysis in Figure 2 . The observed consistency of disordered residues may not necessarily be the same at the level of entire disordered regions . Figure 3 ( right ) shows the percentage of disordered regions that were found as ordered in their entirety when the same or similar proteins were crystallized in independent experiments . When all crystallographic parameters were similar , 13% of regions were found as completely ordered . On the other hand , when all parameters were different we estimated that close to 50% of the regions were lost ( P = 1 . 7⋅10−10; Wilcoxon test ) . To understand whether a loss of disordered regions could be due to potential ligand binding , we investigated pairs of proteins ( p1 , p2 ) , where p1 contained a disordered region r for which p2 contained all ordered residues in the segment aligned with r . We considered that a ligand influenced disorder-to-order transition if any of its atoms could be found within 10 Å of any of the ordered residues from p2 corresponding to r as well as requiring that the ligand was not present in the model of protein p1 . We found that about 25% of disordered regions that underwent order-disorder transition were due to direct ligand binding . Thus , ligands in PDB considerably influence the existence of disordered regions . However , their influence appears to be a less significant factor than experimental conditions or sequence variation . The results presented in Table 1 and Figures 2–4 provide estimates regarding the limits of predictability of intrinsically disordered residues . By combining the mean agreement of both ordered and disordered residues in identical protein chains when all experimental conditions agree , we estimate that the prediction accuracy of computational models constructed to predict disordered regions , measured by averaging sensitivity and specificity , is approximately 95% . This accuracy reduces to 90% if the experimental conditions are not taken into consideration , which is closer to the situation used in computational studies . However , since we considered only identical pairs of proteins , both of these limits seem overly optimistic . Thus , we believe that a more realistic estimate is provided when all sequence pairs with identity≥90% are considered and experimental conditions are ignored . The observed agreement of disordered and ordered residues in such a case was 66% and 96% , respectively . Thus , the maximum balanced-sample accuracy is probably about 81% . Interestingly , the best models in CASP7 assessment have reached 74–78% balanced-sample accuracy [21] , so it is unclear whether the current general predictors can be significantly improved . The knowledge of experimental conditions , on the other hand , should be able to improve the predictability of disordered residues by at least 5 percentage points ( Figure 2 ) . In addition , structures of solved homologs and mutants could provide an additional increase if the points of low stability can be identified .
This study addresses the relationship between intrinsically disordered protein regions , protein sequence , and parameters of crystallographic structure determination . The existence , position , and length of disordered regions in highly similar proteins was shown to strongly depend on variation in amino acid sequence as well as the parameters of crystallographic experiments , such as temperature , pH , and salt concentration . For identical protein chains , most of the observed rearrangements in the crystal lattice can be explained by variation in experimental conditions . For highly similar chains , both experimental conditions and the intrinsic change of protein structure were significant factors . However , we are hesitant to assign relative importance to these factors since the observed sequence differences in PDB are likely to be non-random ( for example , mutations with functional or phenotypic significance are frequently of interest for structure determination ) . The presence/absence of ligands appeared to be less significant in our analysis . The presence of a disordered region under one set of experimental conditions and absence under another can be understood through the framework of the probabilistic theory of protein folding . At every time instant , a protein can be assigned a probability of any particular conformation based on its energy landscape [22] , [23] . For ordered proteins , such energy landscapes are characterized by single ( or a small number of ) deep minima with high probabilities associated with the corresponding conformations . Since the number of conformations in the high energy states is huge and the barriers for moving away from the dominant conformation are relatively large , the energy landscape has a shape of a funnel [23] . This minimum energy state is often associated with protein function and is called the native state . On the other hand , the energy landscapes for disordered proteins are shallower , typically characterized by flat and rugged valleys , i . e . they contain a large number of energy minima with relatively small barriers for transitioning between distinct conformations [24] . Consequently , the probability of each conformation corresponding to an energy minimum is relatively low . The absence of a high probability conformation eventually leads to missing electron density during crystallographic experiments . Thus , the variability in structures of identical proteins solved under different experimental conditions is caused by the environment-driven changes of the energy landscape ( Figure 5 ) . The altered probability distribution over the space of allowed tertiary structures ultimately results in a population shift between ensembles of pre-existing conformational isomers [23]–[25] . The folding funnel theory can not only accommodate both the thermodynamic and the kinetic requirements for protein folding [22] , but also provide a general framework under which folding , binding ( including allostery ) , or effects of mutations and post-translational modifications can be considered [23] , [24] , [26] . For example , folding and binding essentially represent the same phenomenon with a distinction that the chains are disconnected in the case of binding [23] , [25] , [27] . In allostery , a lower probability conformation may be the one preferred for binding . If this complex is the preferred state , the increased probability of a bound conformation will cause a population shift over time from one dominant conformation to the one preferred for binding [26] , [28] . Recently , population shifts were demonstrated for ubiquitin , where all bound conformations available from crystallographic experiments were shown to be accessible in solution by NMR [29] . A limitation of our analysis is that it only included disordered proteins with at least two deposited structures in PDB , and thus may be a non-representative sample . In addition , this data set is enriched for short disordered regions that have distinct sequence biases relative to long regions [30] , [31] . A full analysis including long disordered regions was not possible due to the small number of available protein pairs; however , the overall trends indicate that long disordered regions may be equally sensitive to variation in sequence and experimental conditions . In general , this work provides evidence that disordered protein regions are very sensitive to changes in amino acid sequence and experimental conditions of crystallographic experiments . The success of such crystallographic experiments depends on the complexity of protein's structure and also on a number of experimental or environmental factors including purity of the protein sample , temperature , ionic strength , pH , and precipitants such as ammonium sulfate or polyethylene glycol [32] . Undoubtedly , there are a number of factors that distinguish crystallization conditions from physiological conditions , but there is also a body of evidence that protein structures often correspond to their native states [32] . Therefore , it is reasonable to speculate that a wide range of intracellular and extracellular conditions may have similar effects on the dynamics of protein 3D structure in vivo . The habitats for many living organisms vary from acidic to cold or hot , with various species being able to tolerate wide ranges of environmental conditions . As suggested and quantified by our analysis , any similar variations in cellular environments could have profound effects on protein structure , dynamics , and function . Sensitivity to sequence changes , on the other hand , may facilitate the evolution of function , especially for proteins with the same fold classification .
Our initial data set S comprised of 18 , 884 protein chains from PDB ( March 2008 ) characterized by X-ray crystallography with resolution of at most 2 Å ( Table S2 , Suppl . Data ) . It contained two subsets: D–a set of 14 , 646 chains containing at least one disordered region of length≥3 , identified as those missing C-α atoms in the ATOM fields; and OD–a set of 4 , 238 completely ordered chains such that each sequence was ≥90% identical to one or more sequences in D . For each sequence in S we extracted experimental conditions: temperature , pH value , and concentration of salt ( e . g . ammonium sulfate , potassium sodium tartrate , sodium cacodylate , and a number of others ) , whenever available ( 1 sequence in D and 1502 sequences in OD , did not have any experimental conditions extracted due to differences in file format ) . While temperature and pH value can be obtained from designated fields in PDB , the salt concentration was mined from REMARK200 and REMARK280 fields and manually checked in a number of cases . For simplicity of our analysis , each experimental condition was clustered into two groups , high and low , as discussed in the Results section ( Figure S1 , Suppl . Data ) . Temperature was clustered into group high ( Th ) , containing temperatures greater than or equal to 200 K and group low ( Tl ) , containing temperatures below 200 K at the time of experiment . pH value was clustered into Ph and Pl based on threshold 6 . 5 , while the salt concentration was clustered into Sh and Sl based on the threshold of 100 mM . To construct the non-redundant data sets , the initial set D was split into overlapping subsets , where each subset set Di contained proteins crystallized at experimental conditions Ei ∈ {Th , Tl , ThPh , ThPl , … , TlPlSl} . More specifically , data set containing proteins crystallized at conditions ThPh , had proteins solved at high temperature and high pH value , but the salt concentration could be from the entire range or unknown . Each data set Di was also filtered into a non-redundant set Di−nr such that no two chains had sequence identity greater than or equal to 25% on a global level ( BLOSUM62 matrix , gap opening penalty = −11 , and gap extension penalty = −1 ) . This approach of defining non-redundant sets was used for estimating the overlap of disordered regions between classes Ei and Ej . The size of each data set is shown in Table 2 . Consistency of disordered residues and regions was estimated by calculating the mean overlap of ordered and disordered regions in similar or identical protein chains , crystallized at the same or different experimental conditions . Two protein chains were considered to be similar if their global sequence identity was ≥90% . This threshold was selected to ensure not only similar 3-D structure between two proteins [33] , but also similar function [34] . The mean overlap between two globally aligned proteins p ∈ Di–nr and q ∈ Sj , where the sequence identity ( si ) between p and q was greater than or equal to threshold t1 and lower than t2 , was calculated as follows . Let Op and Dp be the sets of positions of ordered and disordered residues in protein p , and Oq and Dq sets of positions of ordered and disordered residues in protein q , respectively , as shown in Figure 6 . The residue positions are calculated after the alignments are completed . The indices corresponding to insertions and deletions , as well as the indices corresponding to disordered regions of length below 3 , were ignored . We calculate the overlap between ordered ( oo ) and disordered regions ( od ) as Note that q can be a completely ordered sequence , while p is guaranteed to contain at least one disordered region . The average overlap of ordered and disordered regions for a pair ( p , q ) is calculated as We use the term accuracy for the mean overlap due to its similarity to a prediction process in which ordered and disordered regions in one protein serve as predictions for the other protein . The overlaps between pairs of proteins are then generalized to the level of data sets . An average accuracy for chain p is first calculated over all sequences q that are within the sequence identity range [t1 , t2 ) from p , denoted by si ( p , q ) ∈ [t1 , t2 ) . Then , the average accuracy between data sets Di–nr and Sj , corresponding to experimental conditions Ei and Ej , is calculated as the mean over all proteins p . We formalize the entire calculation aswhere and is the number of sequences q ∈ Sj that when aligned to p have sequence identity in range [t1 , t2 ) . Assuming that the maximum prediction accuracy of intrinsically disordered regions is limited by an empirically observed agreement in similar proteins , this approach provides an estimate of the upper limit of the balanced sample accuracy over the given two sets of experimental conditions . The results for several groups of experimental conditions were obtained by simple group averages . The number of pairs for each group of experimental conditions is listed in Table S3 ( Suppl . Data ) . To quantify the agreement of disordered regions for two sets of experimental conditions Ei and Ej , we used a conceptually similar approach . For each protein p ∈ Di–nr we calculated the fraction of regions for which the overlap with sequence q ∈ Sj was zero . The fraction of such regions in p was then averaged over all proteins from q ∈ Sj where si ( p , q ) ∈ [t1 , t2 ) . Finally , the fraction of regions that undergo order-disorder transition between two sets of experimental conditions Ei and Ej was further averaged over all proteins p ∈ Di–nr . Statistical confidence for the estimates was calculated by bootstrapping the non-redundant data sets Di–nr 10 , 000 times .
|
Intrinsically disordered proteins , proteins that exist as conformational ensembles without time-invariant residue positions , have emerged as an important and common class of proteins in all kingdoms of life . Disordered proteins are characterized by distinct amino acid preferences , distinct mechanisms of binding , distinct substitution patterns and rates of evolution , and functional roles predominantly related to signaling and regulation . In recent years , disordered proteins have also been linked to human disease , both through conformational diseases or via host-pathogen interactions . However , despite increased importance , most studies of disordered proteins do not consider the environmental context in which the protein is found or the level of sequence change that would strongly influence the property of being disordered . To address this , we studied and quantified the variability of intrinsically disordered protein regions under different external conditions , such as temperature or pH , and compared them to the variability introduced by small sequence changes . We found that both have a strong impact on the existence of disordered regions , thus potentially regulating protein function by environmental factors or facilitating evolutionary change .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"computational",
"biology/protein",
"structure",
"prediction",
"computational",
"biology"
] |
2009
|
Influence of Sequence Changes and Environment on Intrinsically Disordered Proteins
|
Disrupted circadian rhythms and reduced sleep duration are associated with several human diseases , particularly obesity and type 2 diabetes , but until recently , little was known about the genetic factors influencing these heritable traits . We performed genome-wide association studies of self-reported chronotype ( morning/evening person ) and self-reported sleep duration in 128 , 266 white British individuals from the UK Biobank study . Sixteen variants were associated with chronotype ( P<5x10-8 ) , including variants near the known circadian rhythm genes RGS16 ( 1 . 21 odds of morningness , 95% CI [1 . 15 , 1 . 27] , P = 3x10-12 ) and PER2 ( 1 . 09 odds of morningness , 95% CI [1 . 06 , 1 . 12] , P = 4x10-10 ) . The PER2 signal has previously been associated with iris function . We sought replication using self-reported data from 89 , 283 23andMe participants; thirteen of the chronotype signals remained associated at P<5x10-8 on meta-analysis and eleven of these reached P<0 . 05 in the same direction in the 23andMe study . We also replicated 9 additional variants identified when the 23andMe study was used as a discovery GWAS of chronotype ( all P<0 . 05 and meta-analysis P<5x10-8 ) . For sleep duration , we replicated one known signal in PAX8 ( 2 . 6 minutes per allele , 95% CI [1 . 9 , 3 . 2] , P = 5 . 7x10-16 ) and identified and replicated two novel associations at VRK2 ( 2 . 0 minutes per allele , 95% CI [1 . 3 , 2 . 7] , P = 1 . 2x10-9; and 1 . 6 minutes per allele , 95% CI [1 . 1 , 2 . 2] , P = 7 . 6x10-9 ) . Although we found genetic correlation between chronotype and BMI ( rG = 0 . 056 , P = 0 . 05 ) ; undersleeping and BMI ( rG = 0 . 147 , P = 1x10-5 ) and oversleeping and BMI ( rG = 0 . 097 , P = 0 . 04 ) , Mendelian Randomisation analyses , with limited power , provided no consistent evidence of causal associations between BMI or type 2 diabetes and chronotype or sleep duration . Our study brings the total number of loci associated with chronotype to 22 and with sleep duration to three , and provides new insights into the biology of sleep and circadian rhythms in humans .
There are strong epidemiological associations between disrupted circadian rhythms , sleep duration and disease . A circadian rhythm refers to an underlying 24-hour physiological cycle that occurs in most living organisms . In humans , there are clear daily cyclical patterns in core body temperature , hormonal and most other biological systems [1] . These cycles are important for many molecular and behavioural processes . In particular , circadian rhythms are important in regulating sleeping patterns . While each individual has an endogenous circadian rhythm , the timing of these rhythms varies across individuals . Those with later circadian rhythms tend to sleep best with a late bedtime and late rising time and are often referred to as an “owl” or as an “evening” person . Those with earlier rhythms tend to feel sleepy earlier in the night and wake up early in the morning and are referred to as a “lark” or “morning” person . The remainder of the population falls in between these extremes . This dimension of circadian timing , or chronotype , is one behavioural consequence of these underlying cycles . Chronotype can be simply assessed by questionnaire and is considered a useful tool for studying circadian rhythms [2 , 3] . There is substantial evidence for a relationship between short sleep duration , poor quality sleep and obesity and type 2 diabetes [4 , 5] . Eveningness has been associated with poor glycaemic control in patients with type 2 diabetes independently of sleep disturbance [6] and with metabolic disorders and body composition in middle-aged adults [7] . There is evidence from animal models that disruption to circadian rhythms and sleep patterns can cause various metabolic disorders [8–10] . For example , mice homozygous for dominant negative mutations in the essential circadian gene , Clock , develop obesity and hyperglycaemia [10] and conditional ablation of the Bmal1 and Clock genes in pancreatic islets causes diabetes mellitus due to defective β-cell function [9] . Despite this evidence , in humans the causal nature of the epidemiological associations between sleep patterns , circadian rhythms and obesity and type 2 diabetes is unknown . Identifying genetic variants associated with sleep duration and chronotype will provide instruments to help test the causality of epidemiological associations [11] . A previous genome-wide association study ( GWAS ) in 4 , 251 individuals identified a single genetic variant in ABCC9 associated with sleep duration [12] . A subsequent GWAS meta-analysis including 47 , 180 individuals identified a single locus for sleep duration near PAX8 [13] . Fifteen loci associated with chronotype were recently discovered by 23andMe [14] with 7 of these found to be in close proximity to known circadian rhythm regulation genes . The UK Biobank is a study of 500 , 000 individuals from the UK aged between 37 and 73 years with genome-wide SNP analysis and detailed phenotypic information , including chronotype and sleep duration ( http://www . ukbiobank . ac . uk/ ) . The UK Biobank study provides an excellent opportunity to identify novel genetic variants influencing chronotype and sleep duration which will provide insights into the biology of circadian rhythms and sleep and help test causal relationships between circadian rhythm and metabolic traits including obesity .
Using self-reported “morningness” , we generated a binary and a continuous chronotype score . We performed genome-wide association studies on 16 , 760 , 980 imputed autosomal variants . Fig 1 presents the overall results for these GWAS . Table 1 presents details of all 16 loci associated at P<5x10-8 . Analysing UK Biobank data with that from 23andMe provides evidence that at least 13 of the 16 are associated with chronotype . Thirteen of the chronotype signals remained at P<5x10-8 in a meta-analysis including UK Biobank and 89 , 283 individuals from 23andMe [14] , of which 11 reached P<0 . 05 in the same direction in 23andMe alone , and 15 of the 16 UK Biobank signals were in the same direction ( binomial P = 0 . 0002 ) ( Table 1 ) . We also attempted to validate the associations in 6 , 191 European-Ancestry from the Chronogen consortium and 2 , 532 Korean Ancestry individuals from the Insomnia , Chronotype and sleep EEG ( ICE ) consortium that used “Gold standard” chronotype questionnaire ( Munich Chronotype Questionnaire–MCTQ and Morningness-Eveningness Questionnaire—MEQ ) . Given the sample size of 5% of the discovery UK Biobank study we assessed directional consistency rather than testing for replication P-values <0 . 05 or 0 . 05/16 . In the European-Ancestry individuals 11 of the 16 signals were represented . Nine of these 11 variants had the same direction of effect as the discovery UK Biobank cohort ( binomial test P = 0 . 03 ) and one replicated at Bonferroni significance ( rs12140153 , P = 0 . 003 ) . In the Korean study , 9 signals were represented , four of which had the same direction of effect as the discovery UK Biobank cohort ( binomial test P = 1 . 00 ) . The level of directional consistency in these two smaller studies is consistent with what would be expected in cohorts <5% the size of our discovery cohort . A 23andMe study recently identified 15 loci associated with chronotype [14] . All of the 15 signals were replicated in our study with P<0 . 05 in the same direction and had meta-analysis P<5x10-8 ( S1 Table ) . We performed a conditional analysis of our lead chronotype variants by adjusting for the 15 known signals ( S2 Table ) , in order to identify which of our loci coincided with those of Hu et al . [14] . Seven of our 13 replicated signals remained associated at P<5x10-8 ( see Table 1 ) . The addition of these 7 loci brings the number associated with chronotype to 22 ( full list in S3 Table ) . The variant most strongly associated with chronotype , rs516134 ( OR for morningness = 1 . 21 , 95% CI [1 . 16 , 1 . 26] , binary P = 3 . 7x10-12 , continuous P = 8 . 9x10-13 ) occurs near RGS16 , which is a regulator of G-protein signalling and has a known role in circadian rhythms [16] ( Table 1 and Fig 2 ) . Another signal occurs near PER2 ( lead variant rs75804782 , OR = 1 . 09 , 95% CI [1 . 06 , 1 . 12] , binary P = 7 . 2x10-10 , continuous P = 3 . 2x10-7; Fig 3 ) . PER2 is a well-known regulator of circadian rhythms [17–22] and contains a variant , rs75804782 , recently shown to be associated with iris formation [23] that is in LD ( r2 = 0 . 65 , D’ = 0 . 97 ) with our reported lead SNP . As there is a reported link between season and reported chronotype [24] , we carried out a sensitivity analysis in which we adjusted for month of attendance ( to assessment centre ) ; all associations remained genome-wide significant for the reported variants . We tested for enrichment of specific biological and molecular pathways using MAGENTA ( Meta-Analysis Gene-set Enrichment of variaNT Associations ) [25] but none had a clear link to circadian rhythms ( S4 Table ) . We performed genome-wide association studies on a binary sleep phenotype and a continuous sleep duration score for 16 , 761 , 225 imputed variants . Fig 4 presents the overall results for these GWAS . Three loci reached genome-wide significance . The most strongly associated variant was rs62158211 with an average 2 . 6 minute ( 95% CI [1 . 9 , 3 . 2] , P = 5 . 7x10-16 ) per-allele change in sleep duration and occurs at the previously reported association signal near PAX8 [13] . We identified two , novel , conditionally independent , signals that were located ~900kb apart , one upstream and the other downstream of VRK2 . The downstream variant , rs17190618 , has an average per allele effect of 2 . 0 minutes ( 95% CI [1 . 3 , 2 . 7] , P = 1 . 2x10-9 ) on sleep duration . The upstream variant , rs1380703 ( which is not correlated with rs17190618 , r2 = 0 . 002 ) , has an average per allele effect of 1 . 6 minutes ( 95% CI [1 . 1 , 2 . 2] , P = 7 . 6x10-9 ) on sleep duration . On adjusting for month of assessment , we saw similar associations for both rs62158211 ( P = 3x10-16 ) and rs1380703 ( P = 6x10-9 ) , with no change for rs17190618 . Table 2 shows the three sleep duration loci and their lead variants . Fig 5 shows locus zoom plots of the VRK2 association signals . We did not replicate the association of a previously reported variant in ABCC9 [12] with sleep duration ( rs11046205 , 0 . 1mins , 95% CI [-0 . 6 , 0 . 7] , P = 0 . 83 ) . To replicate the two novel sleep duration hits we used data from 47 , 180 individuals from a published study [13] . The variant rs17190618 replicated with effect size = 2 . 1 minutes ( 95% CI [0 . 8 , 3 . 3] , P = 0 . 001 , meta-analysis P = 5x10-12 ) . The variant rs1380703 replicated with effect size = 1 . 3 minutes ( 95% CI [0 . 3 , 2 . 2] , P = 0 . 01 , meta-analysis P = 3x10-10 ) . Using LD-score regression we estimated the heritability of chronotype and sleep duration within UK Biobank to be 0 . 12 ( 0 . 007 ) , and 0 . 07 ( 0 . 007 ) , respectively . There was no evidence of a genetic correlation between sleep duration and chronotype ( rG = 0 . 0177 , P = 0 . 70 ) . Chronotype was nominally genetically correlated with BMI ( rG = 0 . 056 , P = 0 . 048 ) , but not Type 2 diabetes ( rG = 0 . 004 , P = 0 . 99 ) . As the relationship between sleep duration with BMI and risk of T2D is U-shaped ( see S1 Fig ) , we defined two further binary phenotypes; undersleepers ( <7 vs . 7–8 hours ) and oversleepers ( >8 vs . 7–8 hours ) . There was a strong genetic correlation between undersleeping and BMI ( rG = 0 . 147 , P = 1x10-5 ) , but not T2D ( rG = 0 . 022 , P = 0 . 79 ) . There was also a genetic correlation between oversleeping and both BMI ( rG = 0 . 097 , P = 0 . 039 ) and T2D ( rG = 0 . 336 , P = 0 . 001 ) . We also performed LD-score regression analyses against a range of other diseases and traits where GWAS summary statistics are publicly available ( S5 Table ) . Schizophrenia was genetically correlated ( after adjusting for the number of tests ) with hours slept ( rG = 0 . 26 , P = 5x10-10 ) , oversleeping ( rG = 0 . 35 , P = 6x10-8 ) , undersleeping ( rG = -0 . 14 , P = 2x10-3 ) and chronotype ( rG = -0 . 12 , P = 2x10-4 ) . The genetic correlations we observed provide general estimates that capture pleiotropic variants ( those that affect both traits through different pathways ) and associations that are secondary to a variant affecting a trait that causally influences the second trait . Using a genetic risk score of 69 known BMI variants [26] ( listed in S6 Table ) as an instrumental variable , we next performed Mendelian randomisation analyses in the UK Biobank study to test the potential causal role of BMI in chronotype and sleep . Instrumental variables analyses using variants and their effect sizes identified by previous studies [26] provided no consistent evidence that BMI causally affects self-reported “morningness” ( S7 Table ) . Association statistics of the BMI variants with chronotype are given in S6 Table . We repeated these analyses using a genetic risk score consisting of 55 type 2 diabetes SNPs [27] and did not find any evidence of causality . Performing the reciprocal Mendelian randomization analysis using a genetic risk score of the 13 replicated chronotype variants , with effect sizes obtained from 23andMe , we found no consistent evidence in the UK Biobank data that morningness or eveningness leads to higher BMI ( S7 Table ) . Association of the chronotype-associated variants with BMI are given in S8 Table . Using the same genetic risk score of 69 known BMI variants as an instrument , we saw no consistent evidence that higher BMI increased an individual’s likelihood of being an undersleeper ( IVreg2 P = 0 . 95 , IVW P = 0 . 04 ) or an oversleeper ( IVreg2 P = 0 . 29 , IVW P = 0 . 62 ) in the UK Biobank data ( S7 Table ) . Because there were only three genetic variants of small effect associated with sleep duration , we did not perform any Mendelian Randomisation analyses of sleep on BMI or type 2 diabetes risk .
We performed a genome-wide association study of sleep duration and morningness in 128 , 266 individuals from the UK Biobank study . We discovered and replicated two novel loci associated with sleep duration . Through replication in a study of 89 , 823 individuals from 23andMe we found 13 loci associated with chronotype at P<5x10-8 . Together with a recent study from 23andMe [14] this takes the number of replicated loci for being a morning person to 22 ( 7 not reported in the 23andMe study ) . These loci occur in or near circadian rhythm and photoreception genes and provide new insights into circadian rhythm and sleep biology and their links to disease . The two novel sleep duration association signals occur upstream and downstream of VRK2 ( vaccinia related kinase 2 ) . VRK2 is a serine/threonine kinase important in several signal transduction cascades , and variants near VRK2 are associated with schizophrenia [28] and epilepsy [29] . The two sleep duration variants we identified do not represent the same signals as those associated with schizophrenia at genome wide significance but one is associated with schizophrenia ( based on publically available data from the schizophrenia genetics consortium ( rs1380703 ) at P = 2x10-5 , with the allele associated with more sleep being associated with higher risk of schizophrenia ) . Furthermore , the variants associated with epilepsy and schizophrenia at genome wide significance in previous studies are associated with sleep duration in UK Biobank ( epilepsy lead variant rs2947349 [29] , P = 2x10-5 and schizophrenia lead variant [28] rs11682175 P = 3x10-5 ) but did not reach genome wide significance . We also observed genetic correlation between sleep duration and schizophrenia using LD-score regression ( rG = 0 . 26 , P = 5x10-10 ) . Further work is required to determine whether variation in VRK2 either has independent associations with both sleep and schizophrenia or whether there is some causal link between sleep duration and pattern and schizophrenia and epilepsy . Several of the loci that we identified as associated with chronotype contain genes that have a known role in circadian rhythms . The most strongly associated variant , rs516134 , occurs 20kb downstream of RGS16 ( regulator of G protein signalling 16 ) . RGS16 has recently been shown to have a key role in defining 24 hour rhythms in behaviour [16] . In mice , gene ablation of Rgs16 lengthens the circadian period of behavioural rhythm [16] . By temporally regulating cAMP signalling , Rgs16 has been shown to be a key factor in synchronising intercellular communication between pacemaker neurons in the suprachiasmatic nucleus ( SCN ) , the centre for circadian rhythm control in humans . The association signal with lead SNP rs75804782 occurs ~100kb upstream of PER2 ( Period 2 ) . Per2 is a key regulator of circadian rhythms and is considered one of the most important clock genes , and , under constant darkness , Per2 knockout mice show arrhythmic locomotor activity [17–22] . This locus also contains a variant that has recently been shown to be associated with iris furrow contractions [23] . Our signal is very likely to represent the same association and suggests a link between iris function and chronotype ( rs75804782 has an LD r2 = 0 . 65 and D’ = 0 . 97 with the reported lead SNP , rs3739070 ) . Larsson et al . [23] suggest TRAF3IP1 as the most likely candidate gene at the locus because of its critical role in the cytoskeleton and neurogenesis . Further work is needed to elucidate whether the chronotype association at this locus acts through PER2 or TRAF3IP1 . Four of the 22 chronotype loci had missense variants in LD ( r2>0 . 8 ) with the lead variant ( RGS16 , EXD3 , INADL and HCRTR2; see S9 Table ) . The INADL variant association is particularly interesting as INADL ( InaD-like ) encodes a protein that has been thought to be important in organising and maintaining the “intrinsically photosensitive retinal ganglion cells” , cells that are known to communicate directly with the suprachiasmatic nucleus , the primary circadian pacemaker in mammals [30] . This is compelling evidence that INADL is involved in the human circadian rhythm pathway . Several of the variants associated with chronotype are also associated with BMI and we found genetic correlation between chronotype and sleep duration and BMI . There is substantial evidence for a role of sleep disruption and circadian rhythms in metabolic disease [1] . Data from animal models and epidemiology provide strong evidence that sleep quality or disrupted circadian rhythms can cause metabolic diseases including obesity and type 2 diabetes [4–6 , 8–10] . Our Mendelian Randomisation analyses provided no consistent evidence for a role of higher BMI leading to increased self-reported morningness . These Mendelian Randomisation results are consistent with those from the recent study from 23andMe [14] . There are some important limitations to our study . First , chronotype and sleep duration were self-reported and are subject to reporting bias ( e . g . obese individuals may be more likely to falsely claim to be morning people ) . Second , whilst we did not find any evidence that overall chronotype or sleep duration causally lead to obesity or type 2 diabetes , it is possible that sub-pathways of genes involved in , for example , feeding behaviour may be important in both obesity and chronotype regulation . Third , the identified variants only account for a small amount of the variation in chronotype and sleep duration and we therefore had limited power to detect an effect of these variants on BMI or type 2 diabetes risk . The availability of the full UK Biobank study of 500 , 000 will provide further insight into this relationship . In conclusion , we have identified novel genetic associations for chronotype and sleep duration . The chronotype loci cluster near genes known to be important in determining circadian rhythms and will provide new insights into circadian regulation . Our results provide new insights into circadian rhythm and sleep biology and their links to disease .
This study was conducted using the UK Biobank resource . Details of patient and public involvement in the UK Biobank are available online ( www . ukbiobank . ac . uk/about-biobank-uk/ and https://www . ukbiobank . ac . uk/wp-content/uploads/2011/07/Summary-EGF-consultation . pdf ) . No patients were specifically involved in setting the research question or the outcome measures , nor were they involved in developing plans for recruitment , design , or implementation of this study . No patients were asked to advise on interpretation or writing up of results . There are no specific plans to disseminate the results of the research to study participants , but the UK Biobank disseminates key findings from projects on its website . We used 128 , 266 individuals of British descent from the first UK Biobank genetic data release ( see http://biobank . ctsu . ox . ac . uk ) . British-descent was defined as individuals who both self-identified as white British and were confirmed as ancestrally Caucasian using principal components analyses ( http://biobank . ctsu . ox . ac . uk ) . Of these individuals , 120 , 286 were classified as unrelated , with a further 7 , 980 first- to third-degree relatives of these . As the association tests were carried out in BOLT-LMM [31] , which adjusts for relationships between individuals and corrects for population structure , we included all 128 , 266 related white British individuals in the association analyses . We used imputed variants provided by the UK Biobank . Details of the imputation process are provided at the UK Biobank website ( see http://biobank . ctsu . ox . ac . uk ) . For this study we only included the ~16 . 8M imputed variants with an imputation R2 ≥ 0 . 4 , MAF ≥ 0 . 001 and with a Hardy–Weinberg equilibrium P>1x10-5 . To perform the association tests , we used BOLT-LMM [31] to perform linear mixed models ( LMMs ) in the 128 , 266 individuals . We used BOLT-LMM as it adjusts for population structure and relatedness between individuals whilst performing the association tests with feasible computing resources . As it adjusts for population structure and relatedness , it allowed us to include the additional 7 , 980 related individuals and therefore improved our power to detect associations . To calculate the relationships between individuals , we provided BOLT-LMM a list of 328 , 928 genotyped SNPs ( MAF>5%; HWE P>1x10-6; missingness<0 . 015 ) for the individuals included in the association analysis and used the 1000 Genomes LD-Score table provided with the software . As the continuous phenotypes were derived by adjusting for age , gender and study centre , the LMM only included chip ( BiLEVE vs . UKBiobank arrays ) as a covariate at run-time ( see http://www . ukbiobank . ac . uk/wp-content/uploads/2014/04/UKBiobank_genotyping_QC_documentation-web . pdf ) . The binary phenotypes were unadjusted and so included age , gender and chip at run-time . BOLT-LMM reported no improvement of the non-infinitesimal mixed model test over the standard infinitesimal test and so all association results reported in this paper are for the infinitesimal model [31] . Participants ( N = 89 , 283 ) were from the customer base of 23andMe , Inc . The descriptions of the samples , genotyping and imputation are in [14] . Of the 16 chronotype-associated variants for which we attempted replication , 10 were available from imputation from the 1000 Genomes imputation panel phase 1 pilot . An additional 4 were imputed from the phase 1 version 3 1000 Genomes imputation panel . The final two could not be imputed . We used http://analysistools . nci . nih . gov/LDlink/ to find proxies—the best available were rs4729854 for rs372229746 ( r2 = 0 . 33 ) , and rs12621152 for rs70944707 ( r2 = 0 . 33 ) . We meta-analysed P-values from the discovery and replication samples using sample size weighting implemented in METAL [32] . Genotypes consisting of both directly typed and imputed SNPs were used for the individual GWAS [12] . To avoid over-inflation of test statistics due to population structure or relatedness , we applied genomic control for the independent studies and meta-analysis . Linear regression for associations with normalised chronotype was performed ( see [12] for packages used ) under an additive model , with SNP allele dosage as predictor and with age , age2 , gender , normalised sleep duration , season of assessment ( dichotomized based on time of the year , and day-light savings time–DST or standard zone time assessments ) as covariates . A fixed-effects meta-analysis was conducted with GWAMA ( Genome-Wide Association Meta-Analysis ) [33] using the inverse-variance-weighted method and low imputation quality ( Rsq/proper_info < 0 . 3 ) were dropped from the meta-analysis . Pathway analyses were carried out in MAGENTA[25] using all available libraries provided with the software . We included all imputed variants with association P<1x10-5 from the continuous chronotype trait . For the results presented in S4 Table , we used gene upstream and downstream limits of 250Kb , excluded the HLA region ( default setting ) and set the number of GSEA ( Gene Set Enrichment Analysis ) [34 , 35] permutations at 10 , 000 ( default ) . We used HaploReg v4 . 1[36] to annotate any coding variants within LD r2 > 0 . 8 of the lead variant at each locus . LocusZoom plots ( Figs 2 , 3 and 5 ) were created using the LocusZoom tool [37] ( found at http://locuszoom . sph . umich . edu/locuszoom/ ) by uploading summary statistics from the Chronotype and sleep duration GWAS . For the background LD structure , we selected the “1000 Genomes Nov 2014 EUR” panel . Genetic correlations ( see [38] for methodology ) between traits were calculated using the LD Score Regression software LDSC ( available at https://github . com/bulik/ldsc/ ) [39] . Summary statistics of our traits outputted by BOLT-LMM were first “munged” , a process that converts the summary statistics to a format that LDSC understands and aligns the alleles to the Hapmap 3 reference panel , removing structural variants and multi-allelic and strand-ambiguous SNPs . Genetic correlations were then calculated between our phenotypes and a set of 101 phenotypes for which summary statistics are publicly available ( full list in S5 Table ) . We used precomputed LD structure data files specific to Europeans of HAPMAP 3 reference panel , obtained from ( http://www . broadinstitute . org/~bulik/eur_ldscores/ ) as suggested on the LDSC software website . The 13 variants in Table 1 which reached P<5x10-8 in combined analyses were used as chronotype instruments in the Mendelian Randomisation analyses . Where binary and continuous traits shared a locus , we selected the top variant of the continuous trait over that of the binary . For loci that reach GW-significance in the binary trait only , we selected the top variant but used the effect size from the continuous trait . To test for a causal effect of BMI on chronotype and sleep-duration , we selected 69 of 76 common genetic variants that were associated with BMI at genome wide significance in the GIANT consortium in studies of up to 339 , 224 individuals ( S6 Table ) [26] . We limited the BMI SNPs to those that were associated with BMI in the analysis of all European ancestry individuals and did not include those that only reached genome-wide levels of statistical confidence in one sex or one stratum only . Variants were also excluded if known to be classified as a secondary signal within a locus . Three variants were excluded from the score due to potential pleiotropy ( rs11030104 [BDNF reward phenotypes] , rs13107325 [SLC39A8 lipids , blood pressure] , rs3888190 [SH2B1 multiple traits] ) , three due to being out of HWE ( rs17001654 , rs2075650 and rs9925964 ) and the last variant due to not being present in the imputed data ( rs2033529 ) . For testing reverse causality of type 2 diabetes on our sleep phenotypes , we used 55 of 65 common variants ( listed in S6 Table ) known to be associated with type 2 diabetes at genome wide significance in a meta-analysis of 34 , 840 cases and 114 , 981 [27] , excluding those known or suspected to be pleiotropic . We performed the Mendelian Randomisation analysis two ways; firstly using instrumental variables ( IV ) using STATA’s “IVreg2” function [40] and secondly through the inverse-variance weighted ( IVW ) and MR-Egger methods described in [41] . Analyses were performed in STATA 13 . 1 ( StataCorp . 2013 . Stata Statistical Software: Release 13 . College Station , TX: StataCorp LP . ) . In the instrumental variables method , we generated genetic risk scores ( GRS ) for BMI and type 2 diabetes using the published list of associated variants and their respective betas . For Chronotype , we generated a GRS using the thirteen replicated variants and their respective betas from 23andMe summary statistics . Using the IVreg2 command , we performed two-stage least squares estimation to calculate the effect of predicted exposure ( through the GRS ) on the continuous outcome traits . For binary outcomes ( type 2 diabetes , undersleeper and oversleeper ) , we manually carried out the two-stage process by regressing the exposure trait on its GRS and storing both predicted values and residuals . We then used these predicted values and residuals as independent variables in a logistic regression where the dependent variable was the binary outcome . The inverse-variance weighted ( IVW ) method is equivalent to a meta-analysis of the associations of the individual instruments and uses associations between the instruments and both the exposure and the outcome to estimate the additive effect of the instruments combined [41] . The MR-Egger method is a modification to the IVW method that allows the inclusion of “invalid” instruments ( i . e . those that don't satisfy all three conditions ) , by performing Egger regression using the summary data of the variants . The IVW and Egger methods operate under the assumption that all instruments are valid , in that they satisfy the three IV conditions: the genetic variants are 1 ) independent of confounders , 2 ) associated with the exposure and 3 ) independent of the outcome . The MR-Egger method , however , accounts for the fact that genetic variants could be pleiotropic and may influence the outcome via pathways other than through the exposure and therefore the resulting association between genetic instruments and the outcome should not be biased by invalid instruments and pleiotropy . The MR-Egger method was used purely as a sensitivity test for the IVW method and so MR-Egger results were not considered if the IVW result did not reach nominal significance . For the IVW and MR-Egger methods , associations of genetic instruments ( variants ) with both exposure and outcome phenotypes were generated in STATA by regressing the phenotype against the instrument while adjusting for covariates . As a further sensitivity test , we also repeated these analyses by replacing exposure phenotype-variant associations with their respective published betas and found only slight differences in betas and P-values , though all exposure-outcome associations remained non-significant .
|
Numerous studies have identified links between too little or too much sleep and circadian misalignment with metabolic disorders such as obesity and type 2 diabetes . However , cause-and-effect is not easily determined , because of multiple confounding factors affecting both sleep patterns and disease risk . Using the first release of the UK Biobank study , which combines detailed measurements and questionnaire data with genetic data , we investigate the genetics of two self-report sleep measures , chronotype and average sleep duration , in 128 , 266 white British individuals . We replicate previous genetic associations and identify seven and two novel genetic variants influencing chronotype and sleep duration , respectively . Associated variants are located near genes implicated in circadian rhythm regulation ( RGS16 , PER2 ) , near a serotonin receptor gene ( HTR6 ) and another gene ( INADL ) encoding a protein thought to be important in photosensitive retinal cells , cells known to communicate with the brain’s primary circadian pacemaker . Using the genetic risk factors , we estimate the unconfounded causal associations of BMI and type 2 diabetes on sleep patterns ( and vice versa ) through Mendelian Randomisation . However , we find no evidence for causal associations in either direction . The full UK Biobank release of 500 , 000 individuals will boost our power to detect causal associations .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"sleep",
"diabetes",
"mellitus",
"physiological",
"processes",
"endocrine",
"disorders",
"mathematics",
"statistics",
"(mathematics)",
"genome",
"analysis",
"chronobiology",
"type",
"2",
"diabetes",
"research",
"and",
"analysis",
"methods",
"endocrinology",
"mathematical",
"and",
"statistical",
"techniques",
"statistical",
"methods",
"schizophrenia",
"mental",
"health",
"and",
"psychiatry",
"genetic",
"loci",
"metabolic",
"disorders",
"circadian",
"rhythms",
"physiology",
"meta-analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"genetics",
"of",
"disease",
"computational",
"biology",
"human",
"genetics"
] |
2016
|
Genome-Wide Association Analyses in 128,266 Individuals Identifies New Morningness and Sleep Duration Loci
|
Hepatitis C virus ( HCV ) infects over 170 million people worldwide and is the leading cause of chronic liver diseases , including cirrhosis , liver failure , and liver cancer . Available antiviral therapies cause severe side effects and are effective only for a subset of patients , though treatment outcomes have recently been improved by the combination therapy now including boceprevir and telaprevir , which inhibit the viral NS3/4A protease . Despite extensive efforts to develop more potent next-generation protease inhibitors , however , the long-term efficacy of this drug class is challenged by the rapid emergence of resistance . Single-site mutations at protease residues R155 , A156 and D168 confer resistance to nearly all inhibitors in clinical development . Thus , developing the next-generation of drugs that retain activity against a broader spectrum of resistant viral variants requires a comprehensive understanding of the molecular basis of drug resistance . In this study , 16 high-resolution crystal structures of four representative protease inhibitors – telaprevir , danoprevir , vaniprevir and MK-5172 – in complex with the wild-type protease and three major drug-resistant variants R155K , A156T and D168A , reveal unique molecular underpinnings of resistance to each drug . The drugs exhibit differential susceptibilities to these protease variants in both enzymatic and antiviral assays . Telaprevir , danoprevir and vaniprevir interact directly with sites that confer resistance upon mutation , while MK-5172 interacts in a unique conformation with the catalytic triad . This novel mode of MK-5172 binding explains its retained potency against two multi-drug-resistant variants , R155K and D168A . These findings define the molecular basis of HCV N3/4A protease inhibitor resistance and provide potential strategies for designing robust therapies against this rapidly evolving virus .
Hepatitis C virus ( HCV ) is a genetically diverse positive-stranded RNA virus of the Flaviviridae family infecting an estimated 170 million people worldwide [1] , [2] . Based on genetic diversity , HCV is divided into six major genotypes ( genotypes 1–6 ) and numerous subtypes with different geographic distributions; genotypes 1 and 3 are the most prevalent worldwide [3] . HCV infection is the leading cause of chronic liver disease that persists for decades and eventually progresses to cirrhosis , liver failure , or liver cancer [4] . The current anti-HCV standard of care is a combination of pegylated interferon ( Peg-IFN ) , ribavirin ( RBV ) , and boceprevir or telaprevir , two recently approved antiviral agents targeting the viral NS3/4A protease [5] . Sustained virologic response ( SVR ) –which is tantamount to cure–is achieved only in a subset of treated patients , depending on a combination of viral and host-cell genetic factors [6]–[10] . For example , a human polymorphism at the IL28B gene is associated with poor interferon response [11] . Most patients undergoing interferon-based therapies also experience significant adverse effects , including flu-like symptoms , anemia , and depression [12] . Thus , current anti-HCV therapies are often not tolerated and ineffective for many patients , and novel direct-acting antiviral drugs are required for safer , more efficacious treatment . Direct-acting antiviral agents have the potential to improve SVR rates and minimize treatment duration . The HCV NS3/4A protease – a chymotrypsin-like serine protease – is a prime therapeutic target that cleaves four known sites along the virally encoded polyprotein [13] . The NS3/4A protease also hydrolyzes two human proteins , TRIF and MAVS , which are part of the innate immune system , thereby confounding the innate immune response to viral infection [14] , [15] . Pharmaceutical companies have invested significant effort in developing NS3/4A protease inhibitors . Proof-of-concept of antiviral efficacy was first demonstrated in 2002 with the macrocyclic inhibitor BILN-2061 ( ciluprevir ) [16] , [17] , which was later discontinued due to concerns about its cardiotoxicity [18] . As noted above , boceprevir [19] and telaprevir [20] , [21] are two NS3/4A protease inhibitors recently approved by the Food and Drug Administration , marking an important milestone in anti-HCV research and drug development over the past two decades . Both boceprevir and telaprevir are linear ketoamide compounds that form a reversible , covalent bond with the catalytic serine of NS3/4A protease . Several non-covalent xprotease inhibitors have also advanced into human clinical trials; these inhibitors include both linear ( BMS-650032 [22] , BI 201335 [23] ) and macrocyclic compounds , containing either a P1–P3 ( danoprevir [24] , TMC435 [25] ) or a P2–P4 ( vaniprevir [26] , MK-5172 [27] ) macrocycle ( Figure 1 ) . The NS3/4A protease inhibitors rapidly reduce HCV RNA titers when administered as monotherapy [17] , [28]–[31] and substantially improve SVR rates when given in combination with Peg-IFN and RBV [6]–[10] , [32]–[34] . However , the high rate of HCV replication and poor fidelity of HCV's RNA-dependent RNA polymerase lead to heterogeneous virus populations in infected patients [35] , [36] . These viral quasispecies exist at low levels in untreated patients , and resistant populations emerge under the selective pressure of direct-acting antiviral agents [36]–[38] . In the majority of patients undergoing protease inhibitor therapy , resistance develops rapidly due to overlapping but distinct sets of NS3/4A mutations [37] . In patients with genotype 1a , the R155K mutation causes resistance against nearly all inhibitors , but rarely occurs in genotype 1b patients [29] , [30] , [32] , [37]–[42] . Instead , distinct resistance mutations arise in genotype 1b patients depending on the class of protease inhibitor used; A156 mutates in response to treatment with linear ketoamide protease inhibitors [39]–[41] , while macrocyclic inhibitors more commonly select for D168A and R155K variants [29] , [30] , [32] , [42] . Mutations at V36 , T54 , and V36+A155 are also associated with resistance to ketoamide inhibitors [39]–[41] . Variations in the patterns of resistant mutations arise from the complex interplay between genotype , replication rates , mutation rates , and the resulting effect of mutations on viral fitness and drug potency . Clearly , despite the benefits of combination therapy in improving SVR rates , the emergence of resistance challenges the long-term efficacy of NS3/4A protease inhibitors . Most primary drug-resistance mutations in NS3/4A protease occur around the active site in regions where drugs protrude from the substrate binding space , defined as the substrate envelope , because these changes can preferentially disrupt drug binding with minimal effect on substrate binding and viral fitness [43] . The protease inhibitors danoprevir , TMC435 , and boceprevir protrude from the substrate envelope in regions that correlate with known sites of resistance mutations . Notably , the large P2 moieties of danoprevir and TMC435 bind in the S2 subsite and extensively interact with residues R155 , D168 , and A156 [43] , which mutate to confer multi-drug resistance [37] , [38] , [44] . These and other inhibitors with large P2 moieties derive much of their potency from binding in the S2 subsite [45] , but how molecular changes at these residues selectively weaken inhibitor binding without compromising the binding of viral substrates is not clear . Elucidating the underlying molecular mechanisms of NS3/4A protease inhibitor resistance is therefore essential for developing new drugs that are less susceptible to resistance . How single-site mutations at residues R155 , A156 and D168 confer resistance against most protease inhibitors has not been elucidated in atomic detail . In this study , we report that four chemically representative protease inhibitors – telaprevir , danoprevir , vaniprevir and MK-5172 – exhibit distinct susceptibilities to the protease variants R155K , A156T and D168A ( Table 1 ) . Sixteen high-resolution crystal structures of inhibitors in complexwith the wild-type protease and three drug resistant variants reveal the molecular basis underlying the unique resistance profiles of these inhibitors ( Table 2 ) . The P2 quinoxaline moiety of MK-5172 stacks against the protease catalytic triad in a novel conformation , explaining its retained potency against R155K and D168A . The flexible P2 isoindoline moiety of danoprevir containing a P1–P3 macrocycle packs against the mutated surfaces of A156T and D168A variants , explaining its relatively higher activity against both protease variants . However , the isoindoline moiety in vaniprevir is constrained due to the P2–P4 macrocycle , resulting in significantly lower activity against all three variants . Thus , incorporating either quinoxaline or flexible substituents at the P2 proline confers clear advantages . Taken together , these data highlight potential strategies for designing novel drugs that retain potency against a broader spectrum of resistant viral variants .
Drug activities were determined for telaprevir , danoprevir , vaniprevir and MK-5172 against wild-type genotype 1a HCV and resistant variants R155K , D168A , and A156T using viral replicon-based inhibition assays . The antiviral activities against the resistant variants trended with changes in binding affinities measured in enzyme inhibition assays ( Table 1 ) . Against wild-type protease , macrocyclic inhibitors danoprevir , vaniprevir and MK-5172 exhibited antiviral potencies in the sub nM range ( IC50 = 0 . 24 , 0 . 34 and 0 . 11 nM , respectively ) , while telaprevir potency was significantly lower ( IC50 = 1030 nM ) , consistent with previous reports [46] , [47] . Relative to the wild type , R155K caused large reductions in potency for danoprevir and vaniprevir , but MK-5172 remained highly active ( R155K IC50 = 0 . 55 nM ) . Telaprevir potency was slightly better against D168A relative to the wild type , while danoprevir , vaniprevir and MK-5172 lost 100- to 1000-fold potency against D168A . However , both danoprevir and MK-5172 still were significantly more potent than telaprevir against D168A . Among the macrocyclic drugs , danoprevir and MK-5172 retained higher activities against D168A ( D168A IC50 = 48 nM and 13 nM , respectively ) relative to vaniprevir ( D168A IC50>400 nM ) . Danoprevir also retained significantly higher potency against A156T ( A156T IC50 = 5 . 7 nM ) , while the other three drugs incurred large-fold losses in potency . Notably , MK-5172 , though active against the other two variants , lost significant potency against A156T ( A156T IC50 = 108 nM ) . Thus , the four drugs exhibited varied susceptibilities to protease inhibitor-resistant viral variants R155K , D168A and A156T . To elucidate the underlying mechanism by which chemically diverse inhibitors bind to the wild-type protease and drug-resistant variants , crystal structures were determined for 16 inhibitor-protease complexes . These complexes include wild-type protease and resistant variants R155K , D168A and A156T each bound to telaprevir , danoprevir , vaniprevir and MK-5172 , with resolutions ranging from 1 . 10–2 . 50 Å ( Table 2 ) ; S139A protease variants were used except for telaprevir , which requires covalent bond formation with the serine 139 for efficient binding . These high-resolution data sets afforded very detailed structural interpretations of drug-protease binding . The binding conformations of telaprevir , danoprevir , vaniprevir and MK-5172 to the wild-type protease are shown in Figure 2 and Figure S1 . In all complexes , inhibitors formed three common hydrogen bonds with the protease backbone ( Table S1 ) : ( 1 ) the P1 amide nitrogen with the carbonyl oxygen of R155 , ( 2 ) the P3 carbonyl oxygen with the amide nitrogen of A157 , and ( 3 ) the P3 amide nitrogen with the carbonyl oxygen of A157 ( Figures 3A–6A ) . The P5 amide nitrogen of telaprevir formed an additional hydrogen bond with the carbonyl oxygen of S159 . In the telaprevir complex , the catalytic serine ( S139 ) was covalently bound to the C-α carbon of the ketoamide warhead . The ketoamide oxygen sat in the oxyanion hole and interacted with the backbone amide nitrogens of protease residues 137–139 , while the Nε nitrogen of H57 hydrogen bonded with the keto oxygen . The acylsulfonamide groups of danoprevir , vaniprevir and MK-5172 were also positioned in the oxyanion hole , hydrogen bonding with the same set of backbone amide nitrogens , as observed previously for the TMC435 and danoprevir structures [43] , [45] . Meanwhile the Nε nitrogen of H57 interacted with the sulfonamide nitrogen in these complexes , suggesting that the Nε atoms were deprotonated . Thus , many of these classes of inhibitors overlap in several key interactions with the protease . In wild-type complexes involving macrocyclic inhibitors , R155 adopted a conformation distinct from those observed in telaprevir and substrate complexes to allow binding of the extended P2 moieties in the S2 subsite . This R155 conformation is stabilized by hydrogen bond interactions involving D168 and D80 . The conformation has also been observed for protease in complex with TMC435 and danoprevir , where large P2 moieties of inhibitors are positioned over the guanidine side chain , making extensive cation-π stacking interactions [43] , [45] . Vaniprevir , with the P2 isoindoline moiety , bound in a conformation similar to danoprevir , making favorable cation-π stacking interactions with R155 , despite the P2–P4 macrocycle . In contrast , MK5172 adopted a novel conformation with the ether-linked P2 quinoxaline moiety not interacting extensively with R155 and D168 , but stacking instead against H57 and D81 of the catalytic triad ( Figure 2 ) . Thus , the P2 moieties of these three macrocycles pack in a variety of conformations around the active site . To characterize binding patterns of the drugs relative to natural substrates , the wild-type drug complexes were analyzed with respect to the substrate envelope , the consensus binding volume of the substrates [43] ( Figures 3A–6A ) . Inhibitors are generally more vulnerable to resistance where they protrude beyond the substrate envelope and contact residues less essential for substrate recognition and turnover . All four drugs protruded from the substrate envelope in the protease S2 subsite near residues R155 , A156 and D168 , which individually mutate to confer multi-drug resistance [37] , [38] , [44] . Telaprevir , with the small P2 cyclopentylproline moiety , made fewer van der Waals contacts with R155 , A156 and D168 relative to danoprevir and vaniprevir , which contain the carbamate-linked P2 isoindoline moieties that protruded from the substrate envelope and made extensive van der Waals contacts with these residues ( Figures 3A–5A ) . Danoprevir's isoindoline moiety bound in two conformations in the wild-type complex , but adopted a single conformation in mutant complexes . Notably MK-5172 , with an ether-linked P2 quinoxaline moiety , while protruding from the substrate envelope , stacked against the catalytic triad , avoiding direct van der Waals contact with R155 and D168 ( Figure 6A ) . Thus , although each of these drugs protruded from the substrate envelope at the S2 subsite , each formed unique interactions with R155 , A156 and D168 . Mutations at these residues therefore differentially affected drug binding and potency , resulting in a distinct resistance profile for each inhibitor . Telaprevir lost potency against R155K compared to the wild-type protease , although the crystal structures of both complexes were very similar maintaining the covalent bond between the ketoamide moiety and the catalytic serine ( Figure 3B ) . R155K , however , lost interactions with D168 , thereby disrupting the electrostatic network spanning R123 , D168 , R155 and D81 , which is important for telaprevir binding . These rearrangements modulated the charge landscape along the protease surface , disrupting interactions with the adjacent P2 cyclopentylproline and P4 cyclohexylalanine moieties of telaprevir , consistent with previous modeling studies [48] . Interestingly , telaprevir showed better potency against the D168A variant than the wild-type; the crystal structure revealed that the P2 moiety bent considerably and packed closer against the D168A variant . The inhibitor shifted by approximately 0 . 5 Å relative the position in the wild-type complex , resulting in increased interactions with both R155 and A156 ( Figures 3C , 7A ) . However , the A156T mutation resulted in a steric clash with telaprevir's P2 moiety , causing the inhibitor to shift significantly; the inhibitor P2 moiety moved away from R155 , losing van der Waals interactions with the protease ( Figures 3D , 7A ) . Notably , in the A156T-telaprevir complex the covalent bond between the ketoamide warhead and the catalytic serine was extended to greater than 2 Å , suggesting a reduced capacity for covalent modification , consistent with the large loss in potency against A156T ( Table 1 ) . Thus , while telaprevir's flexibility allows adaptation to D168A , it cannot accommodate the disruption by R155K or A156T . The relatively weak binding affinity of telaprevir to wild-type protease results in a potentially narrow range by which resistant mutations can be tolerated . For both danoprevir and vaniprevir , the R155K mutation disrupted the favorable cation-π stacking interactions with the P2 isoindoline moieties ( Figure 4B ) , causing significant reductions in drug potencies ( Table 1 ) . The D168A mutation also disrupted stacking of the P2 moieties with R155 , by disrupting the electrostatic network and therefore the position of R155 for optimal cation-π stacking . In danoprevir , the P2 isoindoline moiety shifted in response to R155K and D168A mutations , making extensive interactions with the catalytic D81 ( Figure 7B ) . For vaniprevir , the rigidity of the P2–P4 macrocycle prevented similar compensatory changes ( Figure 5C ) . Thus , D168A caused losses in danoprevir and vaniprevir potency by disrupting cation-π stacking . However , the flexibility of the P2 moiety of danoprevir compensates somewhat for this loss , explaining danoprevir's greater potency against the D168A variant relative to vaniprevir ( Table 1 ) . The A156T mutation sterically impinges on the binding of danoprevir and vaniprevir . In both complexes with A156T , the P2 moieties shifted toward the catalytic triad and lost cation-π stacking interactions with R155 . However , the flexibility of the P2 moiety of danoprevir permitted a larger shift , which allowed for more compensatory packing against the A156T variant protease surface ( Figure 4D ) . In contrast , the P2–P4 macrocycle of vaniprevir restrained the P2 moiety and inhibitor's ability to accommodate this steric burden , more strongly compromising the activity of vaniprevir . Thus , the flexible P2 moiety of danoprevir allowed it to retain significant potency against A156T variants compared to vaniprevir . Unlike in the danoprevir and vaniprevir complexes with wild type , in the MK-5172-wild-type complex the P2 quinoxaline moiety did not stack on R155 and interacted less with D168 and the electrostatic network involving these residues . Thus , the single-site mutations R155K and D168A only caused very subtle changes in the MK-5172 binding conformations ( Figures 6B and 6C ) . This subtle effect is reflected in the small loss of potency against the R155K variant ( Table 1 ) ; however , MK-5172 exhibited 100-fold lower potency against the D168A variant , likely due to less extensive interactions with D81 and K136 relative to wild-type and R155K ( Figure 7 ) . A156T , the worst of the resistance mutations for MK-5172A , sterically clashed with the P2–P4 macrocycle and caused a large shift in the binding position away from the catalytic triad relative to its wild-type structure ( Figure 6D ) . This altered binding of MK-5172 resulted in fewer van der Waals contacts with D81 and R155 , and is likely responsible for 1000-fold lower potency against the A156T variant . Overall , analysis of the four crystal structures explains MK-5172's significantly retained potency against R155K and D168A as well as its loss of potency against the A156T variant due to the rigidity of the macrocycle ( Table 1 ) .
Despite the exciting therapeutic success of NS3/4A protease inhibitors , their long-term effectiveness is challenged by drug resistance . In this study we explain the molecular basis of this drug resistance against four NS3/4A protease inhibitors , telaprevir , danoprevir , vaniprevir and MK-5172 , representing the major chemical classes of these inhibitors . Our detailed analysis of 16 high-resolution crystal structures explains the loss of inhibitor potency in the face of resistance mutations . This research supports our substrate envelope model , which stipulates that inhibitors are vulnerable to resistance where they contact protease residues beyond the substrate-binding region and therefore are not essential for substrate binding [43] . These sites can mutate with minimal effect on protease function and viral fitness . Indeed , most resistance mutations occur in regions where drugs protrude from the substrate envelope , as these changes selectively disrupt drug binding with minimal effect on substrate proteolysis . The most potent of the NS3/4A protease inhibitors is MK-5172 . We report here , for the first time , a novel binding conformation for MK-5172 in which the P2 quinoxaline moiety binds far from the S2 subsite and instead stacks against the catalytic residues H57 and D81 . Unlike other inhibitors , MK5172 does not directly interact with R155 and D168 , which mutate to confer multi-drug resistance . This unique binding mode of MK-5172 explains its significantly greater potency against R155K and D168A variants compared to other inhibitors . MK-5172 has a unique barrier to resistance , as neither catalytic residue ( H57 or D81 ) can tolerate mutation . This binding conformation of MK-5172 , combined with its picomolar binding affinity [27] ( Table 2 ) , will likely allow it to retain potency against a broad array of resistant viral variants and genotypes . We define the structural basis for differential drug activities against the resistant variants R155K , D168A , and A156T for four major chemical classes of NS3/4A protease inhibitors . Telaprevir has reduced potency against R155K due to loss of van der Waals contacts but exhibits better potency against D168A as it allows tighter packing in the S2 subsite . R155K and D168A mutations confer danoprevir and vaniprevir resistance by disrupting favorable cation-π stacking interactions with R155 . Interestingly , while both drugs lose considerable potency against R155K , danoprevir retains higher activity against D168A . This difference is likely due to the flexible P2 isoindoline moiety of danoprevir , which lacks P2–P4 cyclization and repacks against the D168A variant . Similarly , vaniprevir and MK-5172 exhibit significantly lower potency against the A156T variant due to direct steric clashes , while danoprevir partially accommodates this steric burden by repacking against the mutated surface . Thus , the flexibility of danoprevir's P2 isoindoline moiety allows it to retain activity against two of the three major drug-resistant variants . Structural analysis of the 16 protease-inhibitor complexes defines the role of all three major drug-resistance mutations . Our results also provide predictions of drug activities against other HCV genotypes and resistant strains . Interestingly , NS3/4A residues around the protease active site , including R155 , A156 , and D168 are highly conserved except genotype 3 viruses which contain the residues Q168 and T123 , instead of D168 and R123 found in other genotypes ( Figure S2 ) . We predict that the terminal amide group of Q168 will be unable to stabilize R155 for stacking against the P2 moieties of danoprevir and vaniprevir , but may interact with T123 instead . Thus , we expect that danoprevir and vaniprevir will exhibit reduced potencies against genotype 3 viruses , while MK-5172 will remain fully active . Indeed , danoprevir and vaniprevir were recently shown to have reduced efficacy against genotype 3 viruses [49] . For genotype 1 strains , our results indicate that MK-5172 is highly active against R155K and D168A variants , while danoprevir is highly active against the A156T variant and to a lesser extent against the D168A variant . Thus , as new inhibitors are developed and HCV resistance testing becomes more available , our findings can help guide anti-HCV treatment regimens for individual patients . Overall our findings correlate with resistance profiles observed in clinical isolates . Most protease inhibitors select for R155K variants in genotype 1a patients as only one nucleotide change is required [29] , [30] , [32] , [37]–[42] . Genotype 1b patients presumably have higher barriers to R155K resistance , requiring two nucleotide substitutions; thus , mutations at A156 and D168 are more readily observed in response to protease inhibitor treatment . The resistance at R155K occurs due to reduced interactions in the S2 subsite . Telaprevir and other linear ketoamide drugs select for A156T variants [39]–[41] by direct steric clashes , while linear ( BI 201335 ) and macrocyclic drugs ( danoprevir , vaniprevir , TMC435 ) with large P2 moieties select for D168A variants [29] , [30] , [32] , [42] by disrupting favorable stacking interactions with R155 . These data also support the converse observation that D168A variants are uncommon in patients treated with telaprevir as the drug can pack tighter in the S2 subsite . Likewise , A156T variants are uncommon in patients treated with macrocyclic drugs containing flexible P2 moieties due to drug repacking against the mutated protease surface [29] , [30] , [32] , [42] . However , drugs such as vaniprevir and MK-5172 containing P2–P4 cyclization likely select for the A156T variant due to the rigidity of their P2 moieties . Whether A156T variants will be found in clinical isolates , however , depends on additional viral factors , such as relative differences in viral fitness between A156T variants and other competing viral variants . Our data thus provide a unique resource for preemptively predicting resistance and choosing the most appropriate protease inhibitor to treat HCV depending on the resistance profile of a particular patient viral population . Whether or not specific mutations arise in clinical isolates is ultimately determined by the complex interplay between drug potency , viral fitness , and genetic barriers to resistance . Thus , depending on the initial viral species altered pathways to resistance will exist . The crystal structures of these NS3/4A inhibitors also provide a key resource to guide future strategies in drug design . The high potency of MK-5172 , for example , derives from interactions with the essential catalytic triad residues , which cannot mutate without severely disrupting viral fitness . In addition , flexible P2 drug moieties – lacking P2–P4 macrocycles – mitigate losses in potency to the A156T and D168A mutations . Similar chemical features can be incorporated in future drugs to potentially evade resistance . Specifically , novel protease inhibitors that incorporate flexible P2 moieties , such as quinoxaline or similar groups , could exploit interactions with the essential catalytic residues and concurrently minimize contact with the P2 subsite , thereby reducing their sensitivities to mutations at R155 , D168 and A156T . Thus , our findings suggest strategies for developing protease inhibitors that retain activity against a wider spectrum of drug-resistant HCV variants .
Danoprevir , vaniprevir and MK-5172 were synthesized in house following reported methods; danoprevir was prepared using our convergent reaction sequence as described [43]; vaniprevir and MK-5172 were prepared following the synthetic methods reported by McCauley et al . [50] and Harper et al . [27] , respectively , with minor modifications . Telaprevir was purchased from A ChemTek , Inc . ( Worcester , MA ) . The HCV genotype 1a NS3/4A protease gene described in a Bristol-Meyers Squibb patent [51] was synthesized by GenScript and cloned into the pET28a expression vector ( Novagen ) . This highly soluble single-chain construct of the genotype 1a NS3/4A protease domain contains a fragment of the cofactor NS4A covalently linked at the N-terminus [51] . A similar protease construct exhibited catalytic activity comparable to that of the authentic full-length protein [52] . All protease variants were generated using the QuikChange Site-Directed Mutagenesis Kit from Stratagene . The codon-optimized genotype 1a helicase sequence ( H77c ) was cloned downstream to the protease gene to generate the full-length protease construct . Geneious [53] was used to generate the sequence alignment of the NS3/4A protease domain from HCV genotypes 1–6 . Single mutations ( R155K , D168A , or A156T ) were introduced into the NS3 region of genotype 1a HCV Con1 luciferase reporter replicon using the mega-primer method of mutagenesis [54] . Replicon RNA of each protease variant was introduced into Huh7 cells by electroporation . Replication was then assessed in the presence of increasing concentrations of protease inhibitors ( telaprevir , danoprevir , vaniprevir or MK-5172 ) by measuring luciferase activity ( relative light units ) 96 hours after electroporation . The drug concentrations required to inhibit replicon replication by 50% ( IC50 ) were calculated directly from the drug inhibition curves . For enzyme inhibition experiments , 5 nM of the genotype 1a HCV NS3/4A protease domain was incubated with increasing drug concentrations for 15 min ( 90 min for telaprevir ) in 50 mM Tris assay buffer ( 5% glycerol , 5 mM TCEP , 6 mM LDAO and 4% DMSO , pH 7 . 5 ) . Proteolysis reactions were initiated by adding 100 nM HCV NS3/4A substrate [Ac-DE-Dap ( QXL520 ) -EE-Abu-ψ-[COO]AS-C ( 5-FAMsp ) -NH2] ( AnaSpec ) and monitored using the EnVision plate reader ( Perkin Elmer ) at excitation and emission wavelengths of 485 nm and 530 nm , respectively . The initial cleavage velocities were determined from sections of the progress curves corresponding to less than 15% substrate cleavage . Apparent inhibition constants ( Ki ) were obtained by nonlinear regression fitting to the Morrison equation of initial velocity versus inhibitor concentration using Prism 5 ( GraphPad Software ) . Data for each drug were generated in triplicate and processed independently to calculate the average inhibition constant and standard deviation . Protein expression and purification were carried out as described [51] , [55] . Briefly , transformed BL21 ( DE3 ) E . coli cells were grown at 37°C and induced at an optical density of 0 . 6 by adding 1 mM IPTG . Cells were harvested after 5 hours of expression , pelleted , and frozen at −80°C for storage . Cell pellets were thawed , resuspended in 5 mL/g of resuspension buffer ( 50 mM phosphate buffer , 500 mM NaCl , 10% glycerol , 2 mM β-ME , pH 7 . 5 ) and lysed with a cell disruptor . The soluble fraction was retained , applied to a nickel column ( Qiagen ) , washed with resuspension buffer , and eluted with resuspension buffer supplemented with 200 mM imidazole . The eluent was dialyzed overnight ( MWCO 10 kD ) to remove the imidazole , and the His-tag was simultaneously removed with thrombin treatment . The nickel-purified protein was then flash-frozen and stored at −80°C . The above-mentioned protein solution was thawed , concentrated to ∼3 mg/mL and loaded on a HiLoad Superdex75 16/60 column equilibrated with gel filtration buffer ( 25 mM MES , 500 mM NaCl , 10% glycerol , 30 µM zinc chloride , and 2 mM DTT , pH 6 . 5 ) . The protease fractions were pooled and concentrated to 20–25 mg/mL with an Amicon Ultra-15 10 kD device ( Millipore ) . The concentrated samples were incubated for 1 hour with 1–3 molar excess of inhibitor . Diffraction-quality crystals were obtained overnight by mixing equal volume of concentrated protein solution with precipitant solution ( 20–26% PEG-3350 , 0 . 1 M sodium MES buffer , 4% ammonium sulfate , pH 6 . 5 ) in 24-well VDX hanging drop trays . X-ray diffraction data were collected at Advanced Photon Source LS-CAT 21-ID-F , GM/CA-CAT 23-ID-D or with the in-house RAXIS IV X-ray system . Diffraction intensities were indexed , integrated and scaled using the program HKL2000 [56] . All structure solutions were generated using simple isomorphous molecular replacement with PHASER [57] . The B chain model of viral substrate product 4A4B ( 3M5M ) [43] was used as the starting model for all structure solutions . Initial refinement was carried out in the absence of modeled ligand , which was subsequently built in during later stages of refinement . Subsequent crystallographic refinement was carried out within the CCP4 program suite , with iterative rounds of TLS and restrained refinement until convergence was achieved [58] . The protein crystals of the wild-type protease and drug-resistant variants R155K and D168A in complex with MK-5172 grew as pseudo-merohedral twins . Amplitude-based twinned refinement was carried out during restrained refinement for all pseudo-merohedral twins . The final structures were evaluated with MolProbity [59] prior to deposition in the Protein Data Bank . To limit the possibility of model bias throughout the refinement process , 5% of the data were reserved for the free R-value calculation [60] . Interactive model building and electron density viewing was carried out using the program COOT [61] . Fobs−Fcalc ligand omit maps were generated with the ligand excluded from the phase calculation using the program PHENIX [62] . Superpositions were performed in PyMOL [63] using the Cα atoms of the active site protease residues 137–139 and 154–160 . The wild-type-danoprevir complex was used as the reference structure for each alignment . The NS3/4A viral substrate envelope was computed as described using the full-length NS3/4A structure ( 1CU1 ) [64] and product complexes 4A4B ( 3M5M ) , 4B5A ( 3M5N ) and 5A5B ( 3M5O ) [43] . Van der Waals contact energies between protease residues and peptide products were computed using a simplified Lennard-Jones potential as described [65] . Briefly , the Lennard-Jones potential ( Vr ) was calculated for each protease-drug atom pair where r , ε and σ represent the interatomic distance , vdW well depth , and atomic diameter , respectively:Vr was computed for all possible protease-drug atom pairs within 5 Å , and potentials for non-bonded pairs separated by less than the distance at the minimum potential were equated to −ε .
|
Hepatitis C virus ( HCV ) infects over 170 million people worldwide and is the leading cause of chronic liver diseases , including cirrhosis , liver failure , and liver cancer . New classes of directly-acting antiviral agents that target various HCV enzymes are being developed . Two such drugs that target the essential HCV NS3/4A protease are approved by the FDA and several others are at various stages of clinical development . These drugs , when used in combination with pegylated interferon and ribavirin , significantly improve treatment outcomes . However HCV evolves very quickly and drug resistance develops against directly-acting antiviral agents . Thus , despite the therapeutic success of NS3/4A protease inhibitors , their long-term effectiveness is challenged by drug resistance . Our study explains in atomic detail how and why drug resistance occurs for four chemically representative protease inhibitors –telaprevir , danoprevir , vaniprevir and MK-5172 . Potentially with this knowledge , new drugs could be developed that are less susceptible to drug resistance . More generally , understanding the underlying mechanisms by which drug resistance occurs can be incorporated in drug development to many quickly evolving diseases .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"biochemistry",
"infectious",
"diseases",
"medicinal",
"chemistry",
"chemistry",
"biology",
"chemical",
"biology",
"biophysics"
] |
2012
|
The Molecular Basis of Drug Resistance against Hepatitis C Virus NS3/4A Protease Inhibitors
|
We present a technique applicable in any dynamical framework to identify control-robust subsets of an interacting system . These robust subsystems , which we call stable modules , are characterized by constraints on the variables that make up the subsystem . They are robust in the sense that if the defining constraints are satisfied at a given time , they remain satisfied for all later times , regardless of what happens in the rest of the system , and can only be broken if the constrained variables are externally manipulated . We identify stable modules as graph structures in an expanded network , which represents causal links between variable constraints . A stable module represents a system “decision point” , or trap subspace . Using the expanded network , small stable modules can be composed sequentially to form larger stable modules that describe dynamics on the system level . Collections of large , mutually exclusive stable modules describe the system’s repertoire of long-term behaviors . We implement this technique in a broad class of dynamical systems and illustrate its practical utility via examples and algorithmic analysis of two published biological network models . In the segment polarity gene network of Drosophila melanogaster , we obtain a state-space visualization that reproduces by novel means the four possible cell fates and predicts the outcome of cell transplant experiments . In the T-cell signaling network , we identify six signaling elements that determine the high-signal response and show that control of an element connected to them cannot disrupt this response .
A key goal in the study of complex dynamical systems is to extract important qualitative information from models of varying specificity ( e . g . , [1 , 2] ) . This has been approached via the construction and analysis of qualitative models ( e . g . , discrete models [3–7] ) and also by analytic techniques applied to continuous systems [8–13] . In this work , we present and implement a new approach to identifying control-robust subsystem behavior that can drive the dynamics of the system as a whole . Our approach applies to a large class of continuous , discontinuous , and discrete models . Interacting systems are partially described by their regulatory networks . In these networks , nodes represent each of the various interacting entities within the system , and signed edges represent direct positive or negative influence . To better understand the temporal character of the system , one can construct a dynamical model on the regulatory network . First order Ordinary Differential Equations ( ODEs ) are a natural choice for such models . The influence upon the value of each entity , xi , is represented as x ˙ i = F i ( x ) , where the dependence of Fi upon xj is consistent with the influence of entity j upon entity i . A validated model can be used to gain practical insights about the system , such as how to drive it into a desired attractor . There are two key challenges to the construction and analysis of ODE models of complex interacting systems . First , there is often large uncertainty in measurements of variable and parameter values . Second , these systems are typically high-dimensional , which complicates phase-portrait visualization and other traditional qualitative analyses . One approach to these challenges is to choose a more qualitative model . Discrete models have been used to successfully model many biological phenomena , including pattern formation and multistability [3 , 4] . Despite the vast reduction in state-space afforded by discretization of variable values , exhaustive searches for dynamical behaviors are computationally infeasible in high-dimensional systems . Several methods for identifying the causal structure of state-space in discrete models have been proposed , including hierarchical transition graphs [14] and prime implicant graphs [15] . An analogous concept in ODE models is that of positive invariant sets ( also called “trap spaces” ) [16 , 17] . These are regions of state space that system trajectories may enter but not exit . By identifying such spaces , one may make predictions about the evolution of a system without integrating the governing ODEs . A second strategy is that of examining features in the dynamical repertoire that arise directly from the associated regulatory network and weak assumptions about the form of the dynamic model . Structural controllability relates branching patterns in the regulatory network to the identification of control targets that are sufficient to drive linear dynamics on the network into any state [18] . This allows one to study system control near a steady state . In many biological and chemical systems , however , the dynamics are nonlinear and large disruptions from equilibrium are of interest . In such systems , even when the dynamics are not specifically known , regulatory feedback loops provide useful information for global control [19–23] . For example , given relatively permissive continuity and boundedness assumptions , an ODE-described system can be driven into any of its attractors by controlling any set of variables whose removal eliminates all feedback loops and external inputs [19 , 21 , 24] . Positive feedback loops in particular are associated with the presence of multistability [8–12] , which has been of particular interest in biomolecular systems because it is necessary for cell-fate branching and decision making [4 , 25–28] . Two existing approaches to identifying the effects of positive feedback loops are especially relevant here . The first of these is the methods put forth by Angeli and Sontag for studying monotone input-output systems ( MIOS ) [29] . Their approach identifies steady states and their stability in systems lacking negative feedback loops or incoherent feed-forward loops ( in the general meaning of two directed paths of opposite sign between two nodes ) . The second is based on the concept of stable motifs of Boolean dynamical systems [22 , 30] . This method constructs an auxiliary network that encodes the regulatory logic within its graph structure ( in a similar vein as logic hypergraphs [5 , 31] ) , enabling efficient identification of the system’s dynamical repertoire . Within this auxiliary network , certain graph structures , called stable motifs , correspond to positive feedback subsystems that sustain steady states that are impervious to influence from the rest of the network ( see S1 Appendix section 1 or [30] for further details ) . In other words , stable motifs determine positive invariant sets . This observation connects the concept of positive invariant sets to the regulatory network in the Boolean case . Our work extends this connection to the continuous case . Our framework encodes the causal relationships between variable constraints as the network structure of an expanded network . An edge from one constraint ( e . g . , x > 0 ) to another ( e . g . , y > 0 ) indicates that the first ( x > 0 ) is sufficient to maintain the second ( y > 0 ) . The expanded network helps to identify low-dimensional subsystems that drive higher-dimensional dynamics . We show that stable modules , source-free expanded subnetworks subject to certain consistency criteria , correspond to control-robust positive-invariant sets of the originating dynamical system . Variables obeying stable module constraints must be directly controlled ( i . e . , either receive exogenous input or be made control variables ) if the constraints are to be broken . This identifies variables that must be controlled to disrupt certain behaviors ( or , equivalently , it identifies variables that cannot be controlled in such a way as to disrupt the behavior ) . It is non-trivial to choose relevant variable constraints for the modeled system , but in practice , the form of the regulatory functions often suggests natural candidates . Furthermore , we leverage MIOS techniques to algorithmically specify meaningful constraints in a class of systems common in biology ( see S1 Appendix section 2 ) . This is implemented ( S1 Source Code ) as code that systematically scans for stable modules in an input ODE system satisfying certain assumptions . Identifying several stable modules in a systematic search highlights “decision points” in subsystems that determine system-wide outcomes .
The core of our analysis strategy lies in the interpretation of an auxiliary network that is constructed from the dynamical system of interest . Following previous work in Boolean systems [3 , 32] , we call this auxiliary network an expanded network . An expanded network must be constructed from a given dynamical system . It is a network on a node set consisting of statements about the values of variables ( or , equivalently , consisting of the regions of state-space in which these statements are true ) . There are two types of directed edges between nodes . The first type , the maintenance edge , indicates that one statement cannot become false while the other is true . The second type , the driving edge , indicates that the sustained truth of the first statement implies that the second statement will eventually become true . In this paper , our focus is on continuous , autonomous ODE systems , although the concepts are presented in such a way as to be readily adapted to other types of dynamical systems . In the following , we describe the nodes and edges of an expanded network in more detail . In S1 Appendix section 3 , we provide a formal mathematical foundation for the following discussion . In an expanded network , there are two types of nodes: virtual and composite . Virtual nodes are statements about the values of dynamic variables that can be assigned a definite truth value at any given time ( e . g . , the virtual node “x > 0” is true only when the value of the variable x is positive ) . Virtual node statements can be viewed as regions of state-space , and are true at time t if x ( t ) is in the corresponding region . Composite nodes also take Boolean values , and correspond to the composition of virtual nodes by “AND” ( ∧ ) rules . Each composite node receives directed edges from its factor virtual nodes . As such , all factors of a composite node must be represented as virtual nodes in the expanded network . For example , the composite node x > 0 ∧ y > 0 is true only when x and y are positive , and there are directed edges from x > 0 and y > 0 to this composite node . In deterministic finite-level systems , it is possible to choose a finite number of statements that fully characterize the state space [33] , but in general , the nodes of an expanded network embody partial information about the system . For a given choice of virtual and composite nodes , the expanded network is unique , however , a different choice of virtual nodes for the same system can lead to different expanded networks . Some choices of virtual nodes are therefore more illuminating than others , and choosing an informative set of virtual nodes is not always straightforward . In the next section , we propose and implement a method to address this difficulty in a particular class of systems . The remainder of this section covers general expanded network properties , which are prerequisite for the methods of the next section . Virtual nodes can receive two types of edges: a maintenance edge or a driving edge , with the latter being a more restrictive version of the former . If a virtual or composite node X must be false before a virtual node Y can change from true to false , we say that X maintains Y and we draw a directed edge from X to Y in the expanded network . Note that if a virtual node X describes a positive-invariant set in state-space that remains positively invariant even under control of variables not involved in the definition of X , then according to this definition X maintains X , which results in a self-loop on X . This can happen if X describes a self-activating variable , for example . In determining whether X maintains Y , we must consider all valid variable values that might disrupt Y when X is true . These variable values are drawn from the region of state-space in which the model is valid and experimentally accessible , e . g . a box in state-space defined by the maximum and minimum values of each variable . By considering values from this region of validity we simultaneously evaluate a large number of system trajectories and control strategies . To explore whether control of one system element can drive the system as a whole into particular regions of state-space , one may also wish to impose the condition that an edge from X to Y indicates that the truth of X implies the truth of Y in finite time ( or , more briefly , X drives Y ) ; this additional constraint is unnecessary when considering self-sustaining behavior . A subnetwork , S , of an expanded network , N , is a stable module if it satisfies three conditions: ( i ) all nodes X in S have a parent ( regulator ) node in S ( possibly X itself if it has a self-loop ) , ( ii ) if a composite node X = ⋀ i = 1 n X i is in S , then Xi is also in S for i = 1‥n , and ( iii ) it is possible for all nodes in S to be simultaneously true . For brevity , we refer to subnetworks satisfying conditions ( i ) , ( ii ) , or ( iii ) , as source-free , composite-closed , or consistent , respectively . Our key result is the following: if all nodes in a stable module are simultaneously true ( i . e . , if in that instant the system is in a region of state-space for which all virtual node statements in the stable module are true ) , then they remain true under all state-space configurations in the region of validity . In the following we will call a stable module whose nodes are simultaneously true an active stable module . To prove our key result , consider by way of contradiction an active stable module , S , that deactivates . Let Y ∈ S be a virtual node that becomes false before or concurrently with any other node in S . Because every node in S has a parent node in S , there is X ∈ S that maintains Y . By the definition of maintenance edges , X ( or one of its factors if it is composite ) must become false before Y does , violating the selection criteria and thereby proving the result . A stable module with no stable submodules is a stable motif . Under the condition that a stable module , S , is active , we can simplify the expanded network by removing any edges that point from a virtual node in S ( e . g . , x > 0 ) to a composite node outside of S ( e . g . , x > 0 ∧ y > 0 ) because the condition expressed by this edge is now satisfied . We can also remove any node that is necessarily false when S is active ( e . g . , if S contains the node x > 0 , the node x < 0 can be removed ) . Stable motifs of the modified expanded network are then identified and added to S in the original expanded network . We thus iteratively form larger stable modules , building a sequence of stabilized subsystems that drive system dynamics . When the activity of a stable module in one sequence implies the inactivity of at least one stable module in another sequence , these sequences are mutually exclusive . Collections of mutually exclusive sequences describe the system’s dynamical repertoire . Our definition of stable motifs encompasses the definitions of stable motifs given in [30] for Boolean systems ( see S1 Appendix section 1 ) and in [33] for multi-level systems . This allows us to generalize many results from discrete modeling to general dynamical systems . In particular , generalizing arguments in [22] , we consider system control via expanded network topology . It is often of interest to identify variables that can activate a stable module ( which may correspond , e . g . , to a healthy cell state ) . This can be achieved by solving the graph-theoretic problem of identifying stable module driver nodes . A module driver node set D of module M in an expanded network is a set of virtual nodes D such that the truth of all nodes in D implies the truth of all nodes in M in finite time . Therefore , identification of a driver node set for a stable module prescribes a control strategy to trigger the module behavior . Conversely , if a stable module represents undesired behavior ( e . g . , a diseased cell state ) , one might seek to disrupt it . Because stable modules are self-sustaining , control of variables not represented in the undesired module can never achieve this goal . Disruption of a stable module requires direct control of at least one of its represented variables . To illustrate the method , and some of its utility , we analyze a toy example ( Fig 1 , Eq 1 ) . In this toy example , we will choose statements for virtual nodes somewhat arbitrarily , with the goal of illustrating how relationships between nodes in the expanded network can be identified and interpreted . In later sections , we introduce a more systematic approach to selecting virtual nodes that does not rely on the intuition of the investigator . u˙=11+z−u3w˙=y−w/2x˙=1+4w+4wz ( 1+2w ) ( 1+2z ) −xy˙=xx+1/2−yz˙=xf ( y ) −zf ( y ) ≥fmin>0 ( 1 ) Here , we have very limited information about f ( y ) ; perhaps it is stochastic or discontinuous . Nevertheless , by uncovering upper and lower bounds on components of the ODE vector field , we can begin to assemble an expanded network one edge at a time . For instance , if x > 1/2 holds , then z ˙ > f m i n / 2 - z is implied . If z is positive and decreasing ( z ˙ < 0 ) , it cannot decrease faster than fmin/2 − z . In this case z would asymptotically approach fmin/2 . As a consequence , z will never fall to zero . Therefore , as long as x is greater than 1/2 , z cannot fall below 0 once it has become positive , and so we say that x > 1/2 maintains z > 0 . A similar argument applies in any case when x is larger than an arbitrary positive value . Furthermore , if z is not positive , then z ˙ is strictly greater than fmin/2 . Therefore z will eventually ( in finite time ) become larger than zero and so we say x > 1/2 drives z > 0 . We can therefore conclude that there is an edge from x > 1/2 to z > 0 in the expanded network . Similarly , we see that x will be maintained above 1/2 if w > 1/2 and z > 0 are both true . We therefore identify a composite node ( w > 1/2 ) ∧ ( z > 0 ) with incoming edges from w > 1/2 and z > 0 , and an outgoing edge to x > 1/2 . We continue to identify edges in the expanded network and search for stable modules . Some of the subgraphs of the expanded network that can be generated in this way are depicted in Fig 1 alongside the traditional network representation of the system . We have identified three stable modules , thereby proving , for example , that if the systems satisfies x , y , w , z > 0 at any time , it will always satisfy those conditions ( as follows from the yellow module in the bottom left of Fig 1 ) . The other two modules contain x , y > 1/2 and z > 0 as well as either w > 1 or w > 1/2 ( see bottom right panel of Fig 1 ) . Thus , if the system satisfies the four conditions given by either module , it will continue to do so for all time . The arguments underlying the construction of the subgraphs of the expanded network hold for any f ( y ) > fmin > 0 , and so we have extracted meaningful qualitative information despite large dynamical uncertainty . In addition to the expanded networks and their subgraphs containing stable modules , many that do not contain stable modules also exist ( e . g . , the top right panel of Fig 1 ) . Such networks contain information regarding the consequences of directly controlling particular nodes so that they satisfy virtual node statements ( e . g . , if we fix y < 0 , we see that w will eventually become negative ) . Choosing virtual nodes defined by inequalities , as is our main focus here , has important implications for how oscillations are observed . If a variable oscillates , but remains above or below some threshold , the statement indicating the variable value relative to that threshold can be part of a stable module . Alternatively , oscillations can manifest in the expanded network as subnetworks with contradictory virtual nodes . For instance , if x ˙ = z + s i n ( y ) - x , then z > z0 ( where z0 is an arbitrary positive number ) maintains ( and drives ) x > z0 − 1 and z < z0 maintains ( and drives ) x < z0 + 1 . The main difficulty in identifying stable modules is determining what statements are most useful for inclusion in the expanded network . If the statements are too general , then either the results will not provide much insight , or the network will be too sparse because the statements are not sufficiently restrictive to imply one another . If a statement is too restrictive , on the other hand , it may have an in-degree of zero in the expanded network , in which case it cannot be part of a stable motif . Despite these challenges , we have found a straightforward approach to analyzing threshold behavior of a large class of biologically relevant systems . We consider a broad class of dynamical systems that take the form x ˙ i = F i ( x ) , ( 2 ) where Fi is continuous , monotonic in each of its arguments , and strictly decreasing in xi . This class of ODEs describes many biological systems ( S1 Appendix section 2 ) and is particularly well-suited to analysis in our framework . The essential steps of the stable module identification process are as follows . First , we identify all subgraphs of the regulatory network that are composed of positive feedback loops . For each such subgraph , we construct two families of candidate stable modules by conjecturing that each variable xi in the regulatory subnetwork has a virtual node of the form xi > Ti or xi < Ti , where Ti is left unspecified ( for brevity , we denote this form by x i ≶ T i α ) . For each candidate module , we construct a “worst-case” monotone system by replacing any variable regulatory effects that would introduce a negative feedback loop or incoherent feedforward loop by constant regulatory effects . This system is analyzed using the techniques of [29] such that equilibria of the worst-case system yield thresholds Ti for which the candidate stable module is genuine . In the following we provide the details of the process . To each variable xi of a regulatory subnetwork under consideration , we assign a set of thresholds { T i α } and consider virtual node statements of the form x i ≶ T i α . At this stage , each T i α may remain parameterized and the statements need not cover the full dynamical range of xi . We create composite nodes ⋀ k = 1 m ( x i k ≶ T i k α k ) as needed . Next , we conjecture that particular edges exist in the expanded network for some ( unspecified ) choice of threshold parameters . For instance , when activity of one variable , x1 , is sufficient for activation of another , x2 , we would hypothesize the formation of an edge x1 > T1 → x2 > T2 . In the conjectured expanded network , we find source-free , consistent , and composite-closed subgraphs , which serve as candidate stable modules . Consider a candidate stable module , Sc , in the conjectured expanded network . Consider also the regulatory subnetwork , Gc , made up of nodes represented in Sc and all incident edges . Some of these incident edges are represented in Sc , while other “external” edges are not . For example , a candidate stable module Sc in Fig 1 might be y > Ty → w > Tw → x > Tx → y > Ty and the corresponding regulatory subnetwork Gc consists of the positive cycle x , y , w and the additional external edge from z to x . We note that external edges may exist between two nodes in Gc if the regulatory relationship between these variables is not part of Sc . To identify bounds for the virtual nodes that ensure that the candidate stable module Sc is genuine , we use the monotone input-output systems ( MIOS ) methods of Angeli and Sontag [29] , which apply to sign-consistent systems ( see S1 Appendix section 4 ) . The relationships represented in Sc constitute a sign-consistent subnetwork ( S1 Appendix section 4 ) . Any sign-inconsistencies in Gc arise from external edges . To construct a sign-consistent modified subsystem for Sc , we consider each variable xi represented in Sc . Any external regulation of xi by yj is held fixed by replacing yj with a “worst-case” value in Fi . The “worst-case” value is chosen such that xi is as close as possible to T i α in the stable module node x i ≶ T i α; because Fi is monotonic in each argument by assumption , this is either yj ≡ inf yj or yj ≡ sup yj ( i . e . , when yj is as large or small as possible within the region of validity ) . For example , if yj negatively regulates xi and x i < T i α is in Sc then we evaluate F i | y j = infy j . Because the resulting modified subsytem is sign-consistent , we can apply the MIOS procedure of Angeli and Sontag ( Theorem 3 of [29] ) . For examples of this process for sign-consistent systems , see [20 , 29] . To do this , we must verify that we can select a variable , xk , called the “MIOS input variable” that has the property that maintaining xk at a constant value drives the system to a single steady state for all initial values of variables other than {xk} [20 , 29] . The form of Eq 2 implies that a node in the modified system satisfies these conditions if its removal makes the modified system acyclic . Once we have verified that a MIOS input variable can be chosen , we can follow Angeli and Sontag [20 , 29] to find the steady states of the modified subsystem . These steady states determine the thresholds that we use for the virtual nodes in Sc . The sign-consistency of the modified subsystem implies that these thresholds describe a positive invariant set of that subsystem ( [29] ) . This sign-consistency together with the monotonicity of the regulatory functions implies that this set remains positively invariant for all possible values of the external regulatory effects because any deviation in these from their worst case values unambiguously drives the system away from the boundary of the stable module subspace and into its interior . Therefore , with these thresholds , Sc is realized as a valid stable module for the original system . We illustrate this method by identifying a candidate module and constructing a worst case system in the example of Eq 1 . First , we recall that we have already shown that the system is restricted to the positive orthant if the initial conditions are within this region , so we assume that this is our region of validity . In general , identification of the region of validity often follows from physical or biological considerations . By inspection , we observe that y activates w , which activates x , which in turn activates y . We thus conjecture that a stable module of the form y > Ty → w > Tw → x > Tx → y > Ty exists . Note that this feedback loop is positive and defines a loop closure of a monotone system when z is viewed as a parameter . To identify valid bounds for this candidate stable module ( if such bounds exist ) , we construct the worst case system for the candidate . As the only regulatory effect not represented in the candidate is the effect of z on x , we must identify the value of z , within the region of validity , for which x ˙ is minimized . In this case , x ˙ is minimized when z is maximized , and so we allow z to tend toward infinity in the worst case system , yielding a worst case system given by x ˙ = 2 w 1 + 2 w - x , along with w ˙ and y ˙ from Eq 1 . The steady state of this system is given by the solution of the feedback characteristic equation x = ( 2 x x + 1 / 2 ) / ( 2 x x + 1 / 2 + 1 / 2 ) , which has solution x = 7/10 , yielding w = 7/6 and y = 7/12 . We thus conclude that y > 7/10 → w > 7/6 → x > 7/10 is a stable module . We provide additional examples in sections 5 and 6 of S1 Appendix . We have algorithmically implemented ( S1 Source Code ) this process by considering intersecting unions of positive feedback loops . For each union , we conjecture two stable modules ( in which one set of nodes is “high” and the other is “low” , and vice versa ) . Using user-specified physical system bounds , we construct a “worst case system” for each candidate stable module , as described above , and test the existence of a MIOS input variable . If such a variable can be found , we use it to numerically find the steady states via the MIOS procedure . If any steady states are within the physical system bounds , we return the corresponding stable module . The above procedure returns a list of stable modules involving threshold statements about subsystem variables connected by positive feedback loops . Note that generally the list of stable modules we generate does not directly correspond to all of the system’s equilibria , or even necessarily to equilibria at all . Rather , it corresponds to “trap” subspaces , i . e . , positive-invariant sets , that are robust to control of regulatory effects external to the subsystem . If the control includes multiple regulatory effects , we assume that these effects can be controlled independently of each other . The list of stable modules generated for each subsystem is in one-to-one correspondence with the equilibria of this subsystem that are robust to such control . This list thus contains the subsystem behaviors that are self-sustaining under all control strategies that preserve the topological structure of the regulatory network . Additional behaviors may be robust to only a subset of these interventions . In the remainder of this paper , we use the above methodology and automation scheme to analyze two systems from the literature . The first , the Drosophila segment polarity gene network , is a prototypical system used to study a broad class of embryonic pattern formation mechanisms . The second example is the T-cell signaling network , which is a characteristic representative of signal transduction networks , which lead to specific cell responses to environmental signals .
The original multicellular model of the Drosophila segment polarity gene network [34] uses coupled ODEs to model the concentrations of mRNAs and proteins of a family of genes that are important for the development of segments in Drosophila melanogaster embryos ( see Fig 2 ) . This family of genes includes engrailed and cubitus interruptus , which encode transcription factors , as well as wingless and hedgehog , whose proteins are secreted and interact with proteins in the neighboring cells [34–36] . We use a modified version of this model ( equations 12-23 in [35] ) , which has incorporated more recent experimental results ( e . g . , on the sloppy-paired protein ) and been recast for a single cell while assuming steady-state values for neighboring cells . Because no measured values of the kinetic parameters in the model are available , and because our purpose here is illustrative , we have simply chosen parameter values from the biologically relevant parameter region ( see S1 Appendix section 7 for parameter values and variable abbreviations ) . We identify several stable modules of biological importance in this model . When neighboring cells have high levels of wingless protein , we find two stable modules distinguished by differential sloppy-paired and engrailed expression ( red and blue nodes in Fig 2 ) . For high concentrations of neighboring hedgehog protein , we find two stable modules involving the wingless sub-network ( yellow and purple nodes in Fig 2 ) . By shading the nodes in the expanded network according to module membership ( as in Fig 2 ) we can visually identify regions of state-space that correspond to different attractors of the system . Specifically , these attractors distinguish the four cell-types observed in the development of Drosophila melanogaster segments , which we label PC1-PC4 [34 , 36] ( see Fig 2 ) . Furthermore , the expanded network highlights the causal chains that link regions of state-space and establish cell fates . By identifying driver node sets for stable modules , we can prescribe control strategies to attain any of the four cell types . For example , drivers of the cell type PC1 ( blue module in Fig 2 ) are high neighboring hedgehog ( Hnbr ) and low sloppy-paired ( sp or SP ) . Furthermore , we can use this information to form hypotheses about the outcome of altering node states . For example , we can make the following prediction about the outcome of a future wet-bench experiment in which a cell of a certain type is transplanted to a region in which neighboring cells express hedgehog and wingless at higher or lower levels relative to the cell’s initial neighbors . Consider a cell of type PC1 ( blue module in Fig 2 ) . If the neighboring wingless ( Enbr ) and hedgehog ( Hnbr ) are reversed in expression level , that disrupts the engrailed-sloppy paired part of the module . As a result , en and sp approach zero and one , respectively . The values of wingless ( wi ) and the two configurations of its protein before transplant are consistent with the stable module characterizing cell type PC2 ( yellow module in Fig 2 ) . Therefore , our analysis of the model ( [35] ) suggests that a qualitative change in cell gene expression from that of the foremost cell of the embryonic segment ( PC1 , blue module ) to that of the second segmental cell ( PC2 , yellow module ) would be observed in such a transplant experiment . Numerical simulations support this conclusion ( S1 Fig ) . Our analysis also identifies the reason for this change: the engrailed-sloppy paired feedback loop is not robust to elimination of neighboring wingless ( Enbr ) . If this prediction is falsified by follow-up experiments , the lack of transition would imply the existence of additional regulation of engrailed and/or sloppy paired . The additional regulation would need to act in such a way as to allow a high expression of engrailed in the absence of neighboring wingless . The second biological example we consider here is a model that describes the cascading activation of transcription factors when T-cell receptors are bound by external molecules [37] . The model was constructed using the Odefy MATLAB toolbox ( [38] ) to transform a pre-existing Boolean model of T-cell activation ( [31] ) into an ODE model x ˙ i = F i ( x ) = ( R i ( x ) - x i ) / τ i , where each Ri is a polynomial of Hill functions with Ri ( x ) ∈ [0 , 1] describing the regulatory effects that influence the production of xi . The parameters τi are the inverse degradation rates of the various biomolecules . To simplify the example , we consider the strongly connected core of the system with saturated input signals , though the precise signal strength has little impact on the analysis . The resulting network is depicted in Fig 3 ( left ) , in which the edges are labeled with the Hill coefficient , n , and disassociation constant , k , of the function Hi ( xi ) for the corresponding regulatory effect . By considering when the activation or inhibition of a given node is sufficient or necessary to cause the activation of other nodes , we have identified the cycle TCRb→Fyn→PAG→Lck→ZAP→cCbl→TCRb as a candidate stable motif depicted in Fig 3 ( right ) . This cycle is a positive feedback loop , but it is embedded in a sign-inconsistent network . As such , before we implement the MIOS approach to determine valid thresholds for the motif , we must address the effects of sign-inconsistent edges ( [29] ) . For instance , in the motif , we expect TCRb and PAG to achieve relatively high values , but there is an inhibitory effect between the two; indeed , τ P A G d P A G d t = ( 1 - H 1 ( T C R b ) ) ( 1 - H 2 ( F y n ) ) + H 2 ( F y n ) - P A G , ( 3 ) where H1 and H2 are Hill functions ( of the form x n x n + k n with n ∈ Z + and k ∈ ( 0 , 1 ) ) . The inhibitory effect is maximized when TCRb attains its maximum value , i . e . , one ( because all variables are normalized to their maximum values in this model ) . It is also possible to consider the possibility that TCRb is delivered to the system via external control , in which case we would evaluate Eq 3 in the limit as TCRb → ∞ . For now , we shall only consider TCRb = 1 in this regulatory function . We therefore replace H1 ( TCRb ) in Eq 3 with H1 ( 1 ) and allow TCRb to evolve according to its natural dynamics in this new network , in which the regulation of PAG is modified . Similar analysis is taken on any edge that either introduces a sign inconsistency , or does not connect two nodes of the stable motif . The resulting modified network is a single positive feedback loop with a single steady state that is easily identified using the MIOS approach [29] . The steady state values of the nodes in the modified network serve as thresholds in the expanded network , and allow us to identify a stable motif ( see Fig 3 ) . In this example , the stable motif we have identified coincides with a global steady state of the system . This observation is in agreement with [17] , in which this system is analyzed by application of theorems regarding the conservation of certain positive invariant sets when a system is described by both a Boolean and an ODE model with Hill regulatory functions . We note that our analysis does not rely on a particular functional form of the regulation or on an explicit companion Boolean model . A novel result of our analysis is that the stable motif behavior cannot be disrupted by manipulating TCRp . We demonstrate the robustness of the stable motif by numerically solving the system ODEs with various constraints placed on TCRp ( Fig 3 ) . In the top left panel of Fig 3 , we show a natural evolution of the system for initial conditions satisfying the stable motif conditions . In the other panels , the value of the TCRp node is subjected to one of three external controls ( absence , saturation , and oscillation ) , and the motif variables continue to respect the stable motif conditions . These simulations illustrate an important conclusion we can draw from the existence of the stable motif: If one wishes to avoid states in which Fyn , PAG , and TCRb are high while Lck , ZAP , and cCbl are low , TCRp is not a viable control target . Biologically , this model predicts that disruption of TCR phosphorylation is not sufficient to disrupt the response of the cell to a high degree of receptor-ligand binding . Instead one must disrupt one of the six motif nodes directly , and furthermore , the motif bounds provide lower bounds on the magnitude of the required disruption . For example , to disrupt the motif via control of PAG , one must lower its value below the threshold of 0 . 69 .
We have presented a new framework , based upon construction of an auxiliary “expanded network” , for identifying self-sustaining subsystems that cannot be controlled via the rest of the system . Full attractor control requires that variables from each of these subsystems be externally manipulated . We have applied our framework to develop an algorithm ( S1 Source Code ) for finding these subsystems that is applicable in many biological ODE models . We have demonstrated our framework and algorithm in two biological systems: the T-cell receptor signaling network and the Drosophila melanogaster segment polarity gene network . The method of expanded networks can extract important qualitative features from quantitative or qualitative models of system behavior . We have emphasized the identification of stable modules , which correspond to state-space regions that , once entered , cannot be exited without directly applying external control on the variables that define the region boundaries . We have also shown , for example in our analysis of the Drosophila melanogaster segment polarity gene network , how the consideration of expanded networks can elucidate meaningful and intuitive partitioning of state-space . In these analyses , we have considered virtual nodes of the form x i ≶ T i α , but other choices for virtual nodes are possible , and can be informative when xi has inherently multi-level behavior . In searching for stable modules , it is important to identify positive feedback loops , as every stable module of the type considered in our automation procedure corresponds to a sign-consistent subgraph that must contain at least one positive cycle . In the examples described here , as well as in every ODE model of biological systems we encountered so far , the number of positive feedback loops is small , and so an exhaustive search is feasible . Even when this number is large , an exhaustive test of all sign-consistent subgraphs is probably still faster than a brute-force simulation approach because this method is testing many control strategies simultaneously for each subsystem , and does not involve integration of any ODEs . Positive feedback loops can be identified using existing software implementations . For each positive feedback loop , the computational complexity of determining the associated thresholds scales linearly with the number of variables in the feedback loop . If necessary , we are also able to limit our search to positive feedback loops of a certain size , allowing for fast identification of small , control-robust subsystems embedded in much larger systems . Our procedure yields all stable modules of threshold statements about variables involved in positive feedback loops . These correspond to positive-invariant sets that remain positively invariant even when regulatory effects external to the feedback loop are manipulated . Because we cannot know a priori which regulatory manipulations are available within a given model , we have chosen to focus on behaviors that are robust to all manipulations of these external regulations . Therefore , the stable modules we identify are robust to control beyond that which can be implemented in practice . In some systems , additional behaviors may exist that are robust only to a biologically relevant subset of control strategies . Nevertheless , knowledge of the fully robust system behaviors reduces ( in some cases , dramatically ) the search space for control targets . For example , in a large system , we might identify a stable module that includes some “undesirable” behavior ( e . g . , a disease state ) and involves a small number of variables . Because the stable module subsystem is robust to all topology-preserving external controls , control targets must be selected from the small number of variables directly involved in the stable module . Many existing results about the analysis of Boolean models via expanded networks remain valid in this more general framework and can therefore be applied to continuous systems . An example is the concept of a driver node set , which is a set of virtual nodes in the expanded network whose truth eventually implies the truth of a given stable module . Identification of driver nodes in the expanded network is related to finding paths in logic hypergraphs [31] . This identification problem has been partially addressed in [30 , 39]; developing a general and fast algorithm for driver node identification in arbitrary expanded networks is a promising direction for future research with applications for control target selection . Some results do not generalize as easily because they rely upon completeness properties of discrete expanded networks; the oscillation analyses in [30 , 33] are an example . Oscillations can manifest in the expanded network as source-free graph components that contain contradictory nodes . Such structures do not always indicate oscillatory behavior , and may instead indicate chaotic behavior or the existence of a steady state that violates all of the contradictory conditions . For example , the simple harmonic oscillator x ˙ = y , y ˙ = - x has an expanded network with contradictory source-free component x > 0 → y < 0 → x < 0 → y > 0; while the system can oscillate between satisfying these conditions , there is also a steady state , x = y = 0 , that violates all four conditions . We are optimistic that results of this type might be recast in more general forms . The expanded network framework shows promise not only for studying the state-space of dynamical systems , as we have emphasized here , but also for the study of parameter space . Statements regarding the value of parameters can be included in an expanded network as statements with self-loops . Because the expanded network approach extracts qualitative information from the system , the inclusion of parameters in this way is conceptually distinct from and complementary to existing methods for probing the parameter space of a dynamical system ( e . g . , [40 , 41] ) . The application of expanded networks to parameter sensitivity analyses is the subject of ongoing work .
|
We show how to uncover the causal relationships between qualitative statements about the values of variables in ODE systems . We then show how these relationships can be used to identify subsystem behaviors that are robust to outside interventions . This informs potential system control strategies ( e . g . , in identifying drug targets ) . Typical analytical properties of biomolecular systems render them particularly amenable to our techniques . Furthermore , due to their often high dimension and large uncertainties , our results are particularly useful in biomolecular systems . We apply our methods to two quantitative biological models: the segment polarity gene network of Drosophila melanogaster and the T-cell signal transduction network .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"signaling",
"networks",
"animals",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"systems",
"science",
"mathematics",
"network",
"analysis",
"experimental",
"organism",
"systems",
"sequence",
"motif",
"analysis",
"drosophila",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"computer",
"and",
"information",
"sciences",
"network",
"motifs",
"system",
"stability",
"animal",
"studies",
"bioinformatics",
"dynamical",
"systems",
"insects",
"arthropoda",
"eukaryota",
"gene",
"regulatory",
"networks",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"organisms"
] |
2018
|
Identifying (un)controllable dynamical behavior in complex networks
|
Sensing invading pathogens early in infection is critical for establishing host defense . Two cytosolic RIG-like RNA helicases , RIG-I and MDA5 , are key to type I interferon ( IFN ) induction in response to viral infection . Mounting evidence suggests that another viral RNA sensor , protein kinase R ( PKR ) , may also be critical for IFN induction during infection , although its exact contribution and mechanism of action are not completely understood . Using PKR-deficient cells , we found that PKR was required for type I IFN induction in response to infection by vaccinia virus lacking the PKR antagonist E3L ( VVΔE3L ) , but not by Sendai virus or influenza A virus lacking the IFN-antagonist NS1 ( FluΔNS1 ) . IFN induction required the catalytic activity of PKR , but not the phosphorylation of its principal substrate , eIF2α , or the resulting inhibition of host translation . In the absence of PKR , IRF3 nuclear translocation was impaired in response to MDA5 activators , VVΔE3L and encephalomyocarditis virus , but not during infection with a RIG-I-activating virus . Interestingly , PKR interacted with both RIG-I and MDA5; however , PKR was only required for MDA5-mediated , but not RIG-I-mediated , IFN production . Using an artificially activated form of PKR , we showed that PKR activity alone was sufficient for IFN induction . This effect required MAVS and correlated with IRF3 activation , but no longer required MDA5 . Nonetheless , PKR activation during viral infection was enhanced by MDA5 , as virus-stimulated catalytic activity was impaired in MDA5-null cells . Taken together , our data describe a critical and non-redundant role for PKR following MDA5 , but not RIG-I , activation to mediate MAVS-dependent induction of type I IFN through a kinase-dependent mechanism .
The innate immune response allows for the rapid production of type I interferons ( IFNs ) and other proinflammatory cytokines to counteract invading viral pathogens . This response relies , in part , on a group of molecules collectively referred to as pattern recognition receptors ( PRRs ) , which recognize pathogen-associated molecular patterns generated during the course of infection . The detection of virus infection is mediated primarily by cytoplasmic sensors for both RNA and DNA , which include members of the RIG-like helicase ( RLH ) family for RNA detection and a variety of cytoplasmic proteins for detection of DNA [1] . To date , there are three members of the RLH class of PRRs , all of which are cytosolic RNA helicases that recognize double-stranded RNA ( dsRNA ) : retinoic acid-inducible gene I ( RIG-I ) [2] , melanoma differentiation-associated gene 5 ( MDA5 ) [3] and laboratory of genetics and physiology-2 ( LGP2 ) [4] . The RLH proteins belong to a family of DExD/H box-containing RNA helicases , and in addition , RIG-I and MDA5 possess two N-terminal caspase recruitment and activation domains ( CARDs ) , and a C-terminal regulatory domain [2 , 3] . Despite this homology , each sensor displays a different affinity for distinct dsRNA ligands and , hence , different viruses [5] . RIG-I , the most extensively studied member of the RLH family , recognizes short dsRNA segments bearing 5' triphosphate structures [6–9] , whereas MDA5 recognizes long dsRNA that likely harbor higher-ordered RNA structures [10 , 11] . LGP2 , unlike RIG-I and MDA5 , lacks the N-terminal CARD domains necessary for activating downstream signaling components and was initially identified as a negative regulator of RIG-I [4 , 12] . However , more current evidence suggests that LGP2 may instead be a positive regulator of both RIG-I and MDA5 [13 , 14] . Although RIG-I and MDA5 recognize different dsRNA motifs , both PRRs converge on a single adaptor protein to stimulate a signaling cascade inducing IFN [15 , 16] . This adaptor protein ( designated MAVS throughout this work ) is known variously as IFNβ promoter stimulator 1 ( IPS-1 ) [17] , mitochondrial antiviral signaling protein ( MAVS ) [18] , virus-induced signaling adaptor ( VISA ) [19] or CARD adaptor inducing IFNβ ( CARDIF ) [20] . MAVS contains a single CARD , a central proline-rich region and a C-terminal hydrophobic region that anchors the protein to the outer mitochondrial membrane [21] . Upon activation , RIG-I or MDA5 bind MAVS via CARD-CARD interactions and ubiquitin chains , resulting in MAVS aggregation on the mitochondrial membrane [22] . Aggregated MAVS provides a platform for recruitment of signaling molecules leading to activation of transcription factors critical for IFN gene induction , including NFκB , ATF2/c-jun and members of the interferon regulatory factor ( IRF ) family [23] . Among the MAVS-recruited adaptors are members of the tumor necrosis factor ( TNF ) receptor-associated factor ( TRAF ) family members , specifically TRAF2 , TRAF3 and TRAF6 [19 , 24 , 25]; which transmit signals downstream to the inhibitor of NFκB kinase ( IKK ) complex to drive nuclear translocation of NFκB [20 , 26] . Additionally , MAVS signals to TRAF family member-associated NFκB activator ( TANK ) binding kinase 1 ( TBK1 ) and IKKε , to phosphorylate IRF3 and IRF7 , and induce transcriptional activation of IFN and other proinflammatory cytokines . IFN can signal in an autocrine or paracrine manner to trigger induction of hundreds of IFN-stimulated genes ( ISGs ) that act to fortify host defenses . Of note is PKR , an ISG whose activity is triggered by the presence of foreign dsRNA . PKR was initially identified for its ability to inhibit viral protein translation in vaccinia virus ( VV ) -infected cells [27] . PKR is a serine/threonine kinase that contains two domains , a N-terminal regulatory dsRNA binding domain , and a C-terminal catalytic domain [28 , 29] . Although PKR expression is inducible by IFN , it is also expressed at a basal level in most cells as an inactive monomer . Latent PKR must be activated to be functional , which is achieved by binding dsRNA or by a variety of other stimuli . Once activated , PKR undergoes a dimerization-dependent conformational change and autophosphorylation , stimulating its catalytic activity toward protein substrates , the best characterized being eukaryotic initiation factor 2 α-subunit ( eIF2α ) [29] . Phosphorylation of eIF2α results in inhibition of protein translation . Although inhibition of translation is the best characterized function for PKR , it has been implicated in additional cellular responses , including apoptosis and autophagy [30] . Mounting evidence also suggests a role for PKR in the production of type I IFN and other proinflammatory cytokines [31] . PKR may play a role in the activation of NFκB , a transcription factor required for IFN induction , presumably via its interaction with the IKK complex [32] . Additionally , PKR has also been shown to interact with several members of the TRAF family , including TRAF2 and TRAF6 [33] , two proteins involved in MAVS signaling . Moreover , a number of studies have shown that PKR is involved in IFN induction in response to polyriboinosinic:polyribocytidylic acid ( pIC ) , a dsRNA mimetic [34–36] . PKR may also be important for IFN induction in response to some , but not all , viral infections [37–40] . Here , we report evidence that PKR is required for type I IFN induction in response to viruses recognized by MDA5 . This role for PKR was independent of its canonical function in translation inhibition and phosphorylation of the initiation factor eIF2α , but nonetheless , dependent on its catalytic activity . Induction of IFN in this MDA5-PKR-dependent manner occurred through IRF3 activation , and likely involved a physical interaction between PKR and MDA5 . Furthermore , PKR activation by virus was impaired in cells lacking MDA5 , and studies involving artificial PKR activation in the absence of virus infection demonstrated IRF3 phosphorylation and IFN expression in a MAVS-dependent , but MDA5-independent fashion . Taken together , these results suggest MDA5 and PKR cooperate to transduce dsRNA-dependent induction of type I IFN production .
Vaccinia virus ( VV ) , like all members of the Orthopoxvirus family , encodes multiple virulence determinants that function by impairing host innate immune responses . In particular , the E3L protein is capable of sequestering viral RNA and inhibiting PKR activation , thus blocking a critical component of IFN-dependent antiviral responses [41] . To assess IFN induction in the absence of this inhibitor , we measured IFNβ expression in cells following infection with a mutant virus lacking this gene , VVΔE3L . Infection of WT ( WT ) mouse embryo fibroblasts ( MEF ) with VVΔE3L resulted in abundant expression of IFNβ ( Fig 1A ) . As expected , IFNβ expression was completely dependent on the presence of the transcription factor IRF3 , as shown by the lack of IFNβ expression following infection of Irf3-/- cells . In an effort to identify the sensor mediating the innate response to VVΔE3L , the expression of IFNβ was measured in the absence of the cytosolic RNA sensors , MDA5 and RIG-I ( Fig 1B ) . In parallel with a known stimulator of MDA5 , encephalomyocarditis virus ( EMCV ) ( Fig 1C ) , VVΔE3L induction of IFN was also dependent on MDA5 ( Fig 1B ) . Interestingly , the alternative RNA sensor , RIG-I , was superfluous for IFNβ expression in response to VVΔE3L infection , since IFNβ expression was normal in infected RigI-/- cells . As expected , IFNβ expression in cells infected by Sendai virus ( SeV ) , a negative-sense RNA virus , showed distinct sensor dependence . SeV-infected cells expressed IFNβ in the absence of MDA5 , but this response was undetectable in the absence of RIG-I ( Fig 1D ) . Therefore , VVΔE3L appears to produce ligands that are recognized by MDA5 , but not by RIG-I , in the absence of the E3L protein , consistent with previous reports suggesting MDA5-dependent inflammatory responses to VV [42] . Vaccinia virus E3L protein and influenza virus NS1 protein are both virulence factors that , due to their affinity for dsRNA , have the ability to inhibit PKR activity [43–47]; infection in their absence efficiently triggers PKR function . Since E3L can abrogate IFN induction [35] and PKR is one of its major targets , we investigated the role of PKR in the cellular response to infection with the VV mutant , VVΔE3L . We infected WT and PKR-null MEF cell lines with VVΔE3L or influenza A virus lacking the NS1 protein ( FluΔNS1 ) . Infection of WT MEFs with either VVΔE3L or FluΔNS1 resulted in the induction of IFNβ mRNA . ( Fig 2A ) However , PKR-null MEFs demonstrated a significant impairment in IFNβ mRNA levels in response to VVΔE3L , but not FluΔNS1 infection . In fact , the absence of PKR boosted IFNβ levels in response to FluΔNS1 infection , possibly due to positive feedback that is otherwise prevented when protein synthesis is inhibited by activated PKR [48] . PKR activity is triggered during viral infection by interaction with dsRNA , which can be mimicked by transfection of pIC [29] . Consistent with previous reports [49 , 50] , cells lacking PKR showed a significant reduction in pIC-induced IFN compared to WT MEFs ( Fig 2B ) . We also assayed the role of PKR in human cell lines by stably reducing its expression in A549 lung adenocarcinoma cells with a short hairpin RNA ( A549 shPKR; S1 Fig ) . Consistent with results in PKR-null MEFs , A549 shPKR cells showed significantly lower levels of IFNβ mRNA following infection with VVΔE3L compared to A549 cells expressing a non-specific shRNA ( A549 shNS; Fig 2C ) . However , levels of IFNβ following FluΔNS1 infection were not impaired in the A549 shPKR cells . Similar to the result with FluΔNS1 in PKR-null MEFs ( Fig 1A ) , IFNβ was induced and even enhanced in response to FluΔNS1 infection following PKR knockdown ( Fig 2C ) . Impaired IFN induction in PKR-null and PKR-knockdown cells during VVΔE3L infection indicated a key role for this enzyme in mediating this response . To assess whether the catalytic function of PKR was required , we treated cells with a PKR inhibitor ( PKR-I ) that prevents RNA-induced PKR autophosphorylation [51] , and measured IFNβ induction following VVΔE3L infection . WT MEFs were treated with PKR-I and infected with either VVΔE3L or FluΔNS1 . Both viruses induced substantial expression of IFNβ in vehicle-treated cells ( DMSO ) that was approximately 100-fold greater than uninfected samples ( Fig 3A ) . PKR-I treatment impaired IFNβ mRNA levels in a dose-dependent manner following infection with VVΔE3L , but not FluΔNS1 . This result demonstrated that PKR catalytic activity contributed to IFNβ induction . A critical component of the antiviral action of PKR is derived from inhibition of host protein translation due to the phosphorylation of the eukaryotic translation initiation factor 2 α-subunit ( eIF2α ) [52] . To assess the contribution of this activity to IFNβ induction during VVΔE3L infection , we examined responses in a cell line that contains a mutation in eIF2α at serine 51 ( eIF2α S51A ) . This mutation blocks its ability to be phosphorylated by eIF2α kinases [53] and renders cells resistant to the translational inhibitory effects of PKR [54] . Interestingly , this mutant cell line showed no deficit in VVΔE3L-induced IFNβ mRNA levels ( Fig 3B ) , indicating that eIF2α phosphorylation and , presumably , PKR-mediated protein translation shutoff were not involved in PKR-mediated IFN induction . Similarly , there was no deficit in IFN induction following infection with the RIG-I-activating viruses SeV and FluΔNS1 . In fact , both viruses induced greater levels of IFN in the absence of eIF2α phosphorylation ( Fig 3B ) , mimicking the heightened induction of IFN observed in response to these viruses in PKR-null and PKR-knockdown cells ( Fig 2A and 2C ) . Phosphorylation of eIF2α leads to formation of stress granules in virus-infected cells , a process that appears to contribute to innate antiviral immunity [55] . Stress granules contain stalled protein translational machinery , due at least in part to translational inhibition from eIF2α inactivation . To determine if stress granule formation due to eIF2α phosphorylation by PKR explained the requirement for PKR during VVΔE3L infection , we probed WT and eIF2α S51A cells for stress granules , using an antibody against the stress granule marker , G3BP1 [56] . As shown in Fig 3C , WT cells displayed stress granule formation in response to infection ( upper panel ) . They also displayed cytoplasmic DAPI-staining bodies that likely represent vaccinia poxvirus factories [57] . In contrast , S51A cells failed to display G3BP1-positivity , although DAPI-staining poxvirus factories were present in infected cells ( Fig 3C , lower panel ) . Approximately 64% of infected WT cells displayed stress granules , while examination of multiple fields of infected mutant cells failed to detect any G3BP1-positive cytoplasmic granules . This observation is consistent with previous reports that virus-induced stress granule formation is dependent on eIF2α S51 phosphorylation [58] . Coupled with the observation that eIF2α S51A cells produced IFN in response to infection ( Fig 3B ) , these data suggest that stress granule formation during VVΔE3L infection is not required for IFN production . [55] To further confirm that host translational shutoff was not the attribute of PKR activation that was required for IFN induction , we infected WT and PKR-null MEFs with VVΔE3L or FluΔNS1 while inhibiting translation pharmacologically with cycloheximide ( CHX ) . If the inability of PKR-null MEFs to shut off translation was causing the deficit in IFN induction , as has been demonstrated for other viruses [59] , then CHX treatment to induce PKR-independent translational arrest might be expected to rescue IFN induction in the absence of PKR . We tested this notion by comparing IFNβ induction in VVΔE3L-infected PKR-deficient cells in the presence and absence of CHX . However , despite elevated basal levels of IFNβ mRNA following CHX treatment that probably reflects mRNA stabilization [60] , VVΔE3L infection was still incapable of inducing IFNβ in PKR-null MEFs ( Fig 3D ) . Taken together , these data suggest that although PKR and its catalytic function are required for VVΔE3L-induced IFN production , its canonical antiviral effector functions involving eIF2α phosphorylation , translational inhibition , and subsequent stress granule formation are not involved . Our data suggest that PKR plays an important role in IFN induction in response to VVΔE3L infection or pIC treatment ( Fig 2B ) , but not FluΔNS1 ( Fig 2A ) . Since IRF3 is required for the induction of IFNβ [61 , 62] , we asked whether its activation was impaired during infection of cells lacking PKR . To this end , WT and PKR-null MEFs were infected with a panel of viruses ( Fig 4 ) , and examined for IRF3 nuclear translocation as an indication of activation . Consistent with data for FluΔNS1 ( Fig 2A and 2C ) , which displayed no dependency on PKR for IFN production , infection of PKR-null MEFs with another RIG-I agonist and strong inducer of IRF3 phosphorylation , Sendai virus ( SeV ) , did not result in a significant defect in IRF3 activation ( Fig 4 ) . In contrast , IRF3 activation in response to either VVΔE3L or EMCV , both MDA5 stimulators , was abrogated in the absence of PKR , as indicated by the inability to detect nuclear IRF3 in extracts from cells infected with these viruses ( Fig 4 ) . PKR-null cells were infected at least as well as WT cells , as assayed by quantitation of viral RNA following infection ( S2 Fig ) , confirming that failure to induce IRF3 translocation was not due to a defect in infectivity of PKR-null cells . Taken together , these data place failure to activate IRF3 , a necessary component of IFN induction ( Fig 1A ) , as a proximal cause of impaired IFNβ induction in the absence of PKR . The requirement for PKR in IFN induction by the various viruses and treatments we tested correlated with the known involvement of MDA5 in signaling . Specifically , the ability of the tested agonists to induce IRF3 phosphorylation or IFNβ expression in the absence of PKR fell into two distinct categories based on the PRR responsible for their recognition: those that require RIG-I for IFN induction , FluΔNS1 and SeV [5] , were PKR independent , while those that require MDA5 , VVΔE3L ( Fig 1B ) , EMCV and pIC [63] were PKR-dependent . Involvement of PKR downstream of MDA5 prompted us to investigate the possibility of direct protein interactions . HEK293T cells were transfected with FLAG-tagged versions of RIG-I , MDA5 , the E3L protein from VV , a known PKR-interacting protein , or STAT2 , a cytoplasmic signaling and transcription factor with no known connection to PKR [64] . Transfected cells were infected with VVΔE3L , and protein interactions were detected by co-immunoprecipitation . Immunoprecipitation with anti-FLAG antibodies and probing for PKR showed co-purification of endogenous PKR with RIG-I , MDA5 , as well as E3L , but not with STAT2 ( Fig 5A ) . The PKR interaction with MDA5 occurred in both uninfected as well as VVΔE3L-infected cells . Interestingly , the interaction between PKR and either RIG-1 or E3L depended on VVΔE3L infection , while PKR interaction with MDA5 did not follow this trend ( Fig 5A ) . Because both MDA5 and PKR are RNA-binding proteins , we explored whether this protein-protein interaction was dependent on nucleic acid . To this end , extracts from transfected cells were incubated with micrococcal nuclease prior to immunoprecipitation . MDA5 and PKR co-immunoprecipitated regardless of nuclease treatment ( Fig 5B ) , indicating that their interaction was not bridged by RNA . To extend these results , we tested the interaction between endogenous proteins in human A549 cells ( Fig 5C ) . Since MDA5 is a low abundance protein , we induced its levels by pretreatment of cells with IFN ( Fig 5C , lanes 1 and 2 ) . Under these conditions , immunoprecipitation of endogenous PKR effectively purified endogenous MDA5 ( lane 4 ) , while control purifications using non-specific IgG recovered neither protein ( lane 3 ) . The abundant cellular protein p38 MAPK served as a control for non-specific binding and was not recovered by either antibody treatment . Together with the overexpression studies , these results confirm that MDA5 and PKR associate in a complex independent of viral infection and RNA binding . To directly assess a requirement for PKR downstream of RIG-I and MDA5 , we tested signaling from ectopically expressed and constitutively active RLH proteins . Full-length MDA5 , which is sufficient to induce IFN when over-expressed , or a constitutively-active RIG-I mutant expressing only the N-terminal CARD domains ( RIG-I-CARD ) [2] , were transduced into A549 cells following PKR knockdown ( S1 Fig ) , and levels of IFN mRNA were measured ( Fig 5D ) . The expression of RIG-I-CARD induced comparable levels of IFNβ in the control A549 shNS and knockdown A549 shPKR cells ( Fig 5D ) . However , MDA5 overexpression induced significantly lower levels of IFNβ in the A549 shPKR compared to the A549 shNS control cells . Together , these results are consistent with our previous observations that PKR was required for IFNβ induction by MDA5 , but not by RIG-I ligands . Comparable levels of RIG-I-CARD and MDA5 were transduced into both cell lines , as judged by expression of the linked IRES-mCherry marker ( S3 Fig ) . To further determine whether PKR dependence of MDA5-induced IFN expression was also observed at the level of transcriptional activation , activation of an IFNB-luciferase reporter was measured in response to the constitutively active constructs ( Fig 5E ) . Transfected HEK293T cells depleted for PKR ( S4 Fig ) displayed reduced activation of an IFNβ reporter in response to expression of MDA5 . A potential caveat to the conclusion that PKR mediates MDA5 , but not RIG-I signaling , could be that the constitutively active RIG-I construct lacked the RNA-binding , helicase , and regulatory regions of the protein , while the analogous domains were intact in the MDA5 construct . To address whether non-CARD domain regions could mediate an RLR-PKR connection , we interrogated the role of PKR downstream of full-length RIG-I . To this end , we ectopically expressed a constitutively active full-length RIG-I mutant that retains RNA binding ability [65] . In contrast , but consistent with results in Fig 5D , both RIG-I-CARD and mutated full-length RIG-I mediated induction of IFNβ regardless of the presence or absence of PKR . Similarly , we tested whether the dependence of MDA5 on PKR involved its RNA-helicase domain . Cells were transfected with truncated MDA5 expressing only the CARD domains ( MDA5-CARD ) , analogous to the RIG-I-CARD construct . Interestingly , MDA5-CARD induced IFN expression in a similar manner to RIG-I-CARD , regardless of the presence or absence of PKR ( Fig 5F ) . These data suggest a critical dependence of MDA5 signaling on PKR that is mediated by its RNA helicase domain and is distinct from the action of RIG-I . To directly assess the ability of PKR to induce IFN , we activated PKR catalytic function in the absence of viral infection . Fusing the PKR catalytic domain ( residues 258–551 ) to the dimerization domain of E . coli Gyrase B ( GyrB ) protein ( GyrB . PKR ) creates a conditionally active version of the kinase that responds to coumermycin treatment [66 , 67] . In this context , GyrB , which dimerizes in the presence of coumermycin , replaces the dsRNA-binding domain that normally regulates PKR . Treatment of HT1080-GyrB . PKR cells with coumermycin resulted in an increase in the levels of phosphorylated eIF2α after 8 hours ( h ) in the absence of viral infection ( Fig 6A ) . Coumermycin treatment of HT1080-GyrB . PKR cells induced robust levels of IFNβ expression ( Fig 6B ) , but IFNβ levels remained unaffected when coumermycin was used to dimerize a catalytically inactive form of GyrB . PKR ( K296H ) ( Fig 6B , green bars ) , consistent with the requirement for intact PKR catalytic function ( Fig 3A ) . Treatment with coumermycin induced IRF3 phosphorylation in a manner equivalent to VVΔE3L infection ( Fig 6D ) , suggesting that activation of PKR is sufficient to activate IRF3 . Additionally , coumermycin treatment induced the expression of the NFκB target genes , TNFα and IL1β ( S5 Fig ) , consistent with previous reports linking PKR to NFκB activation [32] . However , coumermycin treatment had no effect on the expression of an IFNγ-induced gene , GBP1 ( Fig 6C ) , which , unlike IFNβ , is dependent on STAT1 and IRF1 rather than IRF3 and NFκB [68] . These data indicated that PKR was necessary and sufficient for IFN production . To probe the epistatic relationship of PKR relative to other signaling components in this pathway , we silenced either MDA5 or MAVS in HT1080-GyrB . PKR cells and activated PKR catalytic function with coumermycin after 8 and 24 h ( Fig 7 ) . Silencing MDA5 ( Fig 7A ) had little effect on IFNβ induction in response to PKR activation after 8 h , but a reproducible difference was detected after 24 h of coumermycin stimulation ( Fig 7B ) . Interestingly , MDA5 loss abrogated IFN induction in response to VVΔE3L infection , indicating that PKR function is epistatic to MDA5 . In contrast , silencing MAVS ( Fig 7D ) largely abrogated the ability of PKR to induce IFN ( Fig 7E ) , as well as impairing responses to viral infection , such as VSV . As expected , IFNγ-induced expression of GBP1 was unaffected by the knockdown of either MDA5 or MAVS ( Fig 7C and 7F ) . That artificial activation of PKR was capable of inducing IFNβ in the absence of MDA5 , but remained dependent on MAVS led us to conclude that it most likely acts during infection to transmit signals between these two elements of the pathway . While PKR may be capable of functioning independently from MDA5 when activated in an RNA-independent manner ( Fig 7B , 8 h ) , its ability to signal to MAVS during viral infection is augmented in an MDA5-dependent manner , suggesting that PKR acts ‘downstream’ of MDA5 and ‘upstream’ of MAVS . Since the catalytic activity of PKR is required for its ability to augment IFNβ induction in VVΔE3L infected cells ( Fig 3 ) , we assayed this function in response to infection to determine its dependence on MDA5 . PKR was immunoprecipitated from VVΔE3L-infected MEFs in the presence or absence of MDA5 , and the immunoprecipitates were assayed for catalytic activity following incubation with 32P-ATP in vitro ( Fig 8A ) . Very little activity was detected in samples from uninfected cells ( lanes 1 and 3 ) , while kinase activity was significantly enhanced following infection of WT cells ( lane 2 ) . Importantly , little increase in activity was observed for PKR isolated from Mda5-/- cells following infection ( lane 4 , upper panel ) , in spite of equal quantities of total protein being recovered from all cells ( lower panel ) . Quantification of multiple independent experiments demonstrated consistently and significantly reduced induction of PKR catalytic activity in Mda5-/- relative to WT cells following infection . To determine if impairment of PKR activation was an intrinsic property of MDA5-null cells or was restricted to viruses capable of activating MDA5 , we infected WT and Mda5-/- cells with the RIG-I activating virus , vesicular stomatitis virus ( VSV ) . VSV infection led to robust activation of PKR autophosphorylation regardless of MDA5 expression ( Fig 8B , lanes 3 and 4 ) , distinct from the impaired activation of PKR observed in VVΔE3L-infected Mda5-/- cells ( lanes 1 and 2 ) . Taken together , these results indicate that not only is PKR required for the MDA5-dependent activation of IRF3 and induction of IFNβ expression , but activation of its catalytic function is dependent on MDA5 during VVΔE3L , but not VSV , infection .
PKR is a key component of the antiviral response to a wide variety of viruses [69 , 70] . This antiviral effect has been largely attributed to its well-characterized ability to phosphorylate and , thereby , inactivate eIF2α , leading to inhibition of translation of both viral and cellular RNA . As a consequence of eIF2α phosphorylation and the subsequent inhibition of translational initiation , PKR activation also leads to the formation of stress granules that appear to have antiviral function [71] . Early studies indicated that PKR may also play a direct role in IFN induction in response to pIC , a dsRNA mimetic [36 , 49]; however , a number of more recent genetic and biochemical studies have suggested that PKR can be superfluous for IFN induction , at least in response to some viruses , throwing into question whether PKR acts as a sensor or an effector in the antiviral response . For instance , IFN induction by pIC or Newcastle disease virus ( NDV ) infection was intact in PKR-null mice in vivo , and an apparent in vitro deficiency could be reversed by IFN priming , suggesting that PKR-independent pathways mediated IFN induction [72] . Discovery of cytoplasmic dsRNA sensors , RIG-I , MDA5 , and LGP2 and their associated signaling pathways , has provided a framework for understanding the primary pathways for innate immune responses to RNA viruses , without a documented role for PKR [73] . Nonetheless , experimental data have suggested that PKR may be a component of IFN production in response to some viruses , including West Nile virus [37] , Theiler's murine encephalitis virus ( TMEV ) , EMCV [38] , measles virus [74] and Semliki Forest virus ( SFV ) [39] . Here , we provide evidence that PKR is involved in MDA5-mediated responses to virus infection and is necessary for IFN induction . Mechanistically , we found that MDA5 and PKR proteins associate in a complex , and that MDA5 augments virus-induced activation of PKR . These results suggest that that PKR participates as part of the viral sensor machinery in concert with MDA5 to recognize foreign RNA and induce IFN . We reexamined the role of PKR in IFN induction by looking specifically at responses to VVΔE3L infection and other MDA5-mediated responses [42] . Consistent with earlier studies , the absence of PKR severely impaired IFN induction following pIC transfection [36 , 49] . Additionally , the absence of PKR resulted in a significant inhibition of VVΔE3L-induced IFN production , a phenotype that strongly resembled that seen in MDA5-null cells [42] . This impaired IFN phenotype was not seen in response to FluΔNS1 or SeV , strong activators of RIG-I signaling [5] . In fact , IFNβ levels were increased following FluΔNS1 infection of PKR-null cells . This result likely reflects restoration of a protein synthesis-dependent positive feedback loop that would be impaired following PKR-mediated host translational shutoff [48 , 75] . Interestingly , eIF2α S51A mutant cells , which also failed to arrest translation due to PKR activation , demonstrated a similar enhancement in FluΔNS1-mediated IFN production . However , increased IFN levels were also observed during FluΔNS1 infection in the presence of CHX , which also inhibits translation . However , it is likely that increased IFN following CHX treatment likely reflects stabilization of short-lived mRNA reported to occur under these conditions [76] . Furthermore , while loss of PKR abrogated MDA5-dependent IFN induction , mutation of eIF2α did not , suggesting that the requirement for PKR during IFN induction is not dependent on its ability to inhibit protein synthesis . One of the antiviral functions of PKR is its role in formation of stress granules , which are triggered by translational stalling that occurs following eIF2α phosphorylation [77] . These structures , which have been associated with both proviral and antiviral phenotypes , form in the cytoplasm of cells following infection with a variety of different viruses , including VVΔE3L [78] . Formation of these granules is dependent on PKR and involves cytoplasmic helicases , including RIG-I , MDA5 , and DHX36 [56 , 58 , 79 , 80] . However , it is unlikely that the stress granule-inducing function of PKR explains its cooperation with MDA5 during activation of IRF3 and IFNβ induction . Consistent with previous reports , we observed that stress granule formation was dependent on the phosphorylation of eIF2α , probably due to the subsequent translational arrest . In our studies , stress granules formed following VVΔE3L infection in WT cells , but not in eIF2α-S51A cells , although IFN was induced in both genetic backgrounds . These data show that phosphorylation of this PKR substrate , translation inhibition , and subsequent stress granule formation were not prerequisites for IFNβ induction ( Fig 3 ) . Although PKR has been implicated in IFN induction , the mechanism behind its involvement is not fully understood . Recently , it was reported that PKR enhances IFNβ production in response to a mutant form of measles virus [36 , 74] . The requirement for PKR was linked to its role in the activation of NFκB [74] and was ascribed to translational shutoff [59] , since the expression of mutant eIF2α S51A also resulted in inhibition of IFN induced by the mutant virus . However , results from experiments with other viruses demonstrated that PKR influences IFN production in response to EMCV and SFV independently of eIF2α [38] , consistent with our data . Our results indicate that , like EMCV and SFV , PKR-dependent IFN induction in response to VVΔE3L is eIF2α phosphorylation-independent . These distinct roles observed for PKR likely reflect the differential requirements for its enzymatic activity that depend on the viral stimulus . Although MDA5 may play a moderate role in response to measles virus infection , RIG-I , rather than MDA5 , is the predominant sensor during infection for this virus [81] . The role for PKR documented here appears to be exclusive for MDA5-activating viruses , and may be circumvented by redundant signaling following activation of RIG-I . To address potential mechanisms of PKR action , we noted that the presence of PKR influences the activated state of IRF3 in response to VVΔE3L and EMCV . Consistent with the hypothesis that PKR is required only for MDA5 signaling , the absence of PKR did not impact the ability of SeV , a strong activator of RIG-I , to activate IRF3 ( Fig 4 ) . Interestingly , the ability of PKR to influence IRF3 activation observed here is distinct from a recent study that reported a PKR-dependent decrease in EMCV-induced IFN protein expression that correlated with a change in polyadenylated ( poly ( A ) ) , but not total IFN mRNA levels [38] . These investigators ascribed the decrease in IFN protein levels to enhanced shortening of IFN mRNA poly ( A ) tails in the absence of PKR , resulting in impaired translation . However , the defect in IRF3 translocation documented here suggests that , in this alternative pathway , PKR is acting upstream of IFN transcription . Furthermore , we observed a loss of IFNβ mRNA transcript levels in the absence of PKR , regardless of poly ( A ) length ( S6 Fig ) . Thus , the role of PKR in IFN production may be complex and involve multiple points of influence . Support for this notion comes from documented responses to SFV infection . Both IFNβ mRNA and protein levels are impaired in response to SFV infection in the absence of PKR . However , the impairment is more pronounced for IFNβ protein levels [39] , suggesting that PKR exerts both a transcriptional and post-transcriptional influence on IFN production . Consistent with our studies , Zhang and Samuel demonstrated that PKR , MAVS , and MDA5 are important for VVΔE3L-induced IRF3 activation [82] , and reported a direct interaction between PKR and MAVS [83] . To address how PKR may influence the MDA5-MAVS-IRF3 pathway , we found that PKR physically interacted in a complex with MDA5 ( Fig 5A , 5B and 5C ) . This association appears to occur independently of VVΔE3L infection , as both ectopic and endogenous MDA5 co-purified with PKR under mock as well as infected conditions . We also demonstrate in Fig 5B that the association between MDA5 and PKR is not bridged by RNA , since pre-treatment of extracts with nuclease had no effect on the interaction . It is , therefore , possible that under basal settings , PKR and MDA5 exist as a complex that must be further activated in response to virus infection . Whether the presence of viral RNA is detected by MDA5 or PKR , this activation event presumably requires PKR’s catalytic function , since PKR autophosphorylation was diminished in the absence of MDA5 . Interestingly , an interaction between PKR and RIG-I was also detected , but only in response to VVΔE3L infection ( Fig 5A ) . This result is consistent with a recently reported interaction between PKR and RIG-I following FluΔNS1 infection [79] , and highlights a distinct mechanism of action between MDA5 and RIG-I in regards to their association with PKR . Taken together , it is tempting to speculate that interactions of PKR with RLH proteins may depend on activation of the cytoplasmic sensors . For instance , engagement of MDA5 with VVΔE3L during infection could lead to PKR-MDA5 interaction , while activation of RIG-I during FluΔNS1 infection could lead instead to PKR-RIG-I interaction . However , at least in our experiments , we saw no requirement for PKR activation during RIG-I signaling , suggesting that its involvement downstream of RIG-I may be redundant with other signals . We further investigated PKR function by ectopically activating it as a GyrB . PKR fusion protein . Coumermycin-mediated activation of PKR was sufficient for IFNβ induction , accompanied by the activation and phosphorylation of IRF3 . Highlighting the requirement for its catalytic function , this effect was no longer observed when an enzymatically inactive , GyrB . PKR ( K296H ) mutant was expressed . Moreover , direct activation of PKR with this approach displayed MAVS- , but not MDA5-dependency , suggesting that the requirement for PKR during VVΔE3L infection lies downstream of MDA5 , but upstream of MAVS . However , the activation of endogenous PKR during VVΔE3L infection remained strictly dependent on MDA5 . Stimulation of PKR catalytic activity was impaired in MDA5-deficient cells , suggesting that MDA5 augments the ability of PKR to be activated by dsRNA . Structural studies of PKR have demonstrated that it is activated through conformational changes and dimerization mediated by interactions between its amino-terminal dsRNA binding domains and pathogen-derived dsRNA species [84 , 85] . Sufficient length requirements are necessary in order for these dsRNA species to induce PKR dimerization and activation [86] . However , recent studies have indicated that a variety of RNA structures can be recognized by PKR , allowing it to be activated by a diverse set of pathogenic signals [87 , 88] . It is possible that MDA5 , which also recognizes pathogen-derived dsRNA [89] , augments the loading and/or activation of PKR by dsRNA , possibly by acting to remove RNA-binding proteins that would otherwise interfere with PKR activation [90] or facilitating the formation of RNA structures with sufficient stimulatory characteristics [88] . In this regard , it is interesting that it is the helicase domain of MDA5 that can augment PKR activation by stripping RNA-binding proteins from IFN-stimulatory RNA , becoming phosphorylated by PKR in the process [90] . This may also be the same domain that cooperates with PKR during VVΔE3L infection , since the isolated MDA5 CARD domains were able to act in a PKR-independent fashion ( Fig 5F ) . In conclusion , our data indicate that PKR is required for robust MDA5-dependent responses , and that MDA5 plays a role in PKR activation during viral infection . It remains to be determined how MDA5 facilitates PKR catalytic activity and how activated PKR augments the MAVS-dependent stimulation of IRF3 and IFNβ transcription . In particular , the direct PKR substrate responsible for signaling has not been identified . MDA5 is one likely candidate , since it has been shown to be a PKR substrate [90] . Similarly , PKR itself may be a relevant substrate , potentially stabilizing its interaction with MDA5 and/or MAVS . Indeed , PKR becomes multiply autophosphorylated following activation , although the function of the majority of these phosphorylation sites remains to be defined [91] . An attractive model , consistent with the studies reported here , would posit that MDA5 , through its RNA helicase activity , facilitates transfer of viral dsRNA to PKR , enhancing its overall catalytic activity in infected cells . Similarly , PKR activity , either through phosphorylation of MDA5 or additional factors , contributes to MDA5 filament formation , a process that is structurally distinct from activated RIG-I and required for downstream signaling [90] . Altogether , the studies reported here describe a critical catalytic function for PKR that responds to viral infection in an MDA5-dependent manner to govern the IFN response . Further studies will be needed to more fully understand how cellular signaling pathways are wired to tailor the innate response to different classes of viral pathogens .
A549 , HEK293T , HT1080-GyrB-PKR , HT1080-GyrB-PKR-K296H , and MEF cells were maintained in Dulbecco's modified Eagle's Medium ( DMEM; Cellgro ) complemented with 10% calf serum . A549 and HEK293T cells stably expressing short-hairpin RNA against either PKR ( shPKR ) or a nonspecific sequence ( shNS ) were maintained in the above media containing 5 ug/mL and 3 ug/mL of puromycin , respectively . MDA5 and RIG-I deficient and control MEFs were kindly provided by Shizuo Akira ( Osaka University ) . IRF3 deficient MEFs were kindly provided from Tadasugu Taniguchi ( University of Tokyo ) . HT1080-GyrB-PKR and HT1080-GyrB-PKR-K296H cells were a generous gift from Antonis Koromilas ( McGill University ) , and eIF2α-S51A cells were a gift from Randall Kaufman ( University of Michigan ) . MEFs have been isolated from two distinct strains of PKR-null mice . Cells used to conduct the majority of studies reported here were kindly provided by Bryan Williams ( Monash University ) , and express a truncated form of the protein lacking the RNA binding domain [92] . Similar results showing the requirement for PKR for IFN production in response to VVΔE3L were also obtained with PKR-null MEFs kindly provided by John Bell , which express a truncated a form of the protein lacking the catalytic domain [93] . A549 and HEK293T cells stably expressing short-hairpin RNA were generated by MLP-shPKR and MLP-shNS retroviruses derived by co-transfection with EcoPac [94] and VSV-G ( vesticular stomatitis virus glycoprotein ) into HEK293T cells using the calcium phosphate precipitation method . Supernatant was harvested 36–48 h post transfection , clarified , and used for three sequential infection cycles of either A549 or HEK293T cells followed by selection with 5 ug/ml or 3ug/mL of puromycin , respectively . pLKO . 1-puro pseudotyped lentiviruses expressing hairpins against MDA5 and MAVS were packaged in HEK293T cells . Viral supernatants were used to transduce HT1080-GyrB-PKR cells followed by selection for hairpin expression with puromycin . pLVX-mCherry VSV-G pseudotyped lentiviruses expressing RIG-I-CARD or MDA5 were packaged by co-transfection of HEK293T cells with psPAX2 and pMD2 . G , gifts from Didier Trono . Viral supernatant was harvested at 48 and 72 h post transfection and concentrated by centrifugation at 25 , 000 rpm for 90 minutes . A549 shPKR and A549 shNS cells were subjected to three rounds of infection with concentrated virus and the resulting cells were assayed for IFNβ production by real time PCR and mCherry fluorescence by microscopy after 48 h . The PKR retroviral short hairpin RNA ( shRNA ) vector contained a previously described hairpin sequence [95] within a mir-30 backbone ( 5'-TGCTGTTGACAGTGAGCGAgcagggagtagtacttaaataATAGTGAAGCCACAGATGTATtatttaagtactactccctgcCTGCCTACTGCCTCGGA-3' ) , which was cloned into the MSCV-based retroviral vector MLP . pCAGGs constructs expressing full-length RIG-I and MDA5 , kind gifts from Christopher Basler ( Icahn School of Medicine at Mount Sinai ) , were cloned from their respective vectors into the lentiviral vector pLVX-mCherry . Constitutively active RIG-I ( pEF-BOS-RIG-I-MIII ) was the kind gift of Curt Horvath ( Northwestern University ) [65] . pLKO . 1-puro constructs expressing shRNA against MDA5 ( CCGGCCAACAAAGAAGCAGTGTATACTCGAGTATACACTGCTTCTTTGTTGGTTTTTG ) and MAVS ( CCGGCAAGTTGCCAACTAGCTCAAACTCGAGTTTGAGCTAGTTGGCAACTTGTTTTTTG; ) were provided by the NYU School of Medicine shRNA Core Facility [96] . The Flag-tagged STAT2 construct , pLpC-Flag-Stat2 was previously described [64] . An expression construct for MDA5-CARD was generated by cloning the first 267 amino acids of human MDA5 into pLpC Vaccinia Virus ΔE3L ( VVΔE3L ) derived from the Western Reserve strain was a kind gift from Bertram Jacobs ( Arizona State University ) . Sendai Virus ( SeV ) , Cantell strain was purchased from Charles River . The VSV strain ( VSV-GFP M51R ) used in these studies was a gift from Benjamin tenOever ( Icahn School of Medicine at Mount Sinai ) , and contains a mutation in the matrix viral protein ( M51R ) that renders the virus incapable of inhibiting IFN production [97] . This mutant was used as a positive control for a RIG-I ligand , and as a robust stimulator of type I IFN production . Influenza A ΔNS1 ( Flu ΔNS1 ) was a gift from Adolfo Garcia-Sastre ( Icahn School of Medicine at Mount Sinai ) , and encephalomyocarditis virus ( EMCV ) was a gift from Jan Vilcek ( NYU School of Medicine ) . Unless otherwise indicated , cells were infected at a multiplicity of infection ( MOI ) of approximately 3 , with the exception of FluΔNS1 , which was used at an MOI of 1 . SeV was used at 200 HA units/ml . For all infections , virus was diluted in serum-free DMEM and incubated with target cells at 37°C for 1 h , followed by replacement with standard growth medium . Unless otherwise indicated , virus infected cells were assayed at 8 h post infection . PKR inhibitor ( PKR-I; Calbiochem ) was added to cells at the indicated concentrations and incubated for 1 h . Following incubation , the inhibitor was removed and cells were infected as indicated . Cycloheximide ( CHX ) was added to cells at 50 ug/ml for the final 4 h of viral infection . Cells were treated with either 100 ng/ml or 200 ng/mL of coumermycin ( Sigma ) for 8 or 24 h , or with 5 ng/ml IFNγ ( Amgen Biologicals ) for 8 h . For polyriboinosinic:polyribocytidylic acid ( pIC ) transfections , cells were transfected with 2 ug pIC ( Invivogen ) using Lipofectamine 2000 ( Invitrogen ) and incubated for 4 h at 37°C before RNA extraction . HEK293T cells were seeded onto 12-well dishes and transfected using the calcium phosphate method with 1 ug of firefly luciferase plasmid fused to human IFNβ promoter , 1 . 4 ug of plasmids expressing β-galactosidase or renilla luciferase , and 4 ug of plasmids expressing RIG-I-CARD ( pLVX-RIG-I-CARD or pCDNA3- RIG-I-CARD ) , MDA5-CARD ( pLpC- MDA5-CARD ) , RIG-I-MIII ( pEF-Bos-Flag-RIG-I-MIII ) , or MDA5 ( pLVX-MDA5 ) . Cells were harvested 48 h post transfection and assayed for firefly luciferase and β-galactosidase or renilla luciferase activity . Firefly luciferase levels were normalized to β-galactosidase or renilla luciferase activity and experiments were performed in triplicate . RNA isolated with TriZol ( Invitrogen ) was used to generate cDNA with Moloney murine leukemia virus ( M-MLV ) reverse transcriptase ( Invitrogen ) , according to the manufacturer's instructions . While oligo dT was primarily used to prime cDNA transcripts for qRT-PCR , random hexamers were used for S6 Fig to demonstrate the loss of IFNβ in infected Pkr-/- MEFs was not due to poly ( A ) trimming . Transcripts were then quantified by qRT-PCR with SYBR green ( Molecular Probes ) using the following primers: human GAPDH: F 5' TGGAAGGACTCATGACCACA 3' and R 5' TTCAGCTCAGGGATGACCTT 3'; human IFNβ: F 5' GTCTCCTCCAAATTGCTCTC and R 5' ACAGGAGCTTCTGACACTGA 3'; human GBP1: F 5' TTCTTCCAGATGACCAGCAG 3' and R 5' GCTAGGGTGGTTGTCCTTGA 3'; mouse GAPDH: F 5' TGAGGACCAGGTTGTCTCCT 3' and R 5' CCCTGTTGCTGTAGCCGTAT 3'; mouse IFNβ: F 5' CCCTATGGAGATGACGGA 3' and R 5' CTGTCTGCTGGTGGAGTTC 3'; human TNFα: F 5’ CCTGACATCTGGAATCTGGAGACC 3’ and R 5’ CTGGAAACATCTGGAGAGAGGAAGG 3’; human IL1β: F 5’ GGGCCTCAAGGAAAAGAATC 3’ and R 5’ TTCTGCTTGAGAGGTGCTGA 3’; EMCV 3D: F 5’ GACGCTTGAAGACGTTGTCTTCTTA 3’ and R 5’ CCCTACCTCACGGAATGGGGCAAAG 3’; VV HA: 5’ CATCATCTGGAATTGTCACTACTAAA 3’ and R 5’ ACGGCCGACAATATAATTAATGC 3’; SeV NP: 5’ TGCCTGGAAGATGAGTTAG 3’ and 5’ GCCTGTTGGTTTGTGGTAAG 3’ . Relative expression was determined by comparison to a standard curve generated from serial dilutions of cDNA containing abundant target sequences and normalized to the expression of GAPDH . Data were represented as the IFNβ/GAPDH ratio and the mean ± standard deviation between triplicate samples of a representative experiment were shown . Significance was determined by an unpaired t-test . Data shown are representative of at least three independent experiments . Whole cell extracts were lysed in 1% NP-40 buffer ( 50 mM Tris , pH 7 . 5 , 150 mM NaCl , 30 mM NaF , 5 mM EDTA , 10% glycerol , 1% NP-40 , 1 mM PMSF , 1 mM Sodium Orthovanadate , protease inhibitor ) for 20 min on ice before centrifugation at 4°C for 10 min . Protein concentrations were quantified by Bradford assay . For immunoprecipitations ( IP ) in Fig 5A , HEK293T cells were transfected with FLAG-tagged RIG-I , MDA5 , E3L or STAT2 using the calcium phosphate precipitation method ( Fisher Scientific ) . Thirty-six hours post-transfection , cells were infected with VVΔE3L at an MOI of 10 for 8 h , followed by lysis in 1% NP-40 buffer . IPs were performed with 800ug of total protein and incubated with anti-FLAG M2 affinity gel ( Sigma-Aldrich ) overnight at 4°C . For Fig 5B , lysates from transfected HEK293Ts were treated with 100 units/mL of micrococcal nuclease ( Affymetrix USB ) for 30 min at 37°C prior to incubation with incubation with beads . Beads were washed four times with RIPA buffer ( 50 mM Tris pH 7 . 8 , 150 mM NaCl , 0 . 1% SDS , 0 . 5% sodium deoxycholate , 1% Triton-X-100 ) supplemented with protease inhibitor cocktail ( Sigma-Aldrich ) , 1 mM PMSF , 100 mM sodium fluoride , 2 mM sodium pyrophosphate and 2 mM sodium orthovanadate ) , resuspended in Laemmli buffer , and boiled at 95°C for 5 min . Additionally , 5% of total protein lysate was included as input control . For IPs in Fig 5C , A549 cells were first treated with 150U/ml of human IFNα 2a for 14 hours to induce detectable levels of MDA5 and lysed in 1% NP-40 buffer . Protein A agarose beads ( Thermo Scientific ) were blocked with 5% BSA+ PBS prior to incubation with 2 mg of lysates overnight at 4°C . 5% input and IP reactions were resolved by 8% SDS-PAGE . For isolation of cytoplasmic and nuclear fractions , cells were initially lysed in RSB ( 10 mM Tris pH . 7 . 4 , 10 mM NaCl , 3 mM MgCl2 , 0 . 5 mM DTT , 1 mM sodium orthovanadate , 1 mM PMSF , and protease inhibitor cocktail ( Biotool ) for 20 min on ice with occasional vortexing . Following incubation , cells were centrifuged at 7 , 500 rpm for 5 min at 4°C . Supernatant was retained as the cytoplasmic fraction . The remaining nuclear pellet was washed with 1xPBS and lysed in Extraction Buffer ( 20 mM HEPES pH . 7 . 9 , 420 mM NaCl , 1 . 5 mM MgCl2 , 0 . 2 mM EDTA , 25% glycerol , 0 . 5 mM DTT , 1 mM sodium orthovanadate , 1 mM PMSF , and protease inhibitor cocktail ( Biotool ) for 30 min on ice with occasional vortexing . Cells were centrifuged at 14 , 000 rpm for 20 min at at 4°C . Supernatant was retained as nuclear fraction . Samples were resolved by 10% SDS-PAGE , transferred to PVDF membranes ( Millipore ) and blocked for 1 h in 5% non-fat milk + TBS ( 20 mM Tris-HCl pH 7 . 5 , 150 mM NaCl ) . Membranes were probed with primary antibodies diluted in 5% BSA+TBS-0 . 1% Tween-20 buffer overnight at 4°C: anti-PKR ( Santa Cruz; SC-707 for human , Cell Signaling; D7F7 for human , 1:1000 , SC-1702 for murine , 1:1000 ) , anti-Flag M2 ( Sigma-Aldrich , 1:1000 ) , anti-eIF2α-pSer51 ( Cell Signaling , 1:1000 ) , anti-eIF2α ( Cell Signaling , 1:1000 ) , anti-phosphoIRF3 ( Cell Signaling , 1:1000 ) , anti-IRF3 ( Zymed , 1:2000 ) , anti-Histone 3 ( Abcam , 1:1000 ) , anti-human IRF3 ( Pharmingen , 1:1000 ) , anti-p38α MAP kinase ( Cell Signaling 5F11 , 1:2000 ) , anti-MDA5 ( Cell Signaling , D74E4 , 1:1000 ) and anti-Actin ( Chemicon International , 1:1000 ) . Antibodies against viral products VV-I3L and VSV-G were kind gifts from Ian Mohr and Benjamin tenOever , respectively . Membranes were probed with HRP-linked goat anti-mouse IgG ( KPL; 1:5000 ) or goat anti-rabbit IgG ( Thermo Scientific; 1:5000 ) and developed with enhanced chemiluminescence ( ECL; Millipore ) . WT eIF2α and mutant eIF2α S51A MEFs were seeded onto 8-well chamber slides ( Millipore , EZ Slides ) and infected with VVΔE3L at MOI = 1 . At 8 hpi , cells were fixed with 4% Formaldehyde for 1 hr at room temperature ( RT ) , followed by two washes with PBS , and permeabilization with cold methanol for 5 min . Cells were rinsed twice in PBS before incubation with Blocking buffer ( 5 mg/mL BSA and 0 . 04% Tween 20 in PBS ) for 30 min at RT . Cells were overnight at 4°C incubated in primary anti-G3BP1 ( BD Biosciences , #611126 ) , prepared in Blocking buffer at 3 ug/mL . After three washes in PBS , cells were incubated in anti-mouse secondary ( Life Sciences , AlexaFluor 588 ) at 1:200 in Blocking buffer for 1hr at RT . This was followed by three additional washes in PBS and incubation with DAPI at 1:1000 in PBS for 1min . Finally , cells were washed five times in PBS before mounting ( Vectashield ) . G3BP1 and DAPI expression was visualized using the Nikon DS-Qi1 inverted fluorescent microscope . PKR catalytic activity was assayed as previously described [98] . In brief , infected cells were lysed in IP lysis buffer ( 50 mM Tri-HCl pH 7 . 6 , 150 mM NaCl , 10% glycerol , 1% NP-40 , 5 mM EDTA , 1 mM DTT , 100 mM NaF , 2 mM sodium pyrophosphate , 2 mM sodium orthovanadate , 1 mM PMSF ) . Equal amounts of protein were incubated with anti-PKR antibody ( Santa Cruz ) for 30 min on ice and then incubated with protein A agarose ( Thermo Scientific ) overnight at 4°C . The beads were washed twice in IP lysis buffer and twice in DBGA ( 10 mM Tris-HCl pH 7 . 6 , 50 mM KCl , 2 mM MgOAc , 20% glycerol , 7 mM beta-mercaptoethanol ) . The beads were resuspended in 50 μl DBGB ( DBGA containing 2 . 5 μM MnCl2 ) followed by the addition of 5 μl ATP mix [10 μCi [γ-32P]ATP ( 6000Ci/mmol ) , 0 . 5 μl 100 μM ATP , 3 . 5 μl DBGA] . The mixture was incubated at 30°C for 30 min , and the beads were resuspended in 1X Laemmli buffer and boiled . Samples were resolved by 10% SDS-PAGE and visualized by phosphorimager analysis .
|
Induction of type I interferon ( IFN ) during viral infection is a key step in the activation of innate host defense . Critical to this response is the ability to recognize viral nucleic acids in the host cell cytoplasm . Detection of viral RNA is mediated by RNA helicases , including RIG-I and MDA5 , which sense 5’-phosphorylated RNA and long double-stranded viral RNA , respectively . The importance of the IFN response is underscored by the variety of mechanisms through which viruses circumvent signaling by inactivating components of this pathway , including targeting the cellular kinase , protein kinase R ( PKR ) , to evade host defenses . Activation of PKR by virus infection is known to result in an overall block of host translational machinery . Here , we showed that a function of PKR , independent of translation inhibition , was critical for IFN production downstream of MDA5 , but not RIG-I . In addition , MDA5 was required for the stimulation of PKR catalytic activity that occurred in response to infection by an MDA5-restricted virus , but not in response to a RIG-I-dependent virus . These findings identified a previously uncharacterized role for PKR catalytic function that cooperates with MDA5 signaling and highlights an unexpected role for MDA5 in stimulating the enzyme activity of PKR .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"luciferase",
"poxviruses",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"enzymes",
"pathogens",
"messenger",
"rna",
"microbiology",
"enzymology",
"viruses",
"rna",
"viruses",
"immunoprecipitation",
"dna",
"viruses",
"research",
"and",
"analysis",
"methods",
"sendai",
"virus",
"paramyxoviruses",
"proteins",
"medical",
"microbiology",
"oxidoreductases",
"microbial",
"pathogens",
"vaccinia",
"virus",
"precipitation",
"techniques",
"biochemistry",
"rna",
"post-translational",
"modification",
"nucleic",
"acids",
"virology",
"viral",
"pathogens",
"interferons",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2016
|
PKR Transduces MDA5-Dependent Signals for Type I IFN Induction
|
Statistical coupling analysis ( SCA ) is a method for analyzing multiple sequence alignments that was used to identify groups of coevolving residues termed “sectors” . The method applies spectral analysis to a matrix obtained by combining correlation information with sequence conservation . It has been asserted that the protein sectors identified by SCA are functionally significant , with different sectors controlling different biochemical properties of the protein . Here we reconsider the available experimental data and note that it involves almost exclusively proteins with a single sector . We show that in this case sequence conservation is the dominating factor in SCA , and can alone be used to make statistically equivalent functional predictions . Therefore , we suggest shifting the experimental focus to proteins for which SCA identifies several sectors . Correlations in protein alignments , which have been shown to be informative in a number of independent studies , would then be less dominated by sequence conservation .
A fundamental question in biology is the relation between the amino acid sequence of a protein and its function and three-dimensional structure . Given the rapid growth in the sequence data available from many organisms , it has become possible to use statistical sequence analysis to approach this question . Based on sequence similarity , protein sequences can be grouped into families thought to share common ancestry; the proteins in such a family typically perform related functions and fold into similar structures [1 , 2] . It has been shown in many studies that a statistical analysis of a multiple sequence alignment ( MSA ) corresponding to a given protein family can be used to find amino acids that control different aspects of a protein’s function or structure . A basic statistical quantity that can be calculated for a multiple sequence alignment is the distribution of amino acids at each site . In particular , the level of sequence conservation at each site is of biological relevance , since it is expected that conservation is low in the absence of selective pressures . For this reason , conservation has long been used to predict which parts of a protein are most likely to be functionally significant [3–7] . More recently , the availability of large sets of protein sequences has made it possible to also estimate higher-order statistics , such as the correlations between the amino acids found at each pair of sequence positions . In a number of examples , these statistics have been shown to contain information about the structure and function of proteins [8–12] . One way in which pairwise correlations might arise is for a deleterious mutation at a given position to be compensated by a mutation at a different position . This can yield a scenario in which the two individual mutations are relatively rare , but the combination of both is common in natural proteins . Statistical coupling analysis ( SCA ) was introduced by Lockless and Ranganathan in 1999 as a way to infer energetic interactions within a protein from a statistical analysis of a multiple sequence alignment [13] . The authors compared the statistics of an alignment of PDZ domain sequences to measurements of the binding affinity between a particular member of the alignment ( PSD95pdz3 ) and its cognate ligand . The statistical analysis assumed that the frequencies of mutations obey a Boltzmann distribution as a function of binding free energy , allowing estimation of the binding affinity by ΔGi ∼ log fi , where fi is the frequency of an amino acid type at a given site in the alignment . By conditioning on amino acid type at a second site , they calculated the amount by which the effect of a mutation at one site changed depending on the amino acid present at the second site: ΔΔGi∣j = ΔGi∣j − ΔGi ≡ ΔΔGstat . This gave an estimate for the effective coupling between the sites . The assumptions behind the original formulation of SCA are likely to be violated , since the selective pressures acting on a protein are more complex than simply maximizing binding to a ligand . Despite this , the method seemed to be effective . In the original paper [13] , mutant cycle analysis was used to measure ΔΔGbinding , the amount by which the effect of a given mutation on ligand binding affinity of PSD95pdz3 changes when the mutation occurs on a background containing a second mutation . This can be written as ΔΔGbinding = ΔGi∣j − ΔGi , where now ΔG represents a change in the physical free energy as opposed to a statistical construct . The quantity ΔΔGbinding was observed to be well correlated with the statistically-calculated ΔΔGstat . The set of residues identified by SCA to be coupled with a particular site known to be important for binding specificity of the PDZ domain was found to physically connect distal functional sites of the protein [13] . This led to the suggestion that these residues may mediate an allosteric response . Experimental evidence later showed that indeed some of the residues identified by SCA participate in allostery [14–17] . Moreover , in a different study , a large fraction of the artificial WW domains built by conserving the pattern of statistical couplings calculated by SCA were observed to be functional , while sequences built to conserve single-site statistics alone were not [8 , 9] . Motivated by these observations , Halabi et al . reformulated SCA in purely statistical terms , avoiding the assumptions related to energetics [18] . The reformulation amounted to a particular way of combining correlations with conservation . The basic idea was to multiply each element of the covariance matrix Cij by a product ϕiϕj , yielding the “SCA matrix” C ˜ ij = ϕ i ϕ j C ij . The “positional weights” ϕi were a function of the frequency fi of the most prevalent amino acid at each position , and were roughly given by ϕi ∼ log[fi/ ( 1 − fi ) ] . This particular form was chosen to reproduce the results from the original formulation of SCA [18 , 19] . In subsequent work regarding SCA , several variations on this basic idea were used; all of these yield similar though not identical results and are described more precisely in Methods and S1 Text . Running either the original or the reformulated analysis on several examples [8–9 , 13 , 16–18 , 20] , it was noticed that the resulting SCA matrix had an approximate block structure . In analogy to previous work in finance , Halabi et al . analyzed this structure by looking at the top eigenvectors of the SCA matrix [18 , 21] . The corresponding groups of residues were called “protein sectors” because similar clusters observed in the correlations of stock prices were found to correspond to financial sectors . Experiments found that mutating residues in distinct sectors specifically affected different phenotypes of the protein [18] , leading to the suggestion that each SCA sector might comprise a group of amino acids that control a particular phenotype . It is important to note that there are several subtly different meanings that have been attributed to protein sectors ( see Table 1 ) . The description outlined above defines protein sectors as the results of a statistical analysis of a multiple sequence alignment . This definition depends on the statistical method employed; it would , for example , depend on the choice of positional weights in the case of SCA , or on the precise thresholds and methods used for clustering . To distinguish this from other meanings , we will call these statistical sectors ( or SCA sectors when the statistical method is SCA ) . The sectors identified by SCA have also been given an evolutionary interpretation [18 , 20 , 22] , based on the fact that they are defined as groups of residues whose mutations are correlated in an alignment , and the sequences in the alignment are likely to be evolutionarily related . However , this argument is insufficient to prove the evolutionary nature of the statistical sectors , given that their precise composition is dependent on the statistical method employed [23] . Thus it is difficult to assess to what extent the sector’s composition is actually related to the evolutionary process itself , as opposed to the choice of the statistical method . Strikingly , Halabi et al . showed that for an alignment of serine proteases , one of the sectors can be used to distinguish between vertebrates and invertebrates , suggesting that indeed an evolutionary interpretation may be appropriate [18] . However , before concluding that in general SCA sectors have an evolutionary interpretation , it would be important to extend these studies to different alignments . An alternative , more direct , approach would be to perform artificial evolution experiments to check whether the SCA sectors maintain their integrity under strong selection , or whether new sectors can be created in this way . In addition , such experiments would provide data on the evolutionary dynamics of proteins , and thus help to define more precisely the notion of evolutionary sectors . Another surprising property of the groups of residues identified by SCA is that they usually form contiguous structures in the folded protein , although they are not contiguous in sequence [18 , 20 , 22 , 24 , 25] . This suggests the notion of structural sectors , groups of residues having different physical properties compared to their surroundings . An experimental test for such inhomogeneities inside proteins could employ NMR spectroscopy to follow the dynamics of specific atoms while the protein is undergoing conformational change [14 , 26] . In addition , analyzing room-temperature X-ray diffraction data could shed light on residues with coupled mobility or increased fluctuations in an ensemble of structures [27 , 28] ( Doeke Hekstra , personal communication ) . Alternatively , this kind of experiments could be done in silico using for example molecular dynamics simulations to identify correlated motions in the protein [29] ( Olivier Rivoire , personal communication ) . Finally , as mentioned above , a number of mutational studies have suggested yet another interpretation of the sectors identified by SCA as functional sectors , groups of amino acids that cooperate to control certain phenotypic traits of a protein , such as binding affinity [13 , 18 , 20 , 25] , denaturation temperature [18] , or allosteric behavior [14–17 , 24] . It is this aspect of the sectors that has been most emphasized in the literature . In the language we just introduced , we can say that there is some data suggesting that SCA can identify groups of residues that act as evolutionary , structural , and functional sectors in a protein . It is important to note that these aspects can exist independently of one another . As an example , the existence of a physical inhomogeneity overlapping the statistical sector positions would support the idea that SCA can identify structural sectors , but would provide no guarantee that these also have an associated phenotype . For this reason , independent experimental verification is needed to support each of these claims . We focus here on the experimental evidence supporting the hypothesis that SCA sectors act as functional sectors of proteins [8–9 , 18 , 20 , 24 , 25] . We note that with the exception of Halabi et al . [18] , this data refers to proteins in which a single SCA sector was identified , and we show that in this case , within statistical uncertainties , a method based on sequence conservation can identify functional residues as well as SCA . We also give a simple mathematical argument describing why this might happen . Given that conservation information is explicitly used in calculating the SCA matrix , it is not surprising that SCA sectors are related to conservation . However , what we show here is that conservation dominates the SCA calculations in the single-sector case; thus , in order to establish whether the functional significance of SCA sectors is more than what is expected from single-site statistics alone , experiments need to focus on the examples where SCA identifies several sectors . The analysis of serine proteases described above provides such a study [18] , but it is essential to have more data for different protein families to assess the robustness and generality of these observations .
The ability of SCA to identify residues that are important for protein function was recently tested in a high-throughput experiment involving a PDZ domain [25] . Each amino acid of the PSD95pdz3 domain was mutated to all 19 alternatives and the binding affinity of the resulting mutants to the PSD95pdz3 cognate ligand was measured . The measurement involved a bacterial two-hybrid system in which the PDZ domain was fused to the DNA-binding domain of the λ-cI repressor , while the ligand was fused to the α subunit of the E . coli RNA polymerase . This was used to control expression of GFP , which allowed the binding affinity between PSD95pdz3 and its ligand to be estimated using fluorescence-activated cell sorting ( FACS ) . In order to quantify the sensitivity to mutations at a given site , the mutational effects on binding affinity were averaged over all 20 possible amino acids at that site . While mutations at most sites were found to have a negligible effect on ligand binding , 20 sites were identified where mutations had a significant deleterious effect [25] . The sector identified by SCA according to the methodology outlined above was found to indeed contain residues that are more likely to have functional significance than randomly chosen positions in the protein: 14 of the 21 sector residues are functionally significant , or 67% , compared to 25% for the entire protein ( see Fig . 2A ) . This is statistically-significant according to a Fisher exact test ( one-tailed p = 1×10−6 ) , and this result is robust to changing the threshold used to define the sector . This mirrors the results from McLaughlin Jr . et al . [25] , obtained there with a different alignment constructed using a structural alignment algorithm [13] . There is another way of assessing the functional relevance of the sector positions that avoids making a sharp distinction between functional and non-functional residues [25] . The functional effects of mutations at all the positions in the domain were used to define a background distribution showing how likely an effect of a given magnitude was . If the sector is able to identify functionally-relevant positions , then the distribution of functional effects restricted to the sector positions should differ from this background distribution . Fig . 3A shows the comparison for the PDZ experiment . A two-sample Mann-Whitney U test [32] finds that indeed sector positions have a statistically-significant distribution of functional effects compared to all residues . We now test whether we could have obtained similar results by considering only sequence conservation . Indeed , although the 21 most conserved residues are different from the 21 residues identified by SCA ( only about 60% are shared ) , the fraction of these residues that is functionally significant is the same ( see Fig . 2B ) . The histogram of functional effects is also essentially the same between SCA sector residues and conserved residues ( see Fig . 3B ) , and in fact a Mann-Whitney U test confirms that the difference is not statistically significant . McLaughlin Jr . et al . performed a similar analysis and obtained similar histograms as our Fig . 3 ( see Fig . 3a in their paper [25] ) . The reason why our results seem so different is that , due to an error , the top histogram in Fig . 3a in McLaughlin Jr . et al . is missing the data for the five sector residues that do not have a significant mutational effect . These five sector residues are mentioned and taken into account in other parts of the paper by McLaughlin Jr . et al [25] , for example in Table S6a in the supplementary information , but they do not appear in the histogram . Had they been included , the histograms for conserved residues and that for SCA sector residues would look almost identical , in agreement with our results . We stress again that these results do not imply that correlations in protein alignments are not informative . Indeed , as mentioned in the introduction , experimental data on the creation of artificial WW domains showed that ignoring correlations leads to non-functional proteins , while proteins designed based on conservation-weighted correlations can often be functional [8] . Moreover , correlation information was used to provide quite accurate predictions for contact maps and three-dimensional structures of a variety of proteins [10–12] . This is not possible using single-site statistics alone . The question we are asking , however , is whether the particular way in which alignment correlations are used in SCA is more useful for predicting functional information than conservation . The answer seems to be negative for the case of PDZ . All the observations reported above are qualitatively the same when using different alignments , including the alignment employed by McLaughlin Jr . et al . [25] and a Pfam alignment . The observations are also robust to varying the threshold used for defining the sector: in Fig . 4 we show a statistical comparison between the SCA sector and conserved residues calculated for various sizes of the sector . Note that there are some potential caveats for the statistical tests we used . One assumption of both the Mann-Whitney U test and the χ2 test employed above is that the samples analyzed are independent . In our case , the samples are the mutational effects at different residues in a protein domain , which are unlikely to be independent . Designing a statistical test that overcomes this difficulty would require a detailed model of evolutionary dynamics that accurately describes the relation between the binding affinity of PSD95pdz3 to its cognate ligand , and the evolutionary information contained in a multiple sequence alignment . To our knowledge , there is unfortunately no unambiguous way of constructing such a model . Despite these issues , the analysis presented here suggests that , for the top sector , SCA is not significantly better than conservation at predicting functionally-important sites . The case of dihydrofolate reductase ( DHFR ) [24] exhibits some interesting differences from PDZ . The experimental assay in this case involved perturbing the DHFR protein by attaching a light-sensitive domain ( LOV2 ) between the atoms of the peptide bond immediately preceding each surface residue . The experiment used a folate auxotroph mutant of E . coli whose growth was rescued by a plasmid containing DHFR and thymidylate synthetase genes . The growth rate of the bacteria , which was measured with a high-throughput sequencing method , was shown to be approximately proportional to the catalytic efficiency of DHFR . The functional effect of each insertion of the LOV2 domain was measured by the difference in growth rates between lit and dark conditions . Out of the 61 measured surface sites , 14 were found to have a significant functional effect [24] . The effects of the insertion of the LOV2 domain are not localized on a single residue of the protein , which makes the analysis of the functional significance of the SCA sector positions more complicated in the case of DHFR . We follow here the method employed in the original study by Reynolds et al . , which is to define a range around the insertion point within which a residue could conceivably feel the influence of the inserted domain [24] . More specifically , 4 Å spheres were centered on each of the four atoms forming the peptide bond broken by the insertion of LOV2 , and any residues having at least one atom centered within any of these spheres was counted as “touching” the light-sensitive residue . The exact size of the cutoff is not important: we repeated the analysis with the cutoff set to 3 Å and 5 Å and obtained the same qualitative results . Using the methodology described above , the SCA sector identified from the top eigenvector of the SCA matrix is found to “touch” all 14 of the functionally-significant LOV2 insertion sites . A set of conserved residues of the same size as the SCA sector “touches” 12 of the functionally-significant sites , and the difference is not statistically significant ( see Fig . 5 ) . The results we obtained for DHFR are somewhat less robust than those obtained for the other proteins . For the HHblits DHFR alignment , the qualitative result was the same for all sector sizes we tested ( see Fig . 6 ) , but when using the Pfam alignment , very small SCA sectors ( less than 10 residues ) “touched” many more functionally-significant sites than sets of conserved residues of the same size . It is hard to verify whether this is a chance occurrence or a real phenomenon , and it is unclear whether the notion of a sector still makes sense when it comprises such a small part of the protein . One complication arises from the fact that highly conserved residues tend to cluster closer to the core of the protein ( see Fig . 7 ) , and thus are less likely to “touch” its surface . Another dataset on which some work related to SCA has already been performed [31] was collected by Li-Smerin et al . [33] . In their experiments , 127 residues of the drk1 K+ channel were analyzed . For each of the mutants , voltage-activation curves were measured and fit to a two-state model , from which the difference in free energy between open and closed states ΔG0 was estimated . Following Lee et al . [31] , we identified a set of functional sites using the condition ∣ Δ G 0 mut − Δ G 0 wt ∣ ≥ 1 kcal/mol and we compared this set to the SCA sector and to the most conserved residues . As with the other datasets , SCA and conservation turned out to be just as good at identifying functional positions in the voltage-sensing domains of potassium channels ( see Fig . 8 ) . A similar dataset to the PDZ dataset described above is available for the lac repressor protein in E . coli [34] . The authors used amber mutations and nonsense suppressor tRNAs to perform a comprehensive mutagenesis study of lacI . In this study , each one of 328 positions was mutated to 12 or 13 alternative amino acids , and the ability of each mutant protein to repress expression of the lac genes was tested . We summarized this data by recording , for each position , how many of the tested mutations had a significant effect on the phenotype of the lac repressor . We further identified “functionally-significant” sites by considering all the positions for which at least 8 substitutions resulted in loss of function . This threshold can be varied in the whole range from 1 to 10 without significantly altering the results . As before , we observed a significant association between SCA sector positions and functional positions in the lac repressor; see Figs . 9A and 10A . However , again , the set of most conserved positions was equally good at predicting functional sites—see Figs . 9B and 10B . The results were not significantly affected by changing the size of the sector ( see Fig . 11 ) . In the previous sections , we showed that a significant fraction of the sector positions obtained from the top eigenvector of the SCA matrix can be predicted from single-site statistics . This can be attributed to a strong correlation between the components of the top eigenvector and the square root of the diagonal elements of the SCA matrix ( see Fig . 1A ) . In Halabi et al . , the top eigenvector of the SCA matrix was ignored by analogy to finance , where this mode is a consequence of global trends in the market that affect all the stocks in the same way [18] . For proteins , the analogy is suggested to be with parts of sequences that are conserved due to phylogenetic relationships between the sequences in the alignment . Here we show that there is a different mechanism that can generate a spurious top eigenmode of the SCA matrix even when there are no phylogenetic connections between the sequences in the alignment . The main ingredient in this mechanism is a positive bias for the components of the SCA matrix . Suppose that the underlying evolutionary process has no correlations between positions . Due to sampling noise , empirical correlations will typically be non-zero , and will fluctuate in a certain range . We denote the size of these fluctuations by x . The off-diagonal elements of the covariance matrix will have mean zero and variances of order C i j 2 ∼ C i i C j j x 2 . In this case , the reason for the positive bias for the components of the SCA matrix is the fact that typically SCA takes the absolute value of the covariances ( or some norm that produces only non-negative values; see S1 Text ) [18 , 24 , 25] . This implies that the off-diagonal entries of this matrix will have expectation values of order x C i i C j j . Note that the positional weights can be absorbed into the diagonal elements Cii , so we do not write them out explicitly . Even when the absolute value is not used , the correlation between the components of the top eigenmode of the SCA matrix and the diagonal elements of this matrix may also occur; this happens for example for the alignment in Smock et al . [20] . Simulations involving random alignments show that this phenomenon occurs whenever there are weak , uniform correlations between all the positions in an alignment . This can be the result of phylogenetic bias , but could have a different origin . This situation could be distinguished from the one above by looking at how the magnitude x of the off-diagonal correlations scales with alignment size; it should scale roughly like the inverse of the number of sequences if it is due to sampling noise , and be approximately constant otherwise ( we thank D . Hekstra for this observation ) . To try to explain these empirical observations , let us consider a simplified version of the SCA matrix: M = ( Δ 1 d 1 d 2 x ⋯ d 1 d n x d 2 d 1 x Δ 2 ⋯ d 2 d n x ⋯ ⋯ ⋱ ⋯ d n d 1 x d n d 2 x ⋯ Δ n ) . ( 2 ) Writing out the eigenvalue equation and performing some simple algebraic manipulations reveals that the eigenvector components vi corresponding to eigenvalue λ are related to the diagonal elements Δi by Δ i v i ∝ λ − Δ i 1 + x . ( 3 ) When the top eigenvalue is much larger than the other ones , which is usually the case when applying SCA to protein alignments , the following approximation holds: λ top ≈ x 1 + x ∑ i Δ i . ( 4 ) Empirically , this is observed to roughly match the results of SCA on real protein alignments . Given that λtop ≫ Δi , we can also write v i , top ≈ α λ top × Δ i , ( 5 ) where α is a normalization constant . This is the observed linear relation between the top eigenvector and the square root of the diagonal elements of the SCA matrix ( Fig . 1A ) . Note that the SCA matrix for an alignment does not really have the highly symmetric form ( 2 ) ; instead it shows fluctuations in the off-diagonal components . Because of this , we cannot expect to see all the eigenvectors obey eq . ( 3 ) . Indeed , for SCA matrices obtained from protein alignments , eq . ( 3 ) seems to hold only for the top eigenvector . A treatment of this problem in the framework of random matrix theory might help to clear up the expectations one should have for the top eigenvector of the SCA matrix , but such an analysis goes beyond the scope of this paper . The simple argument described above suggests that , under certain conditions that seem to hold in the cases where SCA has been applied , the top eigenvector of the SCA matrix is indeed related to conservation , and is largely independent of correlations between positions . This does not mean that there is no information contained in this top mode , but does imply that most of this information can be obtained by looking at single-site statistics alone . Note again that in our derivation the origin of the off-diagonal entries is not specified . While we showed that they can be a simple artifact of sampling noise , they could also be partly due to a non-trivial phylogenetic structure of the alignment , as previously suggested [18] .
It is perhaps not surprising that conservation is a good indicator of the functionally-important residues in a protein; indeed , this fact is one of the original motivations for using positional weights in SCA that grow with conservation levels [19] . However , as a consequence , for proteins with a single SCA sector , it is difficult to distinguish between the functional significance of sector residues and that of conserved residues . The natural solution to this problem is to focus on proteins with multiple sectors , such as the serine protease family analyzed by Halabi et al . [18] . In the serine protease case , three SCA sectors were identified by placing thresholds on certain linear combinations of eigenvectors of the SCA matrix . The top eigenvector was ignored based on an analogy to finance , and thus the issues outlined in the previous section do not apply here . The three sectors ( called ‘blue’ , ‘red’ , and ‘green’ ) were found to have independent effects on various phenotypes of the protein: the blue sector affected denaturation temperature , the red one affected binding affinity , and the green sector contained the residues responsible for catalytic activity . There are two attractive features of the serine protease data . One is that several different quantities were measured for each mutant , thus allowing for a test of the idea that the protein is split into groups each of which affects different phenotypes . Another important feature is that some double mutants were also measured , showing that mutations in different sectors act approximately independently from each other . Collecting more extensive data of this type for serine proteases and for other proteins should give more weight to the idea that SCA sectors act as functional sectors in proteins . To reduce the amount of work involved , we point out that from our observations , it seems that instead of a complete scan of all 19 alternative amino acids at each position , an alanine scan , involving only mutations to alanine , might be sufficient . Using only alanine replacements , even a complete double-mutant study of PSD95pdz3 would require about 3000 mutants , only a factor of two more than were already studied [25] . For proteins exhibiting multiple SCA sectors , this number could be lowered by focusing only on those double mutants that combine mutations in different sectors , thus testing the independence property . Finding several relevant quantities to measure for each of the mutants might not be an easy task . An ideal system for this would be related to gene expression or signal transduction , allowing measurements to be made in realistic conditions . Furthermore , it would be convenient to have a low-dimensional quantitative description of the protein’s phenotype , so that one could check whether the sectors predicted by SCA correlate with the mutations that affect the parameters in this description . One difficulty in the application of SCA is that the identification of sectors is non-trivial . Halabi et al . used visual inspection to identify linear combinations of eigenvectors to represent the sectors [18] . Independent component analysis ( ICA ) has also been invoked to find the linear combinations [19 , 20 , 22] , but a mathematically rigorous motivation for the application of this procedure is missing . An approach that avoids these difficulties is to check whether a linear regression can approximate the measured quantities for the different mutants with linear combinations of the eigenvectors of the SCA matrix . This seems to work for the case of serine protease ( see S1 Text and S1 Fig . ) , though the small number of data points prevents a statistically rigorous analysis . A similar approach does not work for the PDZ data from McLaughlin Jr . et al . , in which binding to both the cognate ( CRIPT ) ligand and to a mutated T−2F ligand was measured [25] ( see S1 Text and S2 Fig . ) . It also does not work for the potassium channels dataset , in which both the activation voltage V50 and the equivalent charge z were measured for each mutant [33] ( see S1 Text and S3 Fig . ) . This is consistent with the idea that these proteins exhibit a single sector . Conservation alone cannot in general be used to find several distinct groups of residues that have distinct functions . For this reason , finding evidence for functionally significant and independent SCA sectors would automatically favor SCA over a simple conservation analysis . However , it is important to point out that SCA , with the particular set of weights as defined by Halabi et al . [18] , is only one possible procedure for analyzing correlations in sequence alignments . Once more data is available for proteins containing multiple sectors , it will be important to compare different sets of positional weights , or different models altogether , to identify the best approach for analyzing MSAs [23] . We analyzed the available evidence regarding the hypothesis that the residues comprising the sectors identified by statistical coupling analysis are functionally significant . We looked at a number of studies , some directly related to SCA [18 , 24 , 25] , and some unrelated [33 , 34] , and we showed that while the sector positions identified by SCA do tend to be functionally relevant , in the case of single-sector proteins , conserved positions provide a statistically equivalent match to the experimental data . This observation was traced to a property of the SCA matrix that makes the components of its top eigenvector correlate strongly with its diagonal entries . We presented a mathematical model that might explain this correlation . This model suggests that , as a generic property of statistical coupling analysis , the top eigenvector of the SCA matrix does not contain information beyond that provided by single-site statistics . The observation that conservation is an important determinant of the SCA sectors is of course not very surprising , since one of the principles of SCA is to upweight the correlation information for conserved residues compared to poorly-conserved ones . However , this does pose a problem for the interpretation of the large-scale experiments that have been performed in relation to SCA [24 , 25] , given that these provide most of the available evidence for the functional significance of SCA sectors . Our analysis shows that this functional significance might be due to conservation alone . Since function is not the only reason for which protein residues may be conserved [35] , it is not surprising that the overlap with functional residues is not perfect . Once again , it is important to note that our findings do not imply that correlations within MSAs are uninformative; the contrary seems to be supported by experimental data [8 , 10–12] . However , in order to test whether the particular way in which these correlations are used within the SCA framework is useful for making functional predictions about proteins , it will be necessary to go beyond single-sector proteins and measure several different phenotypes . Such data exists [18] , but is too limited at this point to be conclusive . A thorough investigation of the idea that SCA sectors act as functional sectors requires more of this type of data , for a wider class of proteins . Whether small groups of residues inside proteins act as independent “knobs” controlling the various phenotypes is a question that can be asked independently of any statistical analysis of alignments . Such functional sectors could be found by mutagenesis work , as described above . Alternatively , one could look for structural sectors using NMR or X-ray data to search for correlated motions . This has the advantage of not requiring the modification of proteins through mutations . Finally , evolutionary sectors could be searched for by using artificial evolution experiments . If the existence of these functional , structural , or evolutionary sectors is verified with sufficient precision , one could then more easily approach the question of whether a statistical method is capable of inferring their composition from an MSA , and in this case , which method is the most efficient and accurate .
Statistical coupling analysis requires an alignment of protein sequence homologs as input data . This may contain both orthologs and paralogs , and at least moderate sequence diversity within the alignment is necessary , because an alignment of identical sequences will not contain any information about amino acid covariance . The alignments we used were generated using HHblits , with an E-value of E = 10−10 . States with 40% or more gaps were considered insert states , and were later removed from the calculations . The Uniprot IDs of the seed sequences used with HHblits are as follows: DLG4_RAT ( PDZ ) , DYR_ECOLI ( DHFR ) , KCNB1_RAT ( K+ channels ) , and LACI_ECOLI ( lacI ) . To check the robustness of the results , we also ran our analysis on Pfam alignments when available , and on the alignments from McLaughlin Jr . et al . [25] , Reynolds et al . [24] , and from Lee et al . [31] for the PDZ , DHFR , and potassium channels datasets , respectively . The statistical coupling analysis was performed in accordance with the projection method [19 , 25] , which is the default in the newest version of the SCA framework from the Ranganathan lab . The code we used for the analysis can be accessed at https://bitbucket . org/ttesileanu/multicov . Consider a multiple sequence alignment represented as an N×n matrix A in which aki is the amino acid at position i in the kth sequence . We first construct a numeric matrix X ˜ by x ˜ ki = { ϕ i ( a ki ) f i ( a ki ) ∑ b≠gap ϕ i 2 ( b ) f i 2 ( b ) if a k i ≠gap , 0 if a k i = gap , ( 6 ) where ϕi ( a ) is a positional weight , and fi ( a ) the frequency with which amino acid a occurs in column i of the alignment . The positional weights are given by ϕ i ( a ) = log [ f i ( a ) 1 − f i ( a ) 1 − q ( a ) q ( a ) ] , ( 7 ) where q ( a ) is the background frequency with which amino acid a occurs in a large protein database . The SCA matrix is , up to an absolute value , the covariance matrix associated with X ˜ , C ˜ i j = | 1 N ∑ k x ˜ k i x ˜ k j − 1 N 2 ∑ k , l x ˜ k i x ˜ l j | . ( 8 ) Finally , the sector was identified by finding the positions where the components of the top eigenvector of C ˜ ij went above a given threshold . The threshold was chosen so that the sector comprised about 25% of the number n of residues contained in each alignment sequence . More details about this method and descriptions of the other variants of SCA found in the literature can be found in the S1 Text . The conservation level of a position in the alignment is calculated using the relative entropy ( Kullback-Leibler divergence ) , as described in eq . ( 1 ) . A different definition , as the frequency of the most prevalent amino acid at a position , is highly correlated with Di and gives similar results . Note that the calculation of the relative entropy as defined in eq . ( 1 ) requires that ∑a fi ( a ) = 1 and ∑a q ( a ) = 1 . For the first of these relations to hold , we need the sum over a to include the gap , but this requires a value for the background frequency of gaps q ( gap ) . This is not straightforward to estimate or even to define . There are several solutions possible: one is to assume that the background frequency for gaps is equal to the gap frequency in the alignment averaged over all positions . Another approach is to simply ignore the gaps by focusing only on the sequences that do not contain a gap at position i . We chose the former solution , as it is the default one in the SCA framework , but the results are very similar when using the latter choice .
|
Statistical analyses of alignments of evolutionarily related protein sequences have been proposed as a method for obtaining information about protein structure and function . One such method , called statistical coupling analysis , identifies patterns of correlated mutations and uses them to find groups of coevolving residues . These groups , called protein sectors , have been reported to be relevant for various functional aspects , such as enzymatic efficiency , protein stability , or allostery . Here , we reanalyze existing data in order to assess the relative importance of two factors contributing to statistical coupling analysis , namely single-site amino acid frequencies and pairwise correlations . Although correlations have been shown to be informative in other studies , we point out that in existing large-scale data that has been analyzed with statistical coupling analysis , single-site statistics seems to be a dominating factor .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Protein Sectors: Statistical Coupling Analysis versus Conservation
|
Inflammasome activation is important for antimicrobial defense because it induces cell death and regulates the secretion of IL-1 family cytokines , which play a critical role in inflammatory responses . The inflammasome activates caspase-1 to process and secrete IL-1β . However , the mechanisms governing IL-1α release are less clear . Recently , a non-canonical inflammasome was described that activates caspase-11 and mediates pyroptosis and release of IL-1α and IL-1β . Caspase-11 activation in response to Gram-negative bacteria requires Toll-like receptor 4 ( TLR4 ) and TIR-domain-containing adaptor-inducing interferon-β ( TRIF ) -dependent interferon production . Whether additional bacterial signals trigger caspase-11 activation is unknown . Many bacterial pathogens use specialized secretion systems to translocate effector proteins into the cytosol of host cells . These secretion systems can also deliver flagellin into the cytosol , which triggers caspase-1 activation and pyroptosis . However , even in the absence of flagellin , these secretion systems induce inflammasome activation and the release of IL-1α and IL-1β , but the inflammasome pathways that mediate this response are unclear . We observe rapid IL-1α and IL-1β release and cell death in response to the type IV or type III secretion systems of Legionella pneumophila and Yersinia pseudotuberculosis . Unlike IL-1β , IL-1α secretion does not require caspase-1 . Instead , caspase-11 activation is required for both IL-1α secretion and cell death in response to the activity of these secretion systems . Interestingly , whereas caspase-11 promotes IL-1β release in response to the type IV secretion system through the NLRP3/ASC inflammasome , caspase-11-dependent release of IL-1α is independent of both the NAIP5/NLRC4 and NLRP3/ASC inflammasomes as well as TRIF and type I interferon signaling . Furthermore , we find both overlapping and non-redundant roles for IL-1α and IL-1β in mediating neutrophil recruitment and bacterial clearance in response to pulmonary infection by L . pneumophila . Our findings demonstrate that virulent , but not avirulent , bacteria trigger a rapid caspase-11-dependent innate immune response important for host defense .
Antibacterial defense is initiated by germline-encoded pattern recognition receptors ( PRRs ) , which detect conserved pathogen-associated molecular patterns ( PAMPs ) [1]–[3] . Plasma membrane-bound PRRs , such as the Toll-like receptors ( TLRs ) , detect PAMPs present in the extracellular space and endosomal compartments , whereas cytosolic PRRs , such as the NOD-like receptors ( NLRs ) , survey the host cytosol for the presence of invasive pathogens [3]–[7] . Invasive microorganisms or other cellular stresses induce assembly of cytosolic protein complexes known as inflammasomes , which play a critical role in host defense [8]–[11] . Inflammasomes respond to a wide variety of activators , including bacterial pore-forming toxins and bacterial PAMPS , such as flagellin or RNA [12]–[18] . Particular NLRs respond to their cognate stimuli and recruit the adapter protein ASC and pro-caspase-1 through homotypic protein-protein interactions between pyrin domains and caspase recruitment domains ( CARD ) , leading to autoprocessing and activation of caspase-1 [19]–[23] . Caspase-1 is responsible for processing and secreting IL-1 family cytokines and mediates a proinflammatory cell death termed pyroptosis [9] , [11] , [24] , [25] . Caspase-11 participates in the activation of a non-canonical inflammasome that induces cell death and the secretion of IL-1α and IL-1β in response to Gram-negative pathogens , such as Escherichia coli and Vibrio cholerae , and to particular toxins , such as the cholera toxin B subunit [26]–[29] . This non-canonical , caspase-11-dependent response to Gram-negative bacteria is independent of virulence-associated secretion systems that deliver bacterial molecules into the host cytosol and requires LPS-induced TLR4 signaling through the adaptor TIR-domain-containing adaptor-inducing interferon-β ( TRIF ) and TRIF-dependent type I interferon ( IFN ) production . Type I IFN signaling through the type I IFN receptor ( IFNAR ) is required for caspase-11 upregulation and activation , but how type I IFN mediates activation of caspase-11 is not well-defined [27]–[29] . Caspase-11 contributes to NLRP3-dependent activation of caspase-1 and subsequent caspase-1-dependent IL-1β secretion and cell death . Caspase-11 also facilitates an NLRP3- and caspase-1-independent pathway that results in cell death and release of IL-1α [26]–[29] . This caspase-11-dependent , caspase-1-independent pathway is responsible for LPS-induced septic shock in vivo [26] , [30] . Although caspase-11 is activated in response to signals from Gram-negative pathogens and certain pore-forming toxins , whether caspase-11 contributes to inflammasome activation in response to virulence-associated secretion systems that deliver bacterial ligands into host cytosol is unknown . Bacterial pathogens use evolutionarily conserved secretion systems , such as type III or type IV secretion systems ( T3SS or T4SS ) , to translocate effector proteins into the cytosol of host cells [31] , [32] . In addition to bona fide virulence factors , these secretion systems also translocate bacterial molecules such as flagellin or structural components of the secretion machinery itself , which results in inflammasome activation [14] , [16] , [33]–[36] . Legionella pneumophila , an opportunistic pathogen that causes a severe pneumonia known as Legionnaires' disease [37] , [38] , utilizes its dot/icm-encoded T4SS as a virulence factor to translocate bacterial effector proteins into the host cell cytosol and establish a replicative vacuole [39]–[46] . L . pneumophila induces T4SS-dependent inflammasome activation through two genetically distinct pathways [47] . T4SS-mediated translocation of flagellin into the cytosol triggers caspase-1 activation and pyroptosis through the NLR NAIP5 in conjunction with another NLR , NLRC4 [16] , [36] , [47]–[50] . Caspase-1 activation is also triggered independently of the NLRC4/flagellin pathway through the adaptor protein ASC , but the bacterial factor that is recognized and the upstream proteins that regulate this pathway remain unknown [47] , [51] . However , although ASC is necessary for robust secretion of IL-1β in response to L . pneumophila as well as a number of pathogens , such as Salmonella or Yersinia species which employ T3SSs , ASC is dispensable for induction of pyroptosis that is rapidly triggered in response to these infections . We therefore considered the possibility that in addition to its role in delayed inflammasome activation in response to Gram-negative bacteria , caspase-11 might participate in rapid cell death and release of IL-1α in response to the presence of bacterial pathogens that access the host cell cytosol by means of type IV and type III secretion systems . Here , we demonstrate that IL-1α and IL-1β are rapidly released in response to bacterial T4SS activity independently of bacterial flagellin . In this system , we find IL-1β secretion requires caspase-1 , but caspase-1 is dispensable for cell death and IL-1α release in response to a functional L . pneumophila T4SS . Instead , caspase-11 is required for both IL-1α release and cell death in response to L . pneumophila T4SS activity . Consistent with recent findings , caspase-11 contributes to optimal NLRP3-mediated caspase-1 activation and IL-1β secretion in response to L . pneumophila . However , caspase-11-dependent IL-1α release and cell death in L . pneumophila-infected cells are independent of the NAIP5/NLRC4 and NLRP3/ASC inflammasomes . In contrast to the role of TRIF and IFNAR in the response against Gram-negative bacteria , caspase-11 activation and cytokine release in response to the T4SS of L . pneumophila are independent of both TRIF and IFNAR signaling . We further demonstrate that T3SS activity of the unrelated pathogen Yersinia pseudotuberculosis induces a similarly rapid caspase-11-dependent response that also leads to cell death and release of IL-1α and IL-1β . Finally , we find that both IL-1α and IL-1β are critical in vivo for neutrophil recruitment and bacterial clearance . Overall , our data show that caspase-11 is poised to respond robustly to a conserved feature of pathogenic bacteria , bacterial access to the host cytosol through specialized secretion systems . This establishes caspase-11 as a critical regulator of immune system-mediated discrimination of pathogenic and nonpathogenic bacteria .
L . pneumophila infection induces IL-1α and IL-1β secretion that requires T4SS activity [47] , [52] . IL-1β secretion is regulated by a flagellin-dependent NAIP5/NLRC4 inflammasome and a poorly defined ASC inflammasome that both activate caspase-1 [47] , [51] . The mechanisms underlying IL-1α secretion are less clear , but IL-1α secretion is still robustly induced by flagellin-deficient L . pneumophila , which do not activate the NAIP5/NLRC4 inflammasome [52] . Recent studies have described a non-canonical inflammasome triggered in response to Gram-negative bacteria . This non-canonical inflammasome requires lipopolysaccharide ( LPS ) for the upregulation and activation of caspase-11 and subsequent IL-1α and IL-1β release [26]–[29] . Whether caspase-11 is also activated in response to bacteria that use specialized secretion systems to translocate bacterial molecules into the host cytosol is unknown . We thus hypothesized that LPS priming would upregulate caspase-11 , pro-IL-1α , and pro-IL-1β and allow for more robust and rapid IL-1α and IL-1β secretion in response to T4SS activity . To test this , we first compared IL-1α and IL-1β release in unprimed and LPS-primed bone marrow-derived macrophages ( BMDMs ) . As shown previously [48] , [52] , unprimed BMDMs secrete robust levels of IL-1α and IL-1β by 20 hours post-infection with wild-type L . pneumophila ( WT Lp ) ( Figure 1A ) . Slightly attenuated levels of secreted IL-1α and IL-1β are observed with flagellin-deficient L . pneumophila ( ΔflaA Lp ) , which do not activate the NAIP5/NLRC4 inflammasome [17] , [18] . Secretion of both cytokines is significantly diminished during infection with L . pneumophila lacking DotA , an essential component of the T4SS ( ΔdotA Lp ) , and is significantly diminished in caspase-1/caspase-11-deficient ( Casp1−/−Casp11−/− ) macrophages as well ( Figure 1A ) . The diminished IL-1 secretion induced by ΔdotA Lp is not due to a lack of pro-IL-1 production , as ΔdotA Lp and WT Lp induce robust levels of pro-IL-1β ( Figure S1A ) . At 4 hours post-infection , unprimed macrophages do not secrete IL-1 ( Figure 1B ) . However , LPS-primed cells rapidly secrete IL-1α and IL-1β , and this secretion is abrogated in Casp1−/−Casp11−/− macrophages ( Figure 1B ) . Secretion of IL-18 , another IL-1 family cytokine , also requires T4SS activity and is eliminated in Casp1−/−Casp11−/− cells ( Figure S1B ) . Comparable levels of the caspase-1/caspase-11-independent cytokines IL-12 and TNF-α are secreted in the absence and presence of LPS priming ( Figure S1C–D ) . These data suggest that LPS priming upregulates a factor required for rapid IL-1α and IL-1β release in response to L . pneumophila T4SS activity . Secretion of IL-1β in response to both canonical and non-canonical inflammasome activation requires caspase-1 [26] , [53] , [54] . In contrast , IL-1α release downstream of the non-canonical inflammasome depends on caspase-11 , and does not require caspase-1 [26] . To test if the catalytic activity of caspase-1 is required for IL-1α secretion in response to L . pneumophila , we inhibited caspase-1 catalytic activity with the pharmacological inhibitor YVAD-cmk ( YVAD ) . Consistent with previous studies [53] , IL-1β secretion in response to L . pneumophila is substantially inhibited by YVAD . However , YVAD has no effect on IL-1α secretion , indicating that IL-1α release in response to L . pneumophila does not require caspase-1 catalytic activity ( Figure 1C ) , as has been shown for other inflammasome activators [55] . Given that IL-1α secretion occurs more rapidly upon LPS priming , is abrogated in Casp1−/−Casp11−/− macrophages , and does not require caspase-1 catalytic activity , we considered the possibility that caspase-11 might participate in inflammasome activation during L . pneumophila infection . To test the genetic requirement for caspase-11 in the inflammasome response to L . pneumophila , we infected BMDMs from either caspase-1-deficient ( Casp1−/− ) or caspase-11-deficient ( Casp11−/− ) mice . In the absence of flagellin , caspase-11 is required for IL-1α secretion , whereas it is not essential for IL-1β secretion but contributes to maximal secretion ( Figure 2A ) . These data suggest that caspase-11 is activated in response to L . pneumophila infection independently of flagellin . Indeed , there is robust processing and secretion of caspase-11 in response to WT and ΔflaA Lp ( Figure S2 ) . In accordance with previous findings [26] , [53] , caspase-1 is absolutely required for IL-1β secretion . In contrast , we observe robust IL-1α release even in the absence of caspase-1 . Both IL-1α and IL-1β release in response to ΔflaA Lp are caspase-11-dependent in both primed and unprimed macrophages ( Figures 2 , S3A–B ) , making L . pneumophila distinct from other Gram-negative bacteria that require priming to induce robust caspase-11 upregulation and activation [27] . Thus , while caspase-11 contributes to maximal caspase-1-dependent IL-1β secretion , it is both necessary and sufficient for IL-1α release in response to flagellin-deficient L . pneumophila . Cell death in B6 BMDMs is partially flagellin-dependent , but is flagellin-independent in Casp1−/− BMDMs ( Figure 2B ) . Importantly , cell death in response to flagellin-deficient L . pneumophila requires caspase-11 , thus correlating caspase-11-dependent cell death with IL-1α release from host cells . In contrast , and consistent with previous findings [26] , LPS+ATP induces canonical caspase-1-dependent pyroptosis and secretion of IL-1α and IL-1β that is independent of caspase-11 . Because caspase-1 must be processed to mediate IL-1β secretion [53] , we examined whether caspase-1 processing is decreased in the absence of caspase-11 , which could account for the decreased IL-1β secretion in response to ΔflaA Lp . Caspase-1 processing is slightly attenuated but not abrogated in response to ΔflaA Lp in Casp11−/− macrophages , consistent with the slight decrease in IL-1β secretion ( Figures 2C , S3C ) . Thus , flagellin-deficient L . pneumophila trigger a canonical caspase-1-dependent inflammasome as well as a non-canonical caspase-11-dependent inflammasome . The ASC and NAIP5/NLRC4 inflammasomes are required for caspase-1 activation and IL-1β secretion in response to L . pneumophila [47] . To determine if these inflammasomes are also required for caspase-11 activation and IL-1α release , we infected ASC/NLRC4-deficient ( Asc−/−Nlrc4−/− ) BMDMs with L . pneumophila . Asc−/−Nlrc4−/− BMDMs do not secrete IL-1β in response to either WT Lp , ΔflaA Lp , or LPS+ATP . However , Asc−/−Nlrc4−/− BMDMs still release IL-1α in response to ΔflaA Lp in primed and unprimed macrophages ( Figures 3A , S4 ) . Thus , unlike IL-1β , IL-1α is released independently of flagellin , ASC , and NLRC4 . Accordingly , despite an absence of processed caspase-1 p10 , robust levels of processed caspase-11 p26 are detected in the supernatants of Asc−/−Nlrc4−/− cells infected with either WT or ΔflaA Lp but not in response to LPS+ATP ( Figure 3B ) . We next sought to determine whether IL-1α is also released independently of ASC and NLRC4 during in vivo infection . Because flagellin-deficient L . pneumophila do not activate the NLRC4 inflammasome [16] , [17] , [47] , infecting Asc−/− mice with ΔflaA Lp eliminates both the ASC and NLRC4 inflammasome pathways . Importantly , the level of IL-1β in the bronchoalveolar lavage fluid ( BALF ) 24 hours post-infection is significantly attenuated in Asc−/− mice infected with ΔflaA Lp ( Figure 3C ) . In contrast , the level of IL-1α in the BALF is unaffected even in the absence of both the ASC and NLRC4 inflammasomes . Both IL-1α and IL-1β release are significantly diminished in caspase-1/caspase-11-deficient mice ( Figure S5 ) . Collectively , our data indicate that L . pneumophila triggers caspase-11 activation and IL-1α release independently of the ASC and NLRC4 inflammasomes during both in vitro and in vivo infection . L . pneumophila induces caspase-1 activation and IL-1β and IL-18 secretion through two genetically distinct pathways , one involving ASC and one involving NLRC4 ( Figures 4A , S6A–B ) [47] . The upstream host and bacterial components of the ASC-dependent response to L . pneumophila are still unknown , but are independent of the flagellin/NAIP5/NLRC4 pathway ( Figures 4A , S6B ) [47] . Because caspase-11 contributes to maximal IL-1β secretion in response to ΔflaA Lp , we further investigated the ASC-dependent mechanism of inflammasome activation . NLRP3 , an NLR involved in inflammasome-dependent responses to a wide variety of pathogens , requires ASC to mediate caspase-1 processing during both canonical and non-canonical inflammasome activation [9] , [12] , [18] , [26] , [56] . We therefore investigated the role of NLRP3 in the response to ΔflaA Lp . Notably , IL-1β and IL-18 secretion are abrogated during infection of NLRP3-deficient ( Nlrp3−/− ) BMDMs with ΔflaA Lp in both primed and unprimed macrophages ( Figures 4B , S7A–C ) . Consistently , we do not detect processed caspase-1 p10 in the supernatants of Nlrp3−/− macrophages infected with ΔflaA Lp ( Figure 4C ) . Thus , NLRP3 functions together with ASC , caspase-1 , and caspase-11 to control IL-1β secretion in response to flagellin-deficient L . pneumophila . However , IL-1α release and cell death following infection with flagellin-deficient L . pneumophila are independent of NLRP3 ( Figures 4B , S7A ) , indicating that caspase-11 also mediates an NLRP3-independent response towards flagellin-deficient L . pneumophila . Accordingly , NLRP3-dependent IL-1β secretion in response to flagellin-deficient L . pneumophila was inhibited by extracellular potassium , whereas NLRP3-independent caspase-11-dependent IL-1α secretion and cell death were not affected ( Figure S7D–E ) . Recent data demonstrate that caspase-11 activation in response to a wide variety of Gram-negative bacteria requires TLR4 signaling through its adaptor TRIF and subsequent type I IFN production [27]–[29] . To determine if L . pneumophila engages a similar TRIF and type I IFN receptor ( IFNAR ) -dependent pathway for caspase-11 activation , we infected TRIF-deficient ( Trif−/− ) and IFNAR-deficient ( Ifnar−/− ) BMDMs . Unlike the response to E . coli , L . pneumophila infection of unprimed macrophages triggered robust cell death and secretion of IL-1α and IL-1β that was independent of IFNAR and TRIF ( Figure 5A–B ) . Consistently , priming with the TLR1/2 agonist Pam3CSK4 , which results in TRIF- and IFNAR-dependent cytokine secretion and cell death in response to E . coli [27] , still induced cell death and cytokine secretion in TRIF- and IFNAR-deficient cells in response to L . pneumophila ( Figure S8A–B ) . These data suggest that during L . pneumophila infection , caspase-11 is upregulated and activated independently of TRIF and IFNAR signaling . Indeed , caspase-11 is still robustly processed and secreted independently of IFNAR and TRIF ( Figures 5C , S9 ) . Notably , substantially upregulated levels of pro-caspase-11 are not observed in the lysates of cells infected with WT or ΔflaA Lp because both the pro and cleaved forms of caspase-11 are rapidly secreted into the cell supernatant upon infection ( Figures 5C , S9 ) . Accordingly , lysates from IFNAR- and TRIF-deficient macrophages infected with L . pneumophila express comparable levels of pro-caspase-11 to wild-type macrophages , whereas TRIF and IFNAR do contribute to upregulation of pro-caspase-11 in response to E . coli ( Figure S10A–C ) . When the macrophages are primed with LPS prior to infection , there is a moderate contribution of TRIF and IFNAR signaling to inflammasome activation , consistent with the observation that LPS stimulates the TLR4-TRIF-IFNAR axis involved in caspase-11 upregulation ( Figure S8C–D ) . Because the caspase-11-dependent response to L . pneumophila is TRIF-independent , we investigated whether the TLR signaling adaptor MyD88 contributes to caspase-11 upregulation . When immortalized macrophages deficient for both MyD88 and Trif ( iMyd88−/−Trif−/− ) were infected , caspase-11 upregulation was abrogated in response to both WT and ΔflaA Lp ( Figure S11A–B ) , and we were unable to detect caspase-11 activation ( data not shown ) . Thus , although TRIF is not required for caspase-11 activation , a TLR-dependent signal is likely required as the loss of both MyD88 and TRIF eliminates caspase-11 upregulation and activation . Because caspase-11 activation in response to L . pneumophila expressing a functional T4SS is so rapid and robust , we sought to test whether this robust caspase-11-dependent inflammasome activation might be a general response to the activity of specialized secretion systems that allow for bacterial access to the host cytosol . The Yersinia pseudotuberculosis type III secretion system ( T3SS ) induces inflammasome activation independently of bacterial flagellin and the known secreted effector proteins , and this inflammasome activation is important for bacterial clearance [57] . Since wild-type Yersinia induces cell death that is independent of both caspase-1 and -11 and requires the secreted effector YopJ [57] , [58] , we instead infected Casp1−/−Casp11−/− , Casp1−/− , and Casp11−/− BMDMs with a strain of Y . pseudotuberculosis that expresses a T3SS but lacks the six known secreted effectors ( Δ6 Yp ) . Similarly to L . pneumophila infection , both IL-1α and IL-1β release in response to Δ6 Yp are caspase-11-dependent ( Figure 6A ) . Again , caspase-1 is absolutely required for IL-1β secretion , whereas IL-1α is released independently of caspase-1 . Secretion of IL-12 , an inflammasome-independent cytokine , is unaffected ( Figure S12 ) . Cell death in response to Δ6 Yp is both caspase-1 and caspase-11-dependent , with a more dramatic reduction in death in Casp11−/− BMDMs ( Figure 6B ) . Furthermore , Y . pseudotuberculosis-induced release of both IL-1α and IL-1β requires the presence of a functional T3SS , as Y . pseudotuberculosis unable to form a functional T3SS pore in the host cell plasma membrane ( ΔyopB Yp ) do not induce secretion of either cytokine . These data indicate a general role for caspase-11 in the induction of rapid cell death and robust release of IL-1α and IL-1β in response to bacterial secretion systems that are capable of accessing the host cell cytosol , but may be independent of the activities of specific virulence factors per se . As caspase-11 contributes to flagellin-independent IL-1α and IL-1β release from infected macrophages in vitro and IL-1α and IL-1β secretion is flagellin-independent in vivo , we wanted to determine the contribution of IL-1α and IL-1β to host defense against L . pneumophila in vivo . IL-1α and IL-1β both bind the IL-1 receptor ( IL-1R ) , which signals through the MyD88 adaptor protein [59]–[61] . As MyD88 is critical for control of L . pneumophila replication during in vivo infection but deletion of an individual MyD88-dependent TLR or a combination of TLRs does not recapitulate MyD88 deficiency , it is likely that other MyD88-dependent receptors , including the IL-1R , may play a role [62]–[66] . IL-1R signaling contributes to chemokine production by non-hematopoietic cells during infection with wild-type , flagellin-expressing L . pneumophila [67] . However , the role of IL-1R signaling during infection with flagellin-deficient L . pneumophila , which do not activate the NAIP5/NLRC4 inflammasome , has not been investigated . We therefore infected B6 and IL-1R-deficient ( Il1r1−/− ) mice intranasally with ΔflaA Lp and measured bacterial burden in the lung over the course of seven days . Though both B6 and Il1r1−/− mice received similar initial bacterial burdens , Il1r1−/− mice show a defect in bacterial clearance as early as 24 hours post-infection ( Figure 7A ) . Bacterial burden remains elevated in the absence of IL-1R signaling , with the Il1r1−/− mice still exhibiting a log-increase in bacterial load at 120 hours post-infection . Since IL-1R signaling is important for neutrophil recruitment [68] , we examined whether Il1r1−/− mice have a defect in neutrophil recruitment to the pulmonary airway during L . pneumophila infection . Indeed , Il1r1−/− mice exhibit a significant decrease in neutrophil recruitment to the airway 24 hours post-infection , possibly contributing to their inability to efficiently clear the pathogen ( Figure 7B–C ) . The IL-1R signals in response to both IL-1α and IL-1β; however , these cytokines can play non-redundant roles in anti-bacterial defense [69] . To determine the relative contributions of IL-1α and IL-1β to neutrophil recruitment and bacterial clearance during L . pneumophila infection , we utilized neutralizing antibodies to selectively block either IL-1α or IL-1β prior to infection . Specific cytokine neutralization in the BALF could be observed 24 hours post-infection ( Figure S13 ) . Critically , IL-1α neutralization alone significantly diminishes the percentage of neutrophils recruited to the BALF at 24 hours post-infection and results in a half-log increase in bacterial CFUs , in marked contrast to isotype control antibody or neutralization of IL-1β , which on its own did not have a significant effect ( Figure 7D–F ) . However , neutralization of both IL-1α and IL-1β fully recapitulates the magnitude of neutrophil reduction and defect in bacterial clearance observed in the Il1r1−/− mice . Collectively , these data indicate that although there are some overlapping roles for these cytokines during L . pneumophila infection , IL-1α plays a distinct role from IL-1β in driving neutrophil recruitment to the airway and mediating bacterial clearance .
Inflammasomes respond robustly to conserved features of pathogenic microbes , such as pore-forming toxins or specialized secretion systems that access the host cytosol . Inflammasomes therefore play a central role in enabling the immune system to discriminate between virulent and avirulent bacteria [70] . Recent reports show a role for caspase-11 in regulating the activation of a non-canonical inflammasome that promotes cell death as well as IL-1α and IL-1β secretion . This non-canonical inflammasome responds to both pathogenic and non-pathogenic Gram-negative bacteria independently of specialized secretion systems that translocate bacterial molecules into the host cytosol [26]–[29] . This pathway involves the TRIF- and IFNAR-dependent upregulation and activation of caspase-11 and occurs with relatively delayed kinetics in comparison to the response to pathogenic bacteria . Intriguingly , we find that the activity of the L . pneumophila Dot/Icm T4SS leads to rapid and robust caspase-11 activation independently of the TRIF-IFNAR axis , and this activation triggers rapid cell death and release of both IL-1α and IL-1β ( Figure 8 ) . We extend these results to show that the evolutionarily distinct T3SS of another pathogen , Y . pseudotuberculosis , also rapidly triggers caspase-11-dependent responses . Collectively , our findings demonstrate that caspase-11 is critical for inflammasome activation in response to the secretion systems of virulent bacteria that enable bacterial molecules to access the host cell cytosol and demonstrate that IL-1α and IL-1β together play a crucial protective role during acute infection in vivo . We demonstrate that in response to the activity of bacterial secretion systems that enable cytosolic access , caspase-11 contributes to NLRP3-mediated inflammasome activation and caspase-1-dependent IL-1β secretion and to a second ASC and NLRC4-independent pathway that does not require caspase-1 and leads to cell death as well as robust IL-1α release . These L . pneumophila-induced pathways are similar to recent findings with a number of Gram-negative bacterial pathogens , including C . rodentium , E . coli , and S . typhimurium [26]–[29] . However , we observe rapid and robust T4SS-dependent activation of these two caspase-11-mediated pathways by L . pneumophila , whereas the response to Gram-negative bacteria lacking specialized secretion systems occurs less robustly and with much slower kinetics . Intriguingly , we observe a similarly rapid caspase-11-dependent induction of cell death and IL-1 release in response to the structurally and evolutionarily unrelated T3SS of Y . pseudotuberculosis . Importantly , this pathway is independent of host sensing of flagellin , as it is triggered by flagellin-deficient L . pneumophila , and Y . pseudotuberculosis downregulates flagellin expression when the T3SS is expressed [71] . Thus , our data suggest that the caspase-11 inflammasome is poised to respond robustly and rapidly to the activity of bacterial secretion systems that are capable of delivering microbial products to the host cell cytosol and may enable the host to respond to pathogens that evade flagellin-dependent responses . This could have significance for understanding the role of caspase-11 activation at mucosal sites colonized by large numbers of commensal bacteria . At mucosal barriers , it would be expected that the non-canonical inflammasome pathway would not be robustly activated by commensal bacteria but could respond rapidly to the presence of bacterial secretion systems that enable pathogen access to the host cytosol . Our findings are consistent with recent observations that the L . pneumophila Dot/Icm T4SS triggers the caspase-11-dependent non-canonical inflammasome [72] , as well as the finding that bacteria that enter the cytosol either due to failure to maintain integrity of their replicative vacuoles or natural entry into the cytoplasm also trigger rapid caspase-11 activation [73] . Thus the pathway that leads to caspase-11 activation appears to be particularly sensitive to pathogens that ‘violate the sanctity of the cytosol’ [74] , either through the activity of specialized secretion systems that translocate bacterial molecules into the cytosol or through their direct entry into the host cell cytosol . Whether other pathogens that replicate within the cytosol , such as Listeria or Shigella , or cytosolic viruses possess mechanisms to evade this pathway remains to be determined . L . pneumophila T4SS-mediated activation of caspase-11 differs from the other pathways of non-canonical inflammasome activation in several ways . First , L . pneumophila-mediated activation of caspase-11 does not require TRIF or IFNAR signaling . We observe a moderate dependence on TRIF and IFNAR signaling when macrophages are primed with LPS prior to infection , consistent with LPS-dependent upregulation of caspase-11 expression through the TLR4-TRIF-IFNAR axis [27]–[29] . However , in the absence of LPS priming , TRIF and IFNAR signaling are dispensable for L . pneumophila-dependent caspase-11 activation . In this context , it is likely that MyD88 compensates for the absence of TRIF , as cells deficient for both MyD88 and TRIF failed to activate caspase-11 in response to L . pneumophila . Thus , although the TLR4-TRIF-IFNAR axis is required for caspase-11 activation in response to Gram-negative bacteria , a MyD88-dependent signal is sufficient for caspase-11 activation in response to pathogens that utilize virulence-associated secretion systems to translocate bacterial molecules into the host cytosol . It is possible that different signals are capable of activating caspase-11 through distinct pathways , but these pathways occur with distinct kinetics because they may indicate distinct levels of pathogenicity . Thus , while caspase-11 is robustly upregulated by LPS priming , this upregulation alone is insufficient for rapid activation in response to bacteria that lack specialized secretion systems , as ΔdotA or ΔyopB bacteria do not induce rapid cell death even in primed cells . Collectively , these data indicate a two-signal model for rapid caspase-11 activation during infection with virulent bacteria , where bacterial PAMPs induce caspase-11 upregulation , but rapid caspase-11 activation requires a second , secretion system-dependent signal ( Figure 8 ) . The specific secretion system-dependent signals responsible for caspase-11 activation are currently unknown . While rapid activation of caspase-11 requires the presence of a functional type III or type IV secretion system or cytosolic access of the bacteria , whether the signal is an as-yet-undefined translocated bacterial molecule or a cellular response to the pore forming activity of these systems remains to be determined . The delayed NLRP3- and caspase-11-dependent response to Gram-negative bacteria suggests that in addition to LPS-induced upregulation of inflammasome components , bacterial mRNA provides an additional signal for activating the NLRP3 inflammasome [18] , [75] , although the role of caspase-11 in this response has not been formally demonstrated . Activity of the type III or IV secretion systems may bypass the need for bacterial mRNA . Alternatively , these secretion systems may translocate bacterial RNA [70] , [76] , [77] , and the rapid caspase-11-dependent response they induce could be due to more rapid delivery of bacterial mRNA into the host cell cytosol . Furthermore , the host factors required for activation of the NLRP3-independent caspase-11-dependent inflammasome also remain to be identified . As this pathway is independent of flagellin sensing , NLRP3 , ASC , and NLRC4 , an unknown upstream sensor and/or adaptor may be involved in caspase-11 activation in response to a translocated bacterial substrate or an endogenous signal induced by infection . This sensor may also be upregulated by type I IFN signaling itself [27]–[29] . Our data show that IL-1α release during L . pneumophila infection is controlled by two independent pathways , one involving the flagellin-dependent NAIP5/NLRC4 and caspase-1-dependent inflammasome and a second pathway involving the NLRP3-independent caspase-11-dependent inflammasome ( Figure 8 ) . Though we demonstrate that IL-1α release has an important biological consequence in vivo for neutrophil recruitment and bacterial clearance , it is unclear if IL-1α release is regulated by unconventional secretion , as is the case for IL-1β [78] . As both pathways that control IL-1α release also lead to cell death , our data are consistent with a model in which IL-1α is an endogenous alarmin that is released during cell death [79] . Interestingly , caspase-11 also contributes to control of flagellin-expressing L . pneumophila by serving as a component of an NLRC4-dependent inflammasome that promotes trafficking of the L . pneumophila-containing vacuole to lysosomes [80] . Thus , caspase-11 may function in multiple ways to control L . pneumophila infection . Importantly , we find that IL-1α , IL-1β , and IL-1R signaling play an important role in the control of L . pneumophila infection through efficient neutrophil recruitment to the airway . IL-1α and IL-1β play both distinct and overlapping roles in mediating neutrophil recruitment and controlling bacterial replication , as depletion of IL-1α alone showed a more pronounced defect in neutrophil recruitment and bacterial clearance than depletion of IL-1β alone , but loss of both cytokines resulted in a further reduction of neutrophil recruitment and an increased defect in bacterial clearance . Further analysis is required to define the relative contributions of the various caspase-11-mediated effector functions to the control of L . pneumophila replication in vivo . In conclusion , these studies demonstrate that T3SS and T4SS activities trigger rapid and robust activation of caspase-11 . This activation contributes to maximal NLRP3-dependent IL-1β secretion as well as to NLRP3-independent IL-1α release and host cell death . The downstream effector functions of these pathways are important for host defense against L . pneumophila in vivo , as IL-1α and IL-1β promote neutrophil recruitment to L . pneumophila-infected lungs and control pulmonary bacterial replication . Our results highlight the contribution of caspase-11 to rapid inflammasome activation and discrimination between pathogenic and nonpathogenic bacteria .
This study was carried out in strict accordance as defined in the federal regulations set forth in the Animal Welfare Act ( AWA ) , the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health , and the guidelines of the University of Pennsylvania Institutional Animal Use and Care Committee . The protocols were approved by the Institutional Animal Care and Use Committee at the University of Pennsylvania ( protocols #803465 and #803459 ) . Legionella pneumophila serogroup 1 strains were used in all experiments . Macrophages were infected with Lp02 ( thyA ) , a thymidine auxotroph derived from strain Lp01 [40] , or ΔdotA [81] and ΔflaA [16] isogenic mutant strains . For in vivo studies , mice were infected with the Lp02 ΔflaA or the JR32 [82] ΔflaA isogenic mutant strain where indicated . For in vitro and in vivo studies , L . pneumophila were cultured on charcoal yeast extract agar for 48 hours at 37°C prior to infection . Escherichia coli BL21 strain was cultured in LB broth for 16 hours at 37°C prior to infection . The Yersinia pseudotuberculosis strains used were IP2666 ΔyopHOJEMK ( Δ6 ) [58] and ΔyopB [83] . Yersinia were grown overnight with aeration in 2×YT broth at 26°C . The bacteria were diluted into fresh 2×YT containing 20 mM sodium oxalate and 20 mM MgCl2 . Bacteria were grown with aeration for 1 hour at 26°C followed by 2 hours at 37°C prior to infection . C57BL/6 mice were purchased from Jackson Laboratories . Casp1−/−Casp11−/− [84] , Casp1−/− ( unpublished data , T . S . and R . A . F . ) , Casp11−/− [30] , Asc−/− [23] , Nlrc4−/− [85] , Asc−/−Nlrc4−/− [47] , Ifnar−/− [86] , Trif−/− [87] , Il1r1−/− [88] , and Nalp3−/− [89] mice are all on the C57BL/6 background . Asc−/− , Nlrc4−/− , and Nlrp3−/− mice were originally generated by Millenium Pharmaceuticals and were a kind gift of Richard Flavell . Animals were maintained in accordance with the guidelines of the University of Pennsylvania Institutional Animal Use and Care Committee . 8–12 week-old mice were anesthetized by intraperitoneal injection of a ketamine/xylazine/PBS solution at a dose of 100 mg/kg ketamine and 10 mg/kg xylazine . Mice were infected intranasally with 40 µl of a bacterial suspension containing 1×106 CFU L . pneumophila or PBS vehicle control . For antibody neutralization experiments , mice were injected intraperitoneally with 100 µg anti-IL-1α antibody ( clone ALF-161 ) , 100 µg anti-IL-1β antibody ( clone B122 ) , 100 µg of each anti-IL-1α and anti-IL-1β antibody , or 100 µg Armenian hamster IgG1 isotype control antibody ( eBioscience ) 16 hours prior to intranasal infection . At the indicated timepoints after infection , mice were sacrificed , and the bronchoalveolar lavage fluid ( BALF ) and lungs were harvested . To determine bacterial load , the lungs were mechanically homogenized in sterile distilled H2O and a portion of the lysate was spread onto CYE plates . Animal experiments were performed in accordance with approved University of Pennsylvania Institutional Animal Care and Use Committee protocols and procedures . Bone marrow was collected from the femurs and tibiae of mice . Bone marrow cells were differentiated into macrophages by culturing the cells in RPMI containing 30% L929 cell supernatant and 20% FBS at 37°C in a humidified incubator . The macrophages were replated one day prior to infection in RPMI containing 15% L929 cell supernatant and 10% FBS . For experiments involving LPS-primed macrophages , macrophages in 48-well plates ( 2 . 0×105 cells/well ) were pretreated with 0 . 5 µg/mL LPS for 2 . 5 hours and either mock-infected with PBS , infected with L . pneumophila at an MOI = 10 for 4 hours , or treated with 2 . 5 mM ATP for 1 or 4 hours . For experiments performed in the absence of LPS priming , macrophages in 48-well plates ( 2 . 0×105 cells/well ) were either mock-infected with PBS , infected with L . pneumophila at an MOI = 10 for 16 or 20 hours , or infected with E . coli at an MOI = 25 for 1 hour followed by gentamycin treatment for 15 hours . To assess the involvement of caspase-1 catalytic activity , macrophages were treated with 20 µM or 40 µM of the caspase-1 inhibitor YVAD-cmk ( Bachem ) or an equivalent volume of dimethyl sulfoxide ( vehicle control ) 0 . 5 hours prior to infection . For L . pneumophila and E . coli infections , bacteria were centrifuged down onto the macrophages at 1200 RPM for ten minutes prior to incubation . For Y . pseudotuberculosis infection , bacteria were washed three times with pre- warmed DMEM , added to the cells at an MOI = 20 , and centrifuged down onto the macrophages at 1000 rpm for 5 min . Cells were incubated at 37°C for 1 hour post-infection followed by addition of 100 µg/mL gentamicin . Supernatants were harvested 4 hours post infection for ELISA and LDH analysis . Cells were infected or treated as described above , and supernatants were harvested at the indicated times post-infection . Lactate dehydrogenase ( LDH ) release was quantified using the LDH Cytotoxicity Assay Kit ( Clontech ) according to the manufacturer's instructions . Supernatants from infected cells were mixed 1∶1 with 2 X SDS-PAGE sample buffer or infected BMDMs were directly lysed in 1 X SDS-PAGE sample buffer . Samples were boiled , separated by SDS-PAGE , and transferred to Immobilon P membranes ( Millipore ) . Primary antibodies against caspase-1 p10 ( Santa Cruz Biotechnology ) , caspase-11 ( Sigma , clone 17D9 ) , IL-1β ( R&D systems ) , and β-actin ( Sigma ) were used . Detection was performed with HRP-conjugated anti-rabbit IgG ( Cell Signaling Technology ) or anti-rat IgG ( Santa Cruz Biotechnology or Jackson Immuno ) . Harvested supernatants from infected macrophages or the BALF from infected mice were assayed using capture and detection antibodies specific for IL-18 ( MBL ) , IL-1α , IL-1β , and IL-12p40 ( BD Biosciences ) . To determine neutrophil recruitment to the airway , BALF cells were stained with Live/Dead Fixable Dead Cell Stain ( Invitrogen ) , and antibodies specific for CD45 , Gr-1 ( eBioscience ) , and Ly6G ( Biolegend ) . Data were collected with an LSRII flow cytometer ( BD Biosciences ) and post-collection data was analyzed using FlowJo ( Treestar ) . Cells were gated on singlets and live cells . Neutrophils were identified as being CD45+ , Gr-1+ , and Ly6G+ . Plotting of data and statistical analysis were performed using Graphpad Prism software , and statistical significance was determined by the unpaired two-tailed Student's t test , one-way ANOVA with Tukey post-test , or two-way ANOVA with Bonferroni post-test . Differences were considered statistically significant if the P value was <0 . 05 .
|
The inflammasome , a multiprotein complex , is critical for host defense against bacterial infection . The inflammasome activates the host protease caspase-1 to process and secrete IL-1β . Another caspase , caspase-11 , can cause cell death and IL-1α release . The bacterial signals that trigger caspase-11 activation are poorly understood . A common feature of many bacterial pathogens is the ability to inject virulence factors and other bacterial molecules into the host cell cytosol by means of a variety of virulence-associated secretion systems . These secretion systems can introduce bacterial flagellin into the host cytosol , which leads to caspase-1 activation and cell death . However , many bacteria lack or down-regulate flagellin yet still activate the inflammasome . Here , we show that the type IV secretion system of Legionella pneumophila and the type III secretion system of Yersinia pseudotuberculosis rapidly trigger caspase-11 activation in a flagellin-independent manner . Caspase-11 activation mediates two separate inflammasome pathways: one leading to IL-1β processing and secretion , and one leading to cell death and IL-1α release . Furthermore , we find these caspase-11-regulated cytokines are critical for neutrophil recruitment to the site of infection and clearance of non-flagellated Legionella in vivo . Overall , our findings show that virulent bacteria activate a rapid caspase-11-dependent immune response that plays a critical role in host defense .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"gram",
"negative",
"inflammation",
"immune",
"cells",
"monocytes",
"immunity",
"innate",
"immunity",
"immune",
"activation",
"immune",
"defense",
"immunology",
"immunity",
"to",
"infections",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"bacterial",
"pathogens",
"pathogenesis"
] |
2013
|
Caspase-11 Activation in Response to Bacterial Secretion Systems that Access the Host Cytosol
|
Disparate enveloped viruses initiate infection by fusing with endosomes . However , the highly diverse and dynamic nature of endosomes impairs mechanistic studies of fusion and identification of sub-cellular sites supporting the nucleocapsid release . We took advantage of the extreme stability of avian retrovirus-receptor complexes at neutral pH and of acid-dependence of virus-endosome fusion to isolate the latter step from preceding asynchronous internalization/trafficking steps . Viruses were trapped within endosomes in the presence of NH4Cl . Removal of NH4Cl resulted in a quick and uniform acidification of all subcellular compartments , thereby initiating synchronous viral fusion . Single virus imaging demonstrated that fusion was initiated within seconds after acidification and often culminated in the release of the viral core from an endosome . Comparative studies of cells expressing either the transmembrane or GPI-anchored receptor isoform revealed that the transmembrane receptor delivered the virus to more fusion-permissive compartments . Thus the identity of endosomal compartments , in addition to their acidity , appears to modulate viral fusion . A more striking manifestation of the virus delivery to distinct compartments in the presence of NH4Cl was the viral core release into the cytosol of cells expressing the transmembrane receptor and into endosomes of cells expressing the GPI-anchored isoform . In the latter cells , the newly released cores exhibited restricted mobility and were exposed to a more acidic environment than the cytoplasm . These cores appear to enter into the cytosol after an additional slow temperature-dependent step . We conclude that the NH4Cl block traps the virus within intralumenal vesicles of late endosomes in cells expressing the GPI-anchored receptor . Viruses surrounded by more than one endosomal membrane release their core into the cytoplasm in two steps – fusion with an intralumenal vesicle followed by a yet unknown temperature-dependent step that liberates the core from late endosomes .
A large number of enveloped and non-enveloped viruses enter cells through endocytosis [1] . Depending on the nature of cellular receptors and replication strategies , viruses have evolved to utilize alternative entry routes and fuse with distinct intracellular compartments . Preferential entry from early or late endosomes is achieved through adjusting the pH threshold for triggering fusion [2]–[4] or by relying on endosome-specific factors , such as lipids [5]–[7] or lysosomal enzymes [8] , [9] . There is evidence for complex regulation of early and late steps of viral fusion . For instance , viruses , which are activated by mildly acidic pH in early endosomes , may require late endosome-resident factors to complete their fusion process and release the nucleocapsid [5] , [7] . In other words , virus-endosome fusion and capsid release into the cytosol could be spatially and temporally separated . The asynchronous and often rate-limiting steps of virus internalization and trafficking hamper the studies of endosomal fusion and its regulation . In order to gain mechanistic insights into the virus-endosome fusion , it is essential to isolate the virus fusion step from the upstream asynchronous processes and to control the timing of low pH exposure and acidity of endosomal compartments . Avian Sarcoma and Leukosis virus ( ASLV ) initiates fusion via a two-step mechanism that involves priming of Env glycoprotein by cognate receptors ( presumably on the cell surface ) and low pH-dependent fusion with endosomes [3] , [10]–[12] . The receptor priming of Env confers the competence for acid-mediated refolding that drives the merger of viral and endosomal membranes . Subtype A ASLV infects cells expressing either of the alternative isoforms of the TVA receptor , TVA800 and TVA950 , which reside in lipid rafts and liquid-disordered domains , respectively [13]–[15] . Importantly , the transmembrane ( TVA950 ) and GPI-anchored ( TVA800 ) isoforms appear to direct the virus entry through distinct endocytic pathways [16] . ASLV fuses more efficiently with TVA950-expressing than with TVA800-expressing cells [17] . Considering that these isoforms have identical ectodomains [13] , [14] and exhibit a similar glycosylation pattern ( data not shown ) , the difference in fusion efficiency is likely due to the properties of sub-cellular compartments harboring the virus at the time of fusion . A critical feature of ASLV is the extreme stability of the Env-receptor complexes at neutral pH [11] , [16] , [18] , which enables the virus to survive the prolonged fusion block imposed upon raising the endosomal pH by NH4Cl . Removal of NH4Cl after several hours of incubation restores acidic conditions and quickly initiates infection [16] . This arrest/release protocol thus allows one to isolate virus-endosome fusion from the preceding internalization and trafficking steps [16] , [17] . Here , we employed the NH4Cl arrest/release protocol to synchronize the ASLV fusion . Removal of weak base resulted in an immediate and uniform pH drop in all intracellular compartments , thus standardizing the fusion trigger . Synchronized fusion facilitated the detection of viral core release from an endosome as a result of full enlargement of the fusion pore . We defined the sites of core release in cells expressing alternative receptors by ( i ) measuring the mobility of sub-viral particles after spatial separation from endosomes and ( ii ) probing the dynamics of pH changes within the recipient compartments upon removal of NH4Cl . These experiments showed that the viral cores were released into the cytosol of TVA950 cells . By contrast , the cores were delivered into endosomal compartments of TVA800 cells , from which they entered into the cytoplasm after an additional slow temperature-dependent step . Our findings imply that ASLV fusion is redirected to distinct intracellular compartments of cells expressing alternative receptors in the presence of NH4Cl . This approach provides new insights into the morphology and fusion-permissiveness of these compartments .
Removal of ammonium chloride results in acidification of all intracellular compartments , including the cytoplasm [19] . This effect is due to protons left behind by NH4+ , which traverses a membrane in its neutral NH3 form ( Figure S1A ) . Here , we synchronized ASLV fusion by allowing virus internalization in the presence of NH4Cl followed by removal of the fusion block . A quick drop of endosomal pH caused by the NH4Cl removal thus serves as a standardized trigger for ASLV-endosome fusion , irrespective of the distinct trafficking itineraries for this virus through TVA800 and TVA950 isoforms [16] , [17] . This “on demand” ASLV fusion with endosomal compartments offers several advantages for single particle imaging , since it ( i ) bypasses the slow pre-fusion steps and likely allows optimal priming of Env by receptors; ( ii ) eliminates the differences in the pH within distinct intracellular compartments; and ( iii ) simplifies virus tracking . In addition , the NH4Cl arrest redirects the virus entry to distinct intracellular compartments thus providing an opportunity to probe their relative permissiveness to viral fusion . As we will show below , although NH4Cl impairs endosomal trafficking and maturation [20]–[22] , these conditions provide valuable information regarding viral fusion with alternative sub-cellular sites and elucidate early post-fusion steps . We first assessed the effect of NH4Cl removal on endosomal pH by loading intracellular compartments with the endosomal pH indicator , pHrodo dextran [23] . In the presence of weak base , the endosomal pH exhibited a fairly narrow distribution around a neutral value ( Figure S1B ) . Immediately after NH4Cl removal by local perfusion with Hank's buffer ( HBSS ) , the endosomes became uniformly acidic ( Figure S1B , C ) . We next measured the pH changes within virus-carrying endosomes by co-labeling the viral envelope with the GFP-ICAM-1 chimera [17] and the membrane dye , DiD . Viral particles were identified based on the interior marker , MLV Gag-mKO [24] . Triple-labeled pseudoviruses were internalized by cells upon incubation in the presence of NH4Cl . Since the majority of ASLV is internalized by CV-1-derived target cells within 40 min [17] , we choose this time interval for NH4Cl pretreatment . Acidification of endosomal lumen upon removal of NH4Cl quenched the GFP fluorescence ( pKa 6 . 15 [25] ) , while the pH-independent DiD probe served as a reference signal for the ratiometric pH measurements ( Figure S1D-F ) . We found that endosomal pH exponentially decayed throughout the 2 min-perfusion with HBSS reaching 5 . 7±0 . 3 ( n = 20 ) in both TVA800 and TVA950 cells at the end of the pulse ( Figure S2A ) . In parallel experiments , the changes in the cytosolic pH were monitored by labeling the membrane of GFP-transfected CV-1 cells with DiD and measuring the ratio of the intracellular GFP and DiD fluorescence upon addition/removal of NH4Cl . As shown in Figure S2B , the cytosolic pH dropped to ∼6 . 8 at the beginning of HBSS perfusion , but nearly completely recovered before the cells were returned to NH4Cl . This likely occurs as a result of compensatory mechanisms that effectively raise the intracellular pH [19] , [26] , [27] . To conclude , the NH4Cl arrest/release protocol results in quick and uniform acidification of endosomes and the cytosol . This feature is essential for inducing fusion of internalized ASLV irrespective of the identity of virus-bearing endosomes in TVA800 and TVA950 cells . We have previously shown that labeling of pseudoviruses with MLV Gag-GFP permits visualization of single virus fusion in live cells [17] . Gag-GFP is cleaved into defined fragments upon virus maturation , including the GFP-tagged nucleocapsid protein . This fragment is loosely trapped within mature viruses and is readily released upon permeabilization of the viral membrane or as a result of fusion [28] . Our single virus fusion assay is thus based on the loss of GFP content from a viral particle . By contrast , the membrane marker , DiD , undergoes a limited dilution due to its redistribution into an endosomal membrane and thus retains its punctate appearance [17] , [29] . To synchronize the ASLV fusion , Gag-GFP and DiD labeled pseudoviruses were pre-bound to target cells in the cold and incubated at 37°C in the presence of NH4Cl . Cells were mounted on a heated microscope stage and imaged in an NH4Cl-containing buffer for a short period of time before removing the fusion block by perfusion with HBSS . Internalized ASLV particles that failed to undergo fusion with endosomes following the NH4Cl arrest/release protocol exhibited reversible GFP quenching-dequenching pattern ( Figure 1A–C , and video S1 ) reflecting changes in the intraviral pH ( Figure S2C ) . Three types of fusion-related events were induced by the arrest/release protocol in either TVA950 or TVA800 cells ( Table 1 ) . First , nearly 20% of virions fully released their content as a result of fusion ( Figure 1D , E and video S2 ) . Second , about the same fraction of viruses partially lost their content marker ( Figure 1F , G ) . A partial loss of viral content could be due to incomplete cleavage of the Gag-GFP precursor , which would remain in large oligomeric complexes , unable to permeate through a small fusion pore . Consistent with this notion , a similar fraction of viruses ( around 20% ) exhibited a partial loss of the GFP marker following the saponin-mediated lysis ( Figure S3 ) . Third , we observed a combination of a partial GFP release with subsequent separation of the GFP-tagged core from the viral/endosomal membrane labeled with DiD ( Table 1 and Figure 1A , double arrowhead ) . Irrespective of the TVA isoform , all three fusion-related events collectively accounted for ∼50% of viruses internalized in the presence of NH4Cl ( Table 1 ) . By comparison , 16% and 4% of viruses fused with TVA950 and TVA800 cells , respectively , through conventional entry in the absence of NH4Cl [17] . Thus , the arrest/release protocol markedly enhanced the efficiency of ASLV fusion , especially with TVA800 cells . The enhancing effect was likely due to several factors , including the quick acidification of endosomal lumen , and the ease of detection of synchronous fusion events within arrested endosomes . More importantly , the markedly improved probability of fusion could be due to a higher fusion-permissiveness of endosomes harboring the virus in the presence of NH4Cl compared to sites of uninterrupted ASLV fusion . As shown above , ∼20% of pseudoviruses released only a fraction of their GFP marker as a result of fusion ( Figure 1F , G ) . This was likely due to a partial cleavage of Gag-GFP during virus maturation . The presence of particles that contained both a releasable GFP maker and uncleaved Gag-GFP provides an opportunity to monitor the formation and enlargement of Env-mediated fusion pores , respectively . Indeed , a sizeable fraction of events were manifested in a diminution of the GFP signal upon perfusion with HBSS/NH4Cl followed by spatial separation of green and red puncta ( Figures 1A and 2A ) . These events occurred with similar frequencies in TVA950 and TVA800 cells: 13 . 9±5% ( n = 87 ) and 12 . 3±5% ( n = 86 ) , respectively ( Table 1 ) . Upon separation from endosomes , which retained the viral DiD marker , GFP-tagged puncta exhibited accelerated motion and traveled as one entity without losing fluorescence intensity ( Figure 2 , trajectories shown below the graphs , and videos S3 and S4 ) . These results further support the notion that green puncta are sub-viral particles ( SVPs ) containing unprocessed Gag-GFP . We analyzed the spatial separation events by tracking the GFP-labeled particles . The moment of spatial separation was defined based on the loss of the DiD signal which was initially co-localized with the GFP signal ( Figure 2 , blue asterisks ) . In contrast to the synchronized ASLV fusion , we very rarely observed spatial separation of core and membrane markers after a partial loss of a diffusible content marker upon the virus entry through a conventional route . This is likely due to the delayed release of the viral core , which could not be reliably detected because of the loss of the DiD signal ( data not shown ) . The disappearance of DiD marker after redistribution to an endosome probably occurs through its trafficking out of an endosome [17] , [29] . The lack of detectable capsid release through uninterrupted ASLV entry suggests that Env-mediated fusion pores enlarge slowly , so the capsids are released at later times after trafficking to different compartments . It appears that , in the presence of NH4Cl , ASLV is redirected to endosomes that support quick pore enlargement and core release . Regardless of whether or not the released SVPs corresponded to the bona fide viral capsid or to large oligomeric Gag-GFP complexes , their liberation signifies pore dilation and can thus be used to study this final step of fusion . Similar probabilities of the pore formation ( loss of diffusible GFP ) and release of SVPs in cells expressing TVA800 and TVA950 ( ∼13% , Table 1 ) indicate that , under our conditions , both receptor isoforms support equally efficient pore enlargement . We then asked whether the rate of pore enlargement was also independent of the receptor isoform . The time course of pore dilation was deduced from the lag between the onset of HBSS perfusion , which quickly triggered the pore opening , and spatial separation of SVPs from endosomes , as determined by the loss of the DiD signal from double-labeled particles ( Figure 2B and C ( blue asterisks ) , and videos S3 and S4 , respectively ) . The obtained lag times between the onset of HBSS perfusion and SVP release were plotted as cumulative probabilities over time ( Figure 2D ) . This analysis revealed that the SVPs were released ( and thus the pore enlarged ) faster in TVA950 cells compared to cells expressing the GPI-anchored receptor . Approximately 83% of these events were detected within the 2 min window of HBSS perfusion for TVA950 cells , whereas only 40% of cores were released into TVA800 cells within this interval . Together , our results imply that , whereas the arrest/release protocol promotes efficient opening and expansion of fusion pores in cells expressing either receptor isoform , the apparent rate of pore dilation is faster in TVA950 cells ( Figure 2D ) . The more efficient dilation of pores formed in TVA950 cells compared to TVA800 cells is consistent with our previous observation that early fusion pores formed by ASLV are larger in cells expressing TVA950 compared to TVA800 under normal entry conditions [17] . While dissociation of the viral core from the membrane could itself be a slow process [30] , the receptor isoform-dependence of the release kinetics argues against this possibility . On the other hand , we sometimes detected incomplete separation of GFP-tagged sub-viral particles from the viral membrane which could be due to partial pore dilation ( Figure S4 ) . Next we sought to define the cellular compartments into which the SVPs were released . We reasoned that the formation/enlargement of a fusion pore between the virus and the limiting membrane of an endosome should bring the intraviral pH to that of the cytosol . Since the cytosolic and endosomal/intraviral pH exhibit markedly different profiles following the removal of NH4Cl ( Figure S2 ) , the formation of a fusion pore with the limiting membrane of an endosome is expected to raise the intraviral pH . By contrast , if released cores remain inside an endosome ( see Discussion for the model ) , the core-associated GFP will be exposed to a continuously dropping pH throughout the HBSS pulse ( Figure 1B , C , pink lines ) . We found that SVPs released into TVA950 cells during the HBSS perfusion typically exhibited pH responses that were consistent with their release into the cytosol . The GFP signal initially dropped due to the acid-mediated quenching and the release of a diffusible fraction of this marker . However , green fluorescence partially recovered before the end of HBSS perfusion ( Figure 2B , pink lines ) , in spite of the continued decrease of endosomal pH ( Figures 2C , pink line , and S2A ) . The partial recovery of the GFP fluorescence occurred at the time or soon after spatial separation from an endosome and paralleled the increase in cytosolic pH ( Figure 2B , video S3 ) . These results imply that SVPs are released into the cytosol of TVA950 cells . In sharp contrast , spatial separation events in TVA800 cells were not accompanied by recovery of the GFP signal during the HBSS pulse ( Figure 2C , videoS4 ) , consistent with the continuous drop of endosomal pH ( pink line ) . Fluorescence partially recovered only after returning to NH4Cl . This pattern was observed for nearly all spatial separation events in TVA800 cells ( n = 18 , see also Figures S5 and S6 ) , suggesting that cores entered intracellular compartments distinct from the cytoplasm . It is worth pointing out that SVPs were not always released into the cytosol of TVA950 cells: in 3 out of 14 spatial separation events the GFP fluorescence decayed monotonously throughout the HBSS perfusion and partially recovered only after returning to NH4Cl ( exemplified in Figure S7 ) . In other words , the ASLV fusion in NH4Cl-arrested TVA950 cells did not uniformly result in the viral content release into the cytosol . As shown in Figure 2D , a fraction of SPVs were released after the end of the HBSS pulse . Since these delayed events took place under conditions when pH in all intracellular compartments was raised to neutrality by NH4Cl , the sites of core release could not be unambiguously established . To address this issue , a second HBSS pulse was applied . During the second pulse , the signal from SVPs in TVA950 cells exhibited a biphasic behavior similar to that of the cytosolic pH ( Figure 2E , n = 11 ) , which was consistent with the delivery of viral cores into cytoplasm . By comparison , the GFP signal from the late SVPs release events in TVA800 cells remained quenched throughout HBSS perfusion and increased only after returning to NH4Cl ( Figure 2F , n = 6 ) . Collectively , our results indicate that , in cells expressing the GPI-anchored TVA isoform , ASLV Env mediates the release of SPVs into compartments that lack the active buffering capacity of the cytoplasm . Viral cores that entered an endosome are expected to be restricted in their motion . We therefore compared the motion patterns of the newly release SVPs in cells expressing alternative receptor isoforms . We found that motion of virus-carrying endosomes in NH4Cl-arrested cells was restricted , as evidenced by a shallow slope of the mean square displacement ( MSD ) over time ( Figure 3A , B , red circles ) . However , coincident with the time of spatial separation ( blue asterisks ) , the initial MSD slope for green puncta drastically increased ( Figure 3A , B , green circles ) , whereas the movement of DiD-positive puncta ( host endosome ) continued to be restricted . For the liberated SVPs , the average slopes of log ( MSD ) vs . log ( time ) were 0 . 98±0 . 003 and 0 . 80±0 . 01 ( n = 15 ) in TVA950 and TVA800 cells , respectively ( Figure 3C , D ) . The MSD slopes close to 1 are indicative of free diffusion of SVPs released as a result of fusion [31] . To further characterize the motion patterns of these particles , we calculated the effective diffusion coefficients ( D ) from the MSD plots for endosomes , viruses trapped in endosomes , and for liberated sub-viral particles . This analysis confirmed that endosomes and viruses residing within endosomes tended to move slowly ( typical D below 0 . 1 µm2/sec , Figure 3E , in accordance to [32] ) . Only a fraction of Rab5-positive endosomes exhibited significant mobility . In contrast to endosomes , released SVPs were relatively more mobile . Interestingly , SVPs in TVA950 cells moved nearly twice as fast as in TVA800 cells: D = 0 . 30±0 . 1 ( n = 20 ) and 0 . 16±0 . 09 µm2/sec ( n = 20 ) , respectively ( P<0 . 0005 ) . The fact that movement of SVPs released into TVA800 cells was more restricted than in TVA950 cells is consistent with delivery of these particles into endosomal compartments . Analysis of individual particle trajectories confirmed this finding ( Figures S5 and S6 ) . On the other hand , a few SVPs that did not exhibit changes in the GFP signal paralleling the pH changes in the cytosol of TVA950 cells ( e . g . , Figure S7 ) also moved slowly , consistent with their release into endosomes ( Figure 3E , black circles ) . To conclude , single particle tracking further supported our conclusion that viral cores were released into distinct compartments within the NH4Cl-arrested TVA800 and TVA950 cells . How could the apparent lack of SVP release into the cytoplasm of TVA800 cells be rationalized ? Putative compartments harboring SVPs must be relatively large , as suggested by the disappearance of the GFP signal occurring as a result of significant dilution of the content marker . In agreement with this notion , in TVA800 cells the viral cores traveled considerable distances from endosomes with which they fused . Within a few minutes after SPV release the maximum distance between liberated SVPs and the endosome of origin reached 4 . 9±1 . 7 µm ( n = 10 ) and 11 . 3±4 . 2 µm ( n = 10 ) in TVA800 and TVA950 cells , respectively . The markedly different ( P<0 . 001 ) separation of SVPs and endosomes in these cells is another manifestation of the confined core movement in TVA800 cells . These findings are thus in line with the notion that SVPs are released into micron-size subcellular compartments of NH4Cl-arrested TVA800 cells . In order to test whether such unusually large compartments exist in CV-1 cells expressing either receptor , we transiently expressed markers for early ( CFP-Rab5 ) and late ( YFP-Rab7 ) endosomes in TVA800 and TVA950 cells . Figure 4 shows that , whereas these markers were not usually associated with large endosomes in untreated cells , large circular compartments reaching 3–4 µm in diameter became apparent in both cell lines following pre-incubation with NH4Cl . Live cell imaging revealed that large vacuoles started to form as early as after 10 min at 37°C in the presence of NH4Cl ( data not shown ) . It is therefore possible that small ASLV-carrying endosomes are somehow delivered into these abnormally large compartments in NH4Cl-arrested TVA800 , but not in TVA950 cells . To further test the possibility that ASLV is trafficked to different compartments of the two cell lines , we analyzed the virus co-localization with markers of early and late endosomes . Arresting the ASLV fusion by NH4Cl greatly simplifies this analysis , since viruses accumulate in distinct intracellular compartments in cells expressing alternative receptor isoforms . Cells were co-transfected with CFP-Rab5 and YFP-Rab7 and allowed to internalize pseudoviruses labeled with Gag-mKate2 ( red ) for 40 min in the presence of NH4Cl . Co-localization analysis revealed that 82% of ASLV pseudoviruses resided in Rab5-positive and mixed Rab5/Rab7-positive compartments in TVA950 cells ( Figure 5A , C ) . In contrast , the majority ( 63% ) of viruses internalized through TVA800 co-localized with Rab7 and only 25% were found in Rab5-positive endosomes ( Figure 5B , C ) . We also examined the spatial overlap between synchronized ASLV fusion and endosomal markers . ASLV fused with Rab5-positive endosomes of both TVA800 and TVA950 cells ( e . g . , Figure S8 ) . However , dynamic co-localization analysis of fusion with Rab7-positive compartments could not be reliably performed because of their clustering at the perinuclear space of NH4Cl-arrested cells ( Figure 5 ) . We therefore chose not to pursue these experiments . Taken together , the above results support the notion that the two receptor isoforms traffic ASLV into distinct intracellular compartments . Since the NH4Cl arrest-release protocol leads to efficient infection of cells expressing TVA800 ( [16] and data not shown ) , viral cores must eventually enter into the cytoplasm . The post-fusion steps resulting in capsid liberation into the cytoplasm are addressed below . If SVPs are released into endosomal compartments of TVA800 cells , an additional step would be required to deliver these particles into the cytoplasm and establish productive infection ( assuming that SVPs represent bona fide viral cores ) . To address this question , we pseudotyped the HIV-1 core carrying the beta-lactamase ( BlaM ) reporter enzyme with the ASLV Env and used these particles to measure virus-cell fusion [29] , [33] . Transfer of the viral core-incorporated BlaM into the cytosol as a result of fusion allows cleavage of a fluorogenic substrate sequestered in the cytoplasm . This substrate is thus inaccessible to the BlaM trapped within the viral particles that fail to fuse . ASLV pseudoviruses were internalized by TVA800 or TVA950 cells for 45 min at 37°C in the presence of NH4Cl . Fusion was initiated by transferring the cells into HBSS and was stopped after varied times , either by chilling on ice ( referred to as the temperature block , TB ) or by adding NH4Cl to re-neutralize the endosomal pH ( Figure 6A ) . Cells were loaded with the BlaM substrate and incubated overnight at 12°C to allow the substrate cleavage but prevent further fusion . When the NH4Cl block was not lifted ( t = 0 for the chase experiment ) , only a background level BlaM signal was detected ( Figure 6 ) . Consistent with our imaging results , even a very brief removal of NH4Cl allowed fusion to proceed to completion . Adding the NH4Cl back after a few minutes no longer reduced the fusion efficiency in either cell line ( Figure 6B , C ) . Similarly , the TB applied even after a brief removal of NH4Cl failed to block fusion in TVA950 cells ( Figure 6B ) . In sharp contrast , ASLV escape from low temperature after lifting the NH4Cl block was markedly delayed in TVA800 cells ( Figure 6C ) . This delayed resistance to low temperature that occurred with half-time ∼20 min was suggestive of an additional post-fusion step responsible for delivery of viral cores into the cytosol . It is thus conceivable that SVPs are first released into intracellular compartments in NH4Cl-arrested TVA800 cells and then enter into the cytosol through an additional Env-independent , temperature-dependent step .
We have previously found that ASLV fuses far less optimally with TVA800 cells compared to TVA950 cells when entering via a conventional route [17] . Because these receptor isoforms have identical ectodomains , it is unlikely that their Env binding or priming activities differ considerably . On the other hand , a recent study using cells expressing undetectably low levels of TVA800 and TVA950 arrived to the conclusion that more than one GPI-anchored receptor is required for ASLV infection , whereas a single transmembrane receptor is sufficient for productive entry [34] . However , this finding does not rule out the possibility that post-priming steps of entry determine the ASLV fusion efficiency and thus the minimal number of receptors supporting this process . We favor the model that the higher probability of fusion with TVA950 cells is due to ASLV trafficking to more permissive compartments than those in TVA800 cells [17] . This notion is consistent with data supporting ASLV trafficking to distinct intracellular compartments in cells expressing alternative TVA isoforms [16] , [17] . The differences in fusion-permissiveness of distinct endosomes suggest a role for host factors in regulating ASLV fusion and capsid release . Although NH4Cl impairs endosomal trafficking [20]–[22] , the arrest/release protocol employed in this study provides an unprecedented opportunity to redirect the ASLV fusion to different compartments and study this process under controlled conditions . This approach permitted: ( i ) synchronous and uniform acidification of all endosomes harboring the virus; ( ii ) isolation of low pH-dependent stages of fusion from upstream receptor-priming and virus trafficking; and ( iii ) enhancement of the viral core release from endosomes . A uniform acid load delivered upon NH4Cl removal allowed us to meaningfully compare the fusion-permissiveness of endosomes harboring the virus and to infer their overall topology . In summary , the NH4Cl block applied to cells expressing alternative TVA isoforms provided a critical means to direct the ASLV entry through distinct endosomes and gain new insights into early post-fusion processes . We are currently working on visualization of the viral core release upon uninterrupted ASLV entry through alternative TVA isoforms . Novel approaches introduced in this study enabled probing the sites of ASLV fusion . As a rule , sub-viral particles liberated in NH4Cl-arrested TVA950 cells sensed the cytosolic pH , consistent with the ASLV fusion with the limiting membrane of an endosome ( Figure 7 ) . In contrast , cores released into TVA800 cells experienced the pH responses characteristic of endosomal compartments lacking the active buffering ability of the cytosol . Subsequent core release into the cytosol was detected by the BlaM assay ( Figure 6 ) . The nature of this slow temperature-dependent post-fusion step is presently unknown . We propose the following two-step model for ASLV fusion with TVA800 cells following the NH4Cl arrest/release protocol . Since the viral membrane marker does not spread over a larger area as a result of fusion , it is likely that ASLV fuses with a small intralumenal vesicle . However , to account for disappearance of the viral content marker through a marked dilution for increased mobility of the released core , virus has to reside within a small vesicle ( Figure 7 ) . We propose that a small virus-carrying vesicle is in turn enclosed by a large intralumenal vesicle within an endosome . Multilamellar endosomes have been observed under physiological conditions and in the presence of NH4Cl [21] , [35]–[37] . This triple-vesicle enclosure is based on several considerations . First , the interior of an intralumenal vesicle is topologically equivalent to the cytoplasm , so that internalized virus cannot reside within that vesicle . Second , core liberation into the cytosol from the remaining double membrane enclosure can occur through back-fusion [5] ( Figure 7 ) , consistent with a slow temperature-dependent step following the low pH-mediated fusion with TVA800 cells ( Figure 6 ) . The proposed ASLV entry pathway is somewhat similar to that postulated for the Vesicular Stomatitis Virus ( VSV ) . VSV is thought to fuse “laterally” with an intralumenal vesicle ( Figure 7 , question mark ) and release its capsid through back-fusion with the limiting membrane of a late endosome [5] ( but see [38] for an opposite view ) . Our experimental data are difficult to reconcile with the VSV model . First , as pointed out above , fusion with small intralumenal vesicles should not result in sufficient dilution of the viral content to account for disappearance of the GFP signal ( Figure 7 ) . Second , the capsid release into a small vesicle will not allow detectable spatial separation from the viral envelope . Third , back-fusion of virus/vesicle hybrids should spill the viral membrane marker into a large area of the endosome's limiting membrane , an event that has been observed only once in our experiments ( data not shown ) . We therefore favor the “triple-enclosure” model presented above . It would be interesting to elucidate trafficking processes that entrapped the virus within such structure . Results presented in this study revealed the previously unappreciated aspects of viral entry through distinct pathways . Synchronous triggering and real-time visualization of ASLV fusion in cells expressing alternative receptors demonstrates that the identity of virus-harboring compartments determines the efficiency of fusion . Furthermore , the viral core release in NH4Cl-arrested TVA800 cells proceeds through a complex pathway that involves sequential pH-dependent and temperature-dependent steps . The latter post-fusion step ( s ) is unlikely to be mediated by ASLV Env . This provides strong evidence for the existence of back-fusion , which has been implicated in entry of several viruses [5] , [7] and of the anthrax toxin lethal factor [39] . Novel tools developed in this study can be used for probing the sites of fusion and capsid release of other viruses and to gain insights into vesicular trafficking and cargo sorting processes .
African green monkey kidney CV-1 and human embryonic kidney HEK293T/17 cell lines were obtained from ATCC ( Manassas , VA ) . HeLa-derived indicator TZM-bl cells expressing CD4 , CXCR4 and CCR5 ( donated by Drs . J . C . Kappes and X . Wu [40] were obtained from the NIH AIDS Research and Reference Reagent Program . CV-1-derived CV-1/TVA800 and CV-1/TVA950 cells expressing high levels of alternative TVA receptors have been described previously [17] . TZM-bl cells expressing TVA800 or TVA950 were obtained by transduction with VSV-G pseudotyped retroviral vectors pCMMP-TVA950 or pCMMP-TVA800 , as described previously [16] . Cells expressing high levels of either TVA receptor were sorted by flow cytometry using a FACS Aria II ( BD Biosciences , San Jose , CA ) after binding to the subgroup A ASLV SU-IgG fusion protein [16] and a goat anti-rabbit FITC-conjugated secondary antibody ( Sigma , St . Louis , MO ) . CV-1-derived cells were grown in Dulbecco's modified Eagle high glucose medium ( DMEM , Cellgro , Manassas , VA ) supplemented with 10% Cosmic Calf Serum ( HyClone Laboratories , Logan , UT ) and 100 U penicillin-streptomycin ( Gemini Bio-Products , West Sacramento , CA ) . TZM-bl cells were grown in high glucose DMEM supplemented with 10% Fetal Bovine Serum ( FBS , HyClone Laboratories , Logan , UT ) and 100 U penicillin-streptomycin . HEK 293T/17 cells were maintained in high glucose DMEM supplemented with 10% FBS , 100 U penicillin/streptomycin and 0 . 5 mg/ml G418 sulfate . The ASLV ( subgroup A ) Env-derived R99 peptide ( ∼95% purity by HPLC ) was synthesized by Macromolecular Resources ( Fort Collins , CO ) . The lipophilic dye DiD ( 1 , 1′-dioctadecyl-3 , 3 , 3′ , 3′-tetramethylindodicarbocyanine , 4-chlorobenzenesulfonate salt ) , Hoechst-33342 nuclear stain and pHrodo dextran were purchased from Invitrogen ( Carlsbad , CA ) . Ammonium chloride was purchased from Sigma . Hank's balanced salt solution with calcium and magnesium without phenol red ( HBSS ) and Phosphate-Buffered saline ( PBS ) were from Cellgro . Sodium pyruvate was from Hyclone . The subgroup A ASLV Env glycoprotein lacking the cytoplasmic domain ( designated EnvΔCT ) has been described previously [41] . The HIV-1 based packaging vector pR8ΔEnv lacking the env gene was from Dr . D . Trono ( University of Geneva , Switzerland ) . The pMM310 vector expressing BlaM-Vpr ( donated by Dr . M . Miller [42] ) was obtained from the NIH AIDS Research and Reference Reagent Program . Vectors expressing MLV Gag-pol , Gag-GFP , MLV LTR lacZ , pECFP-C1-Rab5 and pEYFP-C1-Rab7 [43] were a gift from Dr . W . Mothes ( Yale University ) . The MLV Gag-mKO expressing vector has been described previously [24] . The mRFP-Rab5 and mRFP-Rab7 expression vectors were obtained from Addgene ( Cambridge , MA ) . The construction of GFP-ICAM-1 was done as follows . The eGFP gene was amplified using KOD Xtreme DNA polymerase ( Novagen ) and the following primers: 5′-TAAGCTTCTCGAG GTGAGCAAGGGCGAGGAGCTGTTC-3′ ( forward ) and 5′-TGAATTCTT CTTGTACAGCTCGTCCATGCCGAGAGTG-3′ ( reverse ) . The amplified fragment was cloned into pCR4blunt-topo vectors using TOPO cloning kit ( Invitrogen ) . After the verification of the sequence of the gene of interest , the EcpH sequence in the EcpH-ICAM-1 vector [29] was replaced with the eGFP fragment using HindIII and EcoRI restriction sites ( italicized regions in the primer sequences ) . The pMLV-Gag-mKate2 plasmid was constructed by amplifying the mKate2 gene by PCR using KOD Xtreme DNA polymerase ( Novagen ) and the following forward and reverse primers: 5′-ATTGCGGATCCGGCGGCGGTGGAGCTAGCGTGAGCGAGCTGATTAAGGAGAAC-3′ and 5′-TTCCGCCGGCTTAGATATCTCTGTGCCCCAGTTTGCTAGGGAG-3′ , which contained the BamHI and NaeI restriction cleavage sites , respectively . The amplified fragment was first clone into the pCR4blunt-topo vector , using TOPO cloning kit ( Invitrogen ) , and its sequence was verified . The YFP gene in pMLV-Gag-YFP vector [44] was replaced with the mKate2 sequence in pCR4blunt-topo by restriction digestion with BamHI ( New England Biolab ) and NaeI ( New England Biolab ) and ligation with T4 DNA ligase ( New England Biolab ) . Fluorescently labeled pseudoviruses were produced in HEK293T/17 cells using PolyFect Transfection reagent ( Qiagen , Valencia , CA ) . Cells grown on a 10 cm dish were transfected with 2 µg MLV-Gag-Pol , 1 µg MLV-Gag-GFP or MLV-Gag-mKO , 3 µg pMLV-LTR-LacZ and 3 µg of the cytoplasmic tail-truncated ASLV-A Env ( EnvΔCT ) . For ASLV Env-pseudotyped virus carrying a membrane-incorporated GFP , 2 µg GFP-ICAM-1 were added to the transfection DNA mixture . Twenty four hours post-transfection , cells were labeled with 10 µM DiD in Opti-MEM ( Invitrogen ) for 4 h in the CO2 incubator at 37°C , as described in [41] , washed , covered with 6 ml of fresh phenol red-free DMEM/10% FBS , and incubated for an additional 24 h . Virus-containing medium was collected 48 h post-transfection , passed through a 0 . 45 µM filter , aliquoted and stored at −80°C . The infectious titer was determined by a β-Gal assay in CV-1 cells expressing TVA800 , as described previously [29] , [45] . To produce pseudoviruses containing the β-lactamase-Vpr ( BlaM-Vpr ) , a 10 cm dish of HEK293T/17 cells was transfected with 2 µg pR8ΔEnv , 2 µg pMM310 vector expressing BlaM-Vpr , 1 µg pcRev , and 3 µg EnvΔCT expression vector , using PolyFect , as described above . The infectious titer was determined by β-Gal assay in TZM-bl cells expressing TVA950 . For transient expression of endosomal markers , CV-1 cells expressing TVA800 or TVA950 were grown to 80% confluency on glass-bottom 35 mm Petri dishes ( Mattek , Ashland , Massachusetts ) in phenol red-free DMEM . Cells were transfected with 0 . 5 µg of CFP-Rab5 and 0 . 5 µg YFP-Rab7 , using Nanofectin transfection reagent ( PAA Laboratories , Dartmouth , MA ) , and used for imaging 24 h post-transfection . EnvΔCT-pseudotyped pseudoviruses bearing the BlaM-Vpr chimera were bound to CV-1 cells expressing TVA800 or TVA950 in 96-well stripwell plates ( Corning , NY ) by centrifugation at 2095×g , 4°C for 30 min ( MOI = 1 ) . Unbound virus was washed off with ice-cold HBSS , and cells were incubated in isotonic HBSS/2% FBS ( pH 7 . 8 ) containing 70 mM NH4Cl were incubated for 45 min at 37°C . Fusion of internalized viruses trapped in endosomes was initiated by transferring the cells into HBSS/2% FBS supplemented with 50 µg/ml of Env-derived R99 peptide ( to prevent fusion of not internalized viruses ) for varied times and stopped either by chilling the cells on ice ( referred to as the temperature block , TB ) or by adding 70 mM NH4Cl containing 50 µg/ml R99 . The medium was removed from wells , and cells were loaded with fluorescent CCF4-AM substrate ( Invitrogen ) and incubated overnight at 12°C . Unless stated otherwise , imaging experiments were performed using the Personal DeltaVision imaging system ( Applied Precision , Issaquah , WA ) equipped with an environmental enclosure that maintained the samples at 37°C and high relative humidity . Double- ( Gag-GFP/DiD ) and triple-labeled ( Gag-mKO/GFP-ICAM/DiD ) viruses and cellular proteins tagged with a monomeric red fluorescent protein ( mRFP-Rab5 and mRFP-Rab7 ) were imaged using an UPlanFluo 40×/1 . 3 NA oil objective ( Olympus ) and a standard DAPI/FITC/TRITC/Cy5 filter set ( Chroma , Bellows Falls , VT ) . The fluorescence emission was collected by an EM-CCD camera ( Photometrics ) . For double-labeled viruses , two consecutive images were collected every 1 . 5 sec for up to 10 min . Triple-labeled viruses were imaged every 2 . 2 seconds for the same period of time . In order to compensate for the axial drift during acquisition , we used the UltimateFocus feature ( Applied Precision ) , which automatically compensates for changes in the coverslip position . CV-1 cells expressing either TVA950 or TVA800 receptor were grown to near confluency on glass-bottom 35 mm Petri dishes ( Mattek ) in phenol red-free growth medium . Cells were placed on ice , washed with cold HBSS , and centrifuged with ∼1 . 5·104 IU of viruses in 100 µl HBSS/2% FBS at 2 , 100×g ( 4°C ) for 20 min . Unbound viruses were removed by washing with cold HBSS , and cells were incubated in isotonic HBSS/2% FBS supplemented with 70 mM NH4Cl ( pH = 7 . 8 ) for 40 min at 37°C to allow virus uptake . For imaging , cells were transferred into serum-free HBSS containing 70 mM NH4Cl and mounted on the microscope stage maintained at 37°C . Solutions around the cells in an image field were changed using a 4-channel miniature perfusion system ( Bioscience Tools , San Diego , California ) controlled through the Worx imaging software . Solutions were applied locally through a 100 µm plastic tip positioned close to the imaged cells , as described in [29] . The following two pre-warmed solutions were applied to cells: HBSS supplemented with 70 mM NH4Cl ( pH = 7 . 8 ) and plain HBSS ( pH = 7 . 2 ) . At the end of imaging experiments , cells were returned to the NH4Cl-containing buffer . Spot-enhancing filter 2D plugin from ImageJ [46] was applied to background-subtracted images to improve the signal to noise ratio . Virus tracking was performed with the 64-bit software module from Imaris ( BitPlane , Zurich , Switzerland ) , using an auto-regressive algorithm . Tracking provided quantitative information regarding the mean fluorescence intensities of the viral content and membrane markers , particle's instantaneous velocity , trajectory and the mean square displacements ( MSD ) . The two-color kymographs were obtained , using the Volume Viewer plugin from ImageJ . The generalized two-dimensional diffusion coefficients ( D ) were obtained from the y-axis intercepts ( y0 ) for the MSD logarithmic plot using the equation , as described in [47]: ( 1 ) Only the linear part of MSD following the core release was used to calculate D . The linear MSD regime typically ranged from 10 to 25 seconds after core release . For co-localization analysis , images were acquired after 40 min incubation with 70 mM NH4Cl at 37°C ( unless stated otherwise ) , using the Zeiss LSM780 microscope ( Carl Zeiss Microscopy ) , and a 63×/1 . 2 NA oil immersion objective . Between 25 and 30 confocal Z sections were acquired ( 0 . 3 µm apart ) using a pinhole set at 1 Airy unit . Spectral unmixing to correct for bleed-through between CFP-Rab5/YFP-Rab7 and 3D image reconstruction were carried out with the Zen and Imaris software , respectively . Image analysis for co-localization was performed after subtracting the background , selecting a meaningful range of apparent particle sizes ( between 0 . 5 and 3 µm ) and comparing the line histograms of CFP-Rab5 , YFP-Rab7 and Gag-mKate2 intensity distributions . A viral particle was considered to be co-localized with an endosomal marker , if the respective peaks of line histograms coincided ( i . e . , an endosomal marker exhibited a local fluorescence maximum coinciding with that of a viral particle ) .
|
Endosomal trafficking and regulation of retrovirus fusion is poorly understood , due in part to heterogeneity of viral particles and their asynchronous entry into an endocytic pathway . Here , we used an avian retrovirus that enters host cells in a receptor- and low pH-dependent manner . This feature allowed capturing the virus in intracellular compartments through raising the endosomal pH . Virus fusion was synchronously initiated upon permitting endosome acidification and visualized in real-time by single particle imaging . We found that different receptor isoforms directed virus into distinct sub-cellular compartments supporting quick release of the viral core . Through tracking individual sub-viral particles released from endosomes , we found that the full length receptor mediated core delivery into the cytoplasm . By contrast , fusion mediated by the GPI-anchored receptor released the core into another endosomal compartment , from which the core entered the cytosol through an additional temperature-dependent step . These findings demonstrate different permissiveness of endosomal compartments to viral fusion and the existence of a post-fusion step leading to the cytosolic release of cores and initiation of infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology"
] |
2012
|
Synchronized Retrovirus Fusion in Cells Expressing Alternative Receptor Isoforms Releases the Viral Core into Distinct Sub-cellular Compartments
|
DNA:RNA hybrid formation is emerging as a significant cause of genome instability in biological systems ranging from bacteria to mammals . Here we describe the genome-wide distribution of DNA:RNA hybrid prone loci in Saccharomyces cerevisiae by DNA:RNA immunoprecipitation ( DRIP ) followed by hybridization on tiling microarray . These profiles show that DNA:RNA hybrids preferentially accumulated at rDNA , Ty1 and Ty2 transposons , telomeric repeat regions and a subset of open reading frames ( ORFs ) . The latter are generally highly transcribed and have high GC content . Interestingly , significant DNA:RNA hybrid enrichment was also detected at genes associated with antisense transcripts . The expression of antisense-associated genes was also significantly altered upon overexpression of RNase H , which degrades the RNA in hybrids . Finally , we uncover mutant-specific differences in the DRIP profiles of a Sen1 helicase mutant , RNase H deletion mutant and Hpr1 THO complex mutant compared to wild type , suggesting different roles for these proteins in DNA:RNA hybrid biology . Our profiles of DNA:RNA hybrid prone loci provide a resource for understanding the properties of hybrid-forming regions in vivo , extend our knowledge of hybrid-mitigating enzymes , and contribute to models of antisense-mediated gene regulation . A summary of this paper was presented at the 26th International Conference on Yeast Genetics and Molecular Biology , August 2013 .
Elevated DNA:RNA hybrid formation due to defects in RNA processing pathways leads to genome instability and replication stress across species [1]–[7] . R loops threaten genome stability and often form under abnormal conditions where nascent mRNA is improperly processed or RNA half-life is increased , resulting in RNA that can hybridize with template DNA , displacing the non-transcribed DNA strand [8] . A recent study also found that hybrid formation can occur in trans via Rad51-mediated DNA-RNA strand exchange [9] . Persistent R loops pose a major threat to genome stability through two mechanisms . First , the exposed non-transcribed strand is susceptible to endogenous DNA damage due to the increased exposure of chemically reactive groups . The second , more widespread mechanism , identified in Escherichia coli , Saccharomyces cerevisiae , Caenorhabditis elegans and human cells , involves the R loops and associated stalled transcription complexes , which block DNA replication fork progression [3] , [4] , [8] , [10] , [11] . R loop-mediated instability is an area of great interest primarily because genome instability is considered an enabling characteristic of tumor formation [12] . Moreover , mutations in RNA splicing/processing factors are frequently found in human cancer , heritable diseases like Aicardi-Goutieres syndrome , and a degenerative ataxia associated with Senataxin mutations [13]–[17] . To avoid the deleterious effects of R loops , cells express enzymes for the removal of abnormally formed DNA:RNA hybrids . In S . cerevisiae , RNH1 and RNH201 , each encoding RNase H are responsible for one of the best characterized mechanisms for reducing R loop formation by enzymatically degrading the RNA in DNA:RNA hybrids [8] . Another extensively studied anti-hybrid factor is the THO/TREX complex which functions to suppress hybrid formation at the level of transcription termination and mRNA packaging [4] , [11] , [18] , [19] . In addition , the Senataxin helicase , yeast Sen1 , plays an important role in facilitating replication fork progress through transcribed regions and unwinding RNA in hybrids to mitigate R loop formation and RNA polymerase II transcription-associated genome instability [5] , [20] . Several additional anti-hybrid mechanisms have also been identified including topoisomerases and other RNA processing factors [2] , [6] , [7] , [9] , [21]–[23] . To add to the complexity of DNA:RNA hybrid management in the cell , hybrids also occur naturally and have important biological functions [24] . In human cells , R loop formation facilitates immunoglobulin class switching , protects against DNA methylation at CpG island promoters and plays a key role in pause site-dependent transcription termination [25]–[28] . Transcription of telomeres by RNA polymerase II also produces telomeric repeat-containing RNAs ( TERRA ) , which associate with telomeres and inhibit telomere elongation in a DNA:RNA hybrid-dependent fashion [29]–[31] . Noncoding ( nc ) RNA such as antisense transcripts , perform a regulatory role in the expression of sense transcripts that may involve R loops [32] . The proposed mechanisms of antisense transcription regulation are not clearly understood and involve different modes of action specific to each locus . Current models include chromatin modification resulting from antisense-associated transcription , antisense transcription modulation of transcription regulators , collision of sense and antisense transcription machineries and antisense transcripts expressed in trans interacting with the promoter for sense transcription [32]–[40] . More recently , studies in Arabidopsis thaliana found an antisense transcript that forms R loops , which can be differentially stabilized to modulate gene regulation [41] . Similarly , in mouse cells the stabilization of an R loop was shown to inhibit antisense transcription [42] . Here we describe , for the first time , a genome-wide profile of DNA:RNA hybrid prone loci in S . cerevisiae by DNA:RNA immunoprecipitation followed by hybridization on tiling microarrays ( DRIP-chip ) . We found that DNA:RNA hybrids occurred at highly transcribed regions in wild type cells , including some identified in previous studies . Remarkably , we observed that DNA:RNA hybrids were significantly associated with genes that have corresponding antisense transcripts , suggesting a role for hybrid formation at these loci in gene regulation . Consistently , we found that genes whose expression was altered by overexpression of RNase H were also significantly associated with antisense transcripts . A small-scale cytological screen found that diverse RNA processing mutants had increased hybrid formation and additional DRIP-chip studies revealed specific hybrid-site biases in the RNase H , Sen1 and THO complex subunit Hpr1 mutants . These genome-wide analyses enhance our understanding of DNA:RNA hybrid-forming regions in vivo , highlight the role of cellular RNA processing activities in suppressing hybrid formation , and implicate DNA:RNA hybrids in control of a subset of antisense regulated loci .
DNA:RNA hybrids have been previously immunoprecipitated at specific genomic sites such as rDNA , selected endogenous loci , and reporter constructs [2] , [5] . Subsequently , DRIP coupled with deep sequencing in human cells has demonstrated the prevalence of R loops at CpG island promoters with high GC skew [26] . To investigate the global profile of DNA:RNA hybrid prone loci in a tractable model , we performed genome-wide DRIP-chip analysis of wild type S . cerevisiae ( ArrayExpress E-MTAB-2388 ) using the S9 . 6 monoclonal antibody which specifically binds DNA:RNA hybrids , as characterized previously [43] , [44] . DRIP-chip profiles were generated in duplicate ( spearman's ρ = 0 . 78 when comparing each of over 2 million probes after normalization and data smoothing , Supplementary Figure S1 ) and normalized to a no antibody control . Overall , our DRIP-chip profiles identified several previously reported DNA:RNA hybrid prone sites including the rDNA locus and telomeric repeat regions ( Figure 1 , Supplementary Tables S1 , S2 ) [2] , [29]–[31] . DNA:RNA hybrids were also observed at 1217 open reading frames ( ORFs ) ( containing greater than 50% of probes above the threshold of 1 . 5 and found in both wild type replicates ) ( Supplementary Table S3 ) . These were generally shorter in length than average ( p = 4 . 29e−58 ) , highly transcribed ( Wilcoxon rank sum test p = 2 . 21e−6 ) , and had higher GC content ( p = 2 . 52e−50 ) ( Figure 2A , 2B and 2C , Supplementary Figure S2 ) . Importantly , despite the correlation between DNA:RNA hybrid association and transcriptional frequency , the wild type DRIP-chip profiles compared to the localization profile of the RNA polymerase II subunit Rpb3 revealed very low correlation ( ρ = 0 . 0097; [45] ) . This suggests that the DRIP-chip method was not unduly biased towards the short DNA:RNA hybrids that could theoretically have been captured within active transcription bubbles . Importantly , because genes with high GC content also have high transcriptional frequencies ( Supplementary Figure S3 ) , it is not clear from our findings whether GC content or transcriptional frequency contributed more to DNA:RNA hybrid forming potential . Furthermore , we observe that DNA:RNA hybrid prone loci do not encode for mRNA transcripts with particularly long half-lives ( Supplementary Figure S2D ) , suggesting that the act of transcription is vital to DNA:RNA hybrid formation and supporting the notion of co-transcriptional hybrid formation as the major source of endogenous DNA:RNA hybrids . Our data also revealed DNA:RNA hybrids highly associated with Ty1 and Ty2 subclasses of retrotransposons ( Figure 2E , Supplementary Table S4 ) . Consistent with our findings at ORFs , the levels of DNA:RNA hybrids correspond well with the known levels of expression of these elements . In general , Ty1 which constitutes one of the most abundant transcripts in the cell has the highest levels of DNA:RNA hybrids . Ty3 and Ty4 that are only slightly expressed have much lower levels of hybrids , and the lone Ty5 retrotransposon which is transcriptionally silent is not enriched for DNA:RNA hybrids ( Figure 2E ) ( [46]–[48] ) . In contrast to the trends observed with ORFs , GC content in retrotransposons is not highly correlated with the levels of expression , suggesting that expression is the main contributor to DNA:RNA hybrid formation . Specifically , Ty3 retrotransposons have the highest GC content but have only modest levels of expression and DNA:RNA hybrids . Certain DNA:RNA hybrid enriched regions identified by our DRIP-chip analysis such as rDNA and retrotransposons are associated with antisense transcripts [49] , [50] . Therefore , we checked if this was a common feature of DNA:RNA prone sites by comparing our list of DNA:RNA prone loci to a list of antisense-associated genes ( [51] ) . Because the expression of antisense-associated transcripts may be highly dependent on environmental conditions , we based our analysis on a list of transcripts identified in S288c yeast grown to mid-log phase in rich media which most closely mirrors the growth conditions of our cultures analyzed by DRIP-chip ( [51] ) . DNA:RNA hybrid enriched genes significantly overlapped with antisense-associated genes , suggesting that DNA:RNA hybrids may play a role in antisense transcript-mediated regulation of gene expression ( Fisher's exact test p = 1 . 03e−12 ) ( Figure 3A , 3B and 3C , Supplementary Table S5 ) . RNase H overexpression reduces detectable levels of DNA:RNA hybrids in cytological screens and suppresses genomic instability associated with R loop formation presumably through the degradation of DNA:RNA hybrids [7] , [52] , [53] . To test for a potential role of DNA:RNA hybrids in antisense-mediated gene regulation , we performed gene expression microarray analysis of an RNase H overexpression strain compared to an empty vector control ( GEO GSE46652 ) . This identified genes that had increased mRNA levels ( upregulated n = 212 ) or decreased mRNA levels ( downregulated n = 88 ) as a result of RNase H overexpression . A significant portion of the genes with increased mRNA levels were antisense-associated ( Fisher exact test p = 2 . 9e−7 ) ( Figure 3D , Supplementary Table S5 ) and tended to have high GC content , similar to DNA:RNA hybrid enriched genes in wild type ( Supplementary Figure S4 ) . However , the genes with increased mRNA levels under RNase H overexpression and the antisense-associated genes enriched for DNA:RNA hybrids in our DRIP experiment both tended towards lower transcriptional frequencies ( Figure 3E ) . These findings suggest that antisense-associated DNA:RNA hybrids moderate the levels of gene expression . Indeed , genes that were both modulated by RNase H overexpression and enriched for DNA:RNA hybrids were all found to be antisense-associated ( Figure 3F ) . The mechanism underlying altered gene expression in cells overexpressing RNase H remains unclear . While the association with antisense transcription is compelling , alternative models exist . One possibility is that the stress of RNase H overexpression triggers gene expression programs that coincidentally are antisense regulated . We analyzed gene ontology ( GO ) terms enriched among genes whose expression was changed by RNase H overexpression . Consistent with previous work , genes for iron uptake and incorporation were strongly activated by RNase H overexpression ( p = 2 . 21e−12 ) ( Figure 4A , Supplementary Table S6 ) and several of these iron transport genes ( i . e . FRE4 , FRE2 , FRE3 , FET3 , FET4 ) are antisense-associated ( [51] , [54] ) suggesting that overexpression of RNase H activates transcription of these genes by perturbing antisense-mediated regulation . Alternatively , changes in RNase H levels may increase the cellular iron requirements since sensitivity to low iron concentration is associated with DNA damage and repair [55] . To test this alternative hypothesis , we tested the RNase H deletion and sen1-1 mutants for sensitivity to low iron conditions compared to a fet3Δ positive control ( Figure 4B ) . The sen1-1 mutant , RNase H depletion or overexpression did not induce sensitivity to low iron ruling out the possibility that the transcriptional response in cells overexpressing RNase H was a result of cellular iron requirement . Collectively , our DRIP-chip and microarray analysis suggest that DNA:RNA hybrids may be an important player in antisense-mediated gene regulation . Transcription-coupled DNA:RNA hybrids have been shown to accumulate in a diverse set of transcription and RNA processing mutants involved in a wide range of transcription related processes ( Table 1 ) . To gain a broader understanding of factors involved in R loop formation , we performed a cytological screen of RNA processing , transcription and chromatin modification mutants for DNA:RNA hybrids using the S9 . 6 antibody . Importantly , previous work in our lab has shown that all of the mutants screened exhibit chromosome instability ( CIN ) , which would be consistent with increased hybrid formation [53] . Significantly elevated hybrid levels were found in 22 of the 40 mutants tested compared to wild type , including a SUB2 mutant which has been previously linked to R loop formation ( Figure 5 , [4] ) . We also assayed some of the well-characterized R-loop forming mutants , RNase H , Sen1 and Hpr1 , as positive controls for elevated DNA:RNA hybrid levels ( Figure 5 ) . In our screen , we detected hybrids in mutants affecting several pathways linked to DNA:RNA hybrid formation such as transcription , nuclear export and the exosome ( Figure 5 , Table 1 ) . Consistent with findings in metazoan cells , we also observed hybrid formation in some splicing mutants ( Figure 5 , Table 1; [56] ) . Several rRNA processing mutants were enriched for DNA:RNA hybrids ( 7 out of the 22 positive hits ) , likely due to DNA:RNA hybrid accumulation at rDNA genes , a sensitized hybrid formation site ( Figure 1; [2] ) . It is possible that , as seen in mRNA cleavage and polyadenylation mutants , DNA:RNA hybrid formation may contribute to their CIN phenotypes [6] . Currently , there are 52 yeast genes whose disruptions have been found to lead to DNA:RNA hybrid accumulation , 21 of which were newly identified by our screen ( Table 1 ) . The success of this small-scale screen suggests that most RNA processing pathways suppress hybrid formation to some degree and that many DNA:RNA hybrid forming mutants remain undiscovered . To better understand the mechanism by which cells regulate DNA:RNA hybrids , we performed DRIP-chip analysis of rnh1Δrnh201Δ , hpr1Δ , and sen1-1 mutants in order to determine if these contribute differentially to the DNA:RNA hybrid genomic profile . The rnh1Δrnh201Δ , hpr1Δ , and sen1-1 mutants are particularly interesting because they have well established roles in the regulation of transcription dependent DNA:RNA hybrid formation . Our DRIP-chip profiles revealed that , similar to wild type profiles , the mutant profiles were enriched for DNA:RNA hybrids at rDNA , telomeres , and retrotransposons ( Figure 6 , Supplementary Tables S1 , S2 , S3 ) . The rnh1Δrnh201Δ , hpr1Δ , and sen1-1 mutants also exhibited DNA:RNA hybrid enrichment in 1206 , 1490 and 1424 ORFs respectively compared to the 1217 DNA:RNA hybrid enriched ORFs identified in wild type ( Supplementary Table S4 ) . Interestingly , in addition to the similarities described above , our profiles also identified differential effects of the mutants on the levels of DNA:RNA hybrids . In particular , we observed that deletion of HPR1 resulted in higher levels of DNA:RNA hybrids along the length of most ORFs with a preference for longer genes compared to wild type ( Figure 7A , 7B and 7C ) . This observation is consistent with Hpr1's role in bridging transcription elongation to mRNA export and its localization at actively transcribed genes ( [4] , [57]–[59] ) . In contrast , mutating SEN1 resulted in higher levels of DNA:RNA hybrids at shorter genes ( Figure 7A and 7B ) , which is consistent with Sen1's role in transcription termination particularly for short protein-coding genes ( [5] , [60] , [61] ) . The rnh1Δrnh201Δ mutant revealed higher levels of DNA:RNA hybrids at highly transcribed and longer genes ( Figure 7A and 7B ) which is supported by a wealth of evidence of RNase H's role in suppressing R loops in long genes to prevent collisions between transcription and replication machineries ( [8] , [62] ) . Further inspection of our profiles also revealed that rnh1Δrnh201Δ and sen1-1 mutants but not the hpr1Δ mutant had increased DNA:RNA hybrids at tRNA genes ( two tailed unpaired Wilcox test p = 1 . 56e−19 in the rnh1Δrnh201Δ mutant and 1 . 68e−15 in the sen1-1 mutant ) ( Figure 8A , 8B and 8C , Supplementary Table S7 ) and this was confirmed by DRIP-quantitative PCR ( qPCR ) of two tRNA genes in wild type and rnh1Δrnh201Δ ( Supplementary Figure S5 ) . Because tRNAs are transcribed by RNA polymerase III , this observation indicates that Hpr1 is primarily involved in the regulation of RNA polymerase II specific DNA:RNA hybrids while RNase H and Sen1 have roles in a wider range of transcripts . Mutation of SEN1 also led to increased levels DNA:RNA hybrids at snoRNA ( two tailed unpaired Wilcox test p = 1 . 81e−6 ) ( Figure 8D , 8E and 8F , Supplementary Table S8 ) consistent with its role in 3′ end processing of snoRNAs ( [63] ) .
Identifying the landscape of genomic loci predisposed to DNA:RNA hybrids is of fundamental importance to delineating mechanisms of hybrid formation and the contributions of various cellular pathways . Although our profiles depend on the specificity of the anti-DNA:RNA hybrid S9 . 6 monoclonal antibody , this aspect has been well characterized [44] and several of our observations are consistent with what has been reported in the literature . Locus specific tests showed that DNA:RNA hybrids occur more frequently at genes with high transcriptional frequency and GC content [4] , [5] , [18] . Moreover , in rnh201Δ cells , there is an inverse relationship between GC content and gene expression levels , suggesting that DNA:RNA hybrids accumulate at regions of high GC content and block transcription in the absence of RNase H [64] . Our work extends the knowledge of DNA:RNA hybrids from a few locus-specific observations to show that , in wild type , there are potentially hundreds of hybrid prone genes that tend to be shorter in length , frequently transcribed and high in GC content [2] , [4] , [56] . The latter is consistent with recent studies in human cells that demonstrated that genomic regions with high GC skew are prone to R loop formation , which plays a regulatory role in DNA methylation [26] , [27] . However , while we determined the relationship between GC content and DNA:RNA hybrid formation , we were unable to do the same analysis for GC skew , likely due to the low level of GC skew and lack of DNA methylation in Saccharomyces . This is unsurprising since the best characterized functional element associated with GC skew , CpG island promoters [26] , , are not found in yeast . Importantly , our findings at retrotransposons support the notion that expression levels and not GC content contribute more to DNA:RNA hybrid forming potential . Additionally , DRIP-chip analysis of wild type cells identified hybrid enrichment at rDNA , retrotransposons , and telomeric regions . Along with previous studies , our DRIP-chip analysis confirms that rDNA is a hybrid prone genomic site and suggests that many factors of rRNA processing and ribosome assembly suppress potentially damaging rDNA:rRNA hybrid formation [2] , [7] . The presence of TERRA-DNA hybrids at telomeres is supported by our observation of significant hybrid signal at telomeric repeat regions across all DRIP-chip experiments . The DRIP-chip dataset is a resource for future studies seeking to elucidate the localization of DNA:RNA hybrids across antisense-associated regions and the impact of DNA:RNA hybrid removal on genome-wide transcription . We observed that genes associated with antisense transcripts were significantly enriched for DNA:RNA hybrids and modulated at the transcript level by RNase H overexpression . Antisense regulation has been reported at mammalian rDNA and yeast Ty1 retrotransposons , loci that were also enriched for DNA:RNA hybrids in our DRIP-chip [49] , [50] . The role of DNA:RNA hybrids and RNase H in antisense regulation is currently unclear . However , there are several non-exclusive models of antisense gene regulation . One model proposes that the physical presence of the antisense transcripts is crucial to antisense gene regulation . For instance , trans-acting antisense transcripts have been shown to control Ty1 retrotransposon transcription , reverse transcription and retrotransposition [65] . Another study has further shown that trans-acting antisense transcripts that only overlap with the sense strand promoter can block sense transcription , potentially by hybridizing with the non-template DNA strand [33] . These suggest that antisense transcription in cis is not necessary as long as the antisense transcript is present . It is possible that DNA:RNA hybrids may be formed by the antisense or the sense transcript with genomic DNA . Moreover , DNA:RNA hybrids may play a functional role in antisense transcription regulation as shown by antisense-associated genes both enriched for DNA:RNA hybrids and affected transcriptionally by RNase H overexpression . Experiments comparing the ratio of antisense versus sense transcripts and determining the amount of DNA:RNA hybrid formation by either transcript under conditions known to regulate the particular gene will further elucidate the role of RNase H and DNA:RNA hybrids in antisense regulation . Our investigation of mutant-specific DNA:RNA hybrid formation sites is consistent with the existing literature on Hpr1 , Sen1 and RNase H . Significantly , the hpr1Δ and rnh1Δrnh201Δ mutants exhibited increased DNA:RNA hybrid levels along the length of long genes , while the sen1-1 mutant exhibited increased DNA:RNA hybrid levels along the length of short genes ( Figure 7A ) . This coheres with Hpr1's function in transcription elongation and mRNA export , and RNase H's role in preventing transcription apparatus and replication fork collisions , which carry greater consequence for long genes ( [4] , [57]–[59] , [62] ) . In contrast , Sen1 is particularly important for transcription termination at short genes ( [61] ) . In addition , the RNase H deletion and sen1-1 mutants had increased hybrids at tRNA genes , suggesting that they are both required to prevent tRNA:DNA hybrid accumulation . Interestingly , a recent study found that the mRNA levels of genes encoding RNA polymerase III and proteins that modify tRNA are increased in an rnh1Δrnh201Δ mutant [64] , which may be in response to a lack of properly processed tRNA transcripts . The finding that both tRNA and snoRNA genes were enriched for hybrids in sen1-1 highlights the role of Sen1 in RNA polymerase I , II and III transcription termination and transcript maturation [60] , [63] , [66] . More broadly , our data and the literature support the notion that transcripts from RNA polymerases I , II and III can be subject to DNA:RNA hybrid formation especially in RNA processing mutant backgrounds . Factors regulating ectopic , genome destabilizing DNA:RNA hybrids are best characterized in yeast , although less is known about the functions of native R loop structures . The genome-wide maps of DNA:RNA hybrids presented here recapitulate the known sites of hybrid formation but also add important new insights to potential functions of R loops . Most importantly , we demonstrate the usefulness of DRIP profiling for detecting biologically meaningful differences in mutant strains . Therefore , DRIP profiling of yeast genomes in various mutant backgrounds will be key to understanding the causes and consequences of inappropriate R loop formation and how these are modulated by other cellular pathways .
All strains are listed in Supplementary Table S9 . For RNase H overexpression experiments , recombinant human RNase H1 was expressed from plasmid p425-GPD-RNase H1 ( 2μ , LEU2 , GPDpr-RNase H1 ) and compared to an empty control plasmid p425-GPD ( 2μ , LEU2 , GPDpr ) [7] . Briefly , cells were grown overnight , diluted to 0 . 15 OD600 and grown to 0 . 7 OD600 . Crosslinking was done with 1% formaldehyde for 20 minutes . Chromatin was purified as described previously [67] and sonicated to yield approximately 500 bp fragments . 40 µg of the anti-DNA:RNA hybrid monoclonal mouse antibody S9 . 6 ( gift from Stephen Leppla ) was coupled to 60 µL of protein A magnetic beads ( Invitrogen ) . For ChIP-qPCR , crosslinking reversal and DNA purification were followed by qPCR analysis of the immunoprecipitated and input DNA . DNA was analyzed using a Rotor-Gene 600 ( Corbett Research ) and PerfeCTa SYBR green FastMix ( Quanta Biosciences ) . Samples were analyzed in triplicate on three independent DRIP samples for wild type and rnh1Δrnh201Δ . Primers are listed in Supplementary Table S11 . For DRIP-chip , precipitated DNA was amplified via two rounds of T7 RNA polymerase amplification ( [68] ) , biotin labeled and hybridized to Affymetrix 1 . 0R S . cerevisiae microarrays . Samples were normalized to a no antibody control sample ( mock ) using the rMAT software and relative occupancy scores were calculated for all probes using a 300 bp sliding window . All profiles were generated in duplicate and replicates were quantile normalized and averaged . Spearman correlation scores between replicates are listed in Supplementary Table S10 . Coordinates of enriched regions are available in Dataset S1/S2/S3/S4/S5/S6/S7/S8 . DRIP-chip data is available at ArrayExpress E-MTAB-2388 . Enriched features had at least 50% of the probes contained in the feature above the threshold of 1 . 5 . Only features enriched in both replicates were reported . Transcriptional frequency [69] , GC content ( [70] ) and gene length were compared using the Wilcoxon rank sum test . Antisense association was analyzed by the Fisher's exact test using R . Statistical analysis of genomic feature enrichment was performed using a Monte Carlo simulation , which randomly generates start positions for the particular set of features and calculates the proportion of that feature that would be enriched in a given DRIP-chip profile if the feature were distributed at random [67] . 500 simulations were run per feature for each DRIP-chip replicate to obtain mean and standard deviation values . These values were used to calculate the cumulative probability ( P ) on a normal distribution of seeing a score lower than the observed value by chance . CHROMATRA plots were generated as described previously ( [71] ) . Relative occupancy scores for each transcript were binned into segments of 150 bp . Transcripts were sorted by their length , transcriptional frequency or GC content and aligned by their Transcription Start Sites ( TSS ) . For transcriptional frequency transcripts were grouped into five classes according to their transcriptional frequency described by Holstege et al 1998 . For GC content transcripts were grouped into four classes according to their GC content obtained from BioMart ( [70] ) . Average gene , tRNA or snoRNA profiles were generated by averaging all the probes that were encompassed by the features of interest . For averaging ORFs , corresponding probes were split into 40 bins while 1500 bp of UTRs and their probes were split into 20 bins . For smaller features like tRNAs and snoRNAs corresponding probes were split into only 3 bins . Average enrichment scores were calculated using in house scripts that average the score of all the probes encompassed by the feature . Gene expression microarray data is available at GEO GSE46652 . Strains harboring the RNase H1 over-expression plasmid or empty vector were grown in SC-Leucine at 30°C . All profiles were generated in duplicate . Total RNA was isolated from 1 OD600 of yeast cells using a RiboPure Yeast kit ( A&B Applied Biosystems ) , amplified , labeled , fragmented using a Message-Amp III RNA Amplification Kit ( A&B Applied Biosystems ) and hybridized to a GeneChIP Yeast Genome 2 . 0 microarray using the GeneChip Hybridization , Wash , and Stain Kit ( Affymetrix ) . Arrays were scanned by the Gene Chip Scanner 3000 7G and expression data was extracted using Expression Console Software ( Affymetrix ) with the MAS5 . 0 statistical algorithm . All arrays were scaled to a median target intensity of 500 . A minimum cut off of p-value of 0 . 05 and signal strength of 100 across all samples were implemented and only transcripts that had over a 2-fold change in the RNase H over-expression strain compared to wild type were considered significant . The correlation between duplicate biological samples was: control ( r = 0 . 9955 ) , RNase H over-expression ( r = 0 . 9719 ) . For statistical analysis , GC content , transcription frequencies and antisense association were analyzed as for DRIP-chip analysis . Cells were grown to mid-log phase in YEPD rich media at 30°C and washed in spheroplasting solution ( 1 . 2 M sorbitol , 0 . 1 M potassium phosphate , 0 . 5 M MgCl2 , pH 7 ) and digested in spheroplasting solution with 10 mM DTT and 150 µg/mL Zymolase 20T at 37°C for 20 minutes similar to previously described ( [72] ) . The digestion was halted by addition of ice-cold stop solution ( 0 . 1 M MES , 1 M sorbital , 1 mM EDTA , 0 . 5 mM MgCl2 , pH 6 . 4 ) and spheroplasts were lysed with 1% vol/vol Lipsol and fixed on slides using 4% wt/vol paraformaldehyde/3 . 4% wt/vol sucrose ( [73] ) . Chromosome spread slides were incubated with the mouse monoclonal antibody S9 . 6 ( 1 µg/mL in blocking buffer of 5% BSA , 0 . 2% milk and 1× PBS ) . The slides were further incubated with a secondary Cy3-conjugated goat anti-mouse antibody ( Jackson Laboratories , #115-165-003 , diluted 1∶1000 in blocking buffer ) . For each replicate , at least 100 nuclei were visualized and manually counted to obtain the fraction with detectable DNA:RNA hybrids . Each mutant was assayed in triplicate . Mutants were compared to wild type by the Fisher's exact test . To correct for multiple hypothesis testing , we implemented a cut off of p<0 . 01 divided by the total number of mutants compared to wild type , meaning mutants with p<0 . 00024 were considered significantly different from wild type . 10-fold serial dilutions of each strain was spotted on 90 µM BPS plates with FeSO4 concentrations of 0 , 2 . 5 , 20 or 100 µM and grown at 30°C for 3 days [55] . A summary of this paper was presented at the 26th International Conference on Yeast Genetics and Molecular Biology , August 2013 [74] .
|
RNA processing factors are mutated in human cancers , inherited developmental disorders and neurodegenerative syndromes . Defects in RNA processing have been associated with increased levels of mutations and DNA damage in part via the formation of DNA:RNA hybrids . Although it is likely that specific regions of the genome are more prone to DNA:RNA hybrid formation , a map of hybrid-prone regions is not available . In this study , we describe the genome-wide distribution of DNA:RNA hybrids in both normal and mutant Saccharomyces cerevisiae cells . The resulting profiles contribute to both our understanding of the general properties of hybrid-forming loci and to our knowledge of hybrid-mitigating enzymes . Interestingly , significant DNA:RNA hybrid enrichment was detected at genes associated with antisense transcription . We show that overexpression of RNase H , which degrades the RNA in hybrids , significantly affects the expression of genes associated with antisense transcripts . These findings support a role for DNA:RNA hybrids in regulation of gene expression by antisense transcripts .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"model",
"organisms",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"molecular",
"cell",
"biology",
"research",
"and",
"analysis",
"methods"
] |
2014
|
Genome-Wide Profiling of Yeast DNA:RNA Hybrid Prone Sites with DRIP-Chip
|
Regulation and maintenance of protein synthesis are vital to all organisms and are thus key targets of attack and defense at the cellular level . Here , we mathematically analyze protein synthesis for its sensitivity to the inhibition of elongation factor EF-Tu and/or ribosomes in dependence of the system’s tRNA and codon compositions . We find that protein synthesis reacts ultrasensitively to a decrease in the elongation factor’s concentration for systems with an imbalance between codon usages and tRNA concentrations . For well-balanced tRNA/codon compositions , protein synthesis is impeded more effectively by the inhibition of ribosomes instead of EF-Tu . Our predictions are supported by re-evaluated experimental data as well as by independent computer simulations . Not only does the described ultrasensitivity render EF-Tu a distinguished target of protein synthesis inhibiting antibiotics . It may also enable persister cell formation mediated by toxin-antitoxin systems . The strong impact of the tRNA/codon composition provides a basis for tissue-specificities of disorders caused by mutations of human mitochondrial EF-Tu as well as for the potential use of EF-Tu targeting drugs for tissue-specific treatments .
Ribosomes and elongation factors EF-Tu are the most important targets of antibiotics inhibiting bacterial protein synthesis because of the crucial roles they play in this vital process [1 , 2] . Ribosomes are molecular machines that use the genetic information stored in messenger RNAs ( mRNAs ) to synthesize proteins . Elongation factors EF-Tu bind aminoacylated tRNAs and guanosine-5’-triphosphate ( GTP ) molecules to form ternary complexes , which deliver the aminoacyl-tRNAs to the ribosomes , see Fig 1 . A major fraction of antibiotics that interfere with protein synthesis is directed against the ribosome [3 , 4] , whereas a minor fraction is directed against EF-Tu [5] . In addition to compounds targeting either the ribosome or EF-Tu , antibiotics of the kirromycin and enacyloxin IIa families inhibit ribosomes and EF-Tu simultaneously by stalling the ternary complex on the ribosome [6–9] , see Table 1 . Fig 1 gives a schematic overview over these three distinct inhibition pathways . Although EF-Tu and the ribosome are both fundamental for bacterial mRNA translation , they are very distinct in terms of , e . g . , function , abundance , structure , or interactions with other components of the bacterial protein synthesis machinery . Therefore , it is not obvious under which conditions which of the three inhibition pathways depicted in Fig 1 is the most efficient one , i . e . , under which conditions EF-Tu is a more suitable target than the ribosome for down-regulation of bacterial protein synthesis . To address this question , it is necessary to understand how sensitively protein synthesis responds to a decline in the availability of EF-Tu and under which conditions it reacts more sensitively to the loss of functional EF-Tu than to the loss of functional ribosomes . Moreover , the sensitivity of protein synthesis to EF-Tu inhibition is not only an important aspect of drug effectiveness . In fact , it might also provide a further and so-far undescribed basis for an anti-drug defense mechanism called bacterial persistence , which is applied by bacteria when they suppress their own growth to defend themselves against antibiotic attacks . Bacterial infections can relapse or become chronic after antibiotic treatments . One major reason for this failure of antibiotics is bacterial persistence [10] . Persistent bacteria are able to tolerate exposure to antibiotics as well as other negative influences from the environment . In contrast to antibiotic resistant cells , persisters are genetically identical to drug-sensitive individuals of the same population . The persistence arises from stochastic phenotypic transitions resulting in strongly reduced growth rates [11] . The switching between a fast-growing and a dormant cell state can be mediated by different toxin-antitoxin systems , as recently reviewed in [12] . A toxin-antitoxin system consists of two components , a growth-inhibiting toxin and an antagonistic antitoxin . When the antitoxin gets degraded , the toxin can fully develop its growth-arresting effect , i . e . , persistence is induced and the cell is protected from the adverse effects of antibiotics . Toxin-antitoxin mediated phenotype transitions from the fast-growing to the dormant phenotype and vice versa were found to be caused by stochastic fluctuations of the abundance of free toxin above and below a certain threshold [13] , where the switching rates need to be fast enough . If the switching rates are too low , the fraction of persister cells is too small to guarantee survival of the population under stress conditions . In addition , the persisters die before regaining full growth because the continuous impact of the toxin leads to cell death [14] . Recently , the phd/doc toxin–antitoxin system was discussed in the context of persister cell formation [15] . The antitoxin Phd inactivates the toxin Doc that in turn was found to inhibit elongation factor EF-Tu by phosphorylating it at position Thr382 [16 , 17] . Phosphorylation of EF-Tu at position Thr382 suppresses ternary complex formation and , thus , protein synthesis [17 , 18] . Therefore , in principle the phd/doc toxin–antitoxin system can facilitate transitions between a persistent phenotype with strongly reduced protein synthesis and a fast-growing phenotype with maximal protein synthesis . EF-Tu is one of the most abundant proteins in bacteria to compensate for limited diffusion caused by molecular crowding [19] . It is unlikely that a major fraction of the EF-Tu molecules gets phosphorylated at the same time . This brings up the question: Does bacterial protein synthesis respond indeed in a highly sensitive manner to EF-Tu inhibition , rendering Doc an efficient toxin despite the vast cellular abundance of its target and making phd/doc-mediated phenotypic transitions fast enough to enable persister formation and survival ? To assess the sensitivity of protein synthesis to EF-Tu inhibition and answer these questions , we apply a previously published computational framework of in-vivo-like bacterial protein synthesis [27 , 28] that was recently further validated by experiment [29] . In particular , we study the effect of variations of the EF-Tu concentration on the translational state of a cell . We compare the behavior of our computational in-vivo-like translation system both to re-evaluated published experimental data for different E . coli strains [30] and to highly simplified artificial bacterial translation systems based on only one or two codons and their cognate tRNAs . We conclude that imbalances between tRNA abundances and codon usages lead to an ultrasensitive dependence of cellular protein synthesis on EF-Tu concentration . We confirm these findings by computer simulations of the in-vitro synthesis of fMetLysHis tripeptides using the “PURE system simulator” [31] and compare the effects of EF-Tu and ribosome inhibition on the synthesis rate .
The simplest translation system consists of one codon and one tRNA that is cognate to the codon . Experimentally , such a one-codon-one-tRNA ( 1C-1T ) model is realized by a cell-free ( in-vitro ) expression system containing for example only poly-U mRNA and Phe-tRNAPhe . In S1 Text , all details on the system of equations describing the 1C-1T model can be found . For such a 1C-1T translation system , the dependence of the overall elongation rate on the EF-Tu concentration follows a Michaelis-Menten-like behavior: At first , the overall elongation rate increases linearly with increasing EF-Tu concentration and then levels off once it has reached a certain saturation value , see Fig 3A ) . We conclude that a 1C-1T translation system is not ultrasensitive to the abundance of EF-Tu . Instead , at lower values the overall elongation rate is approximately proportional to the EF-Tu concentration; in contrast to in-vivo translation , which requires a substantial amount of EF-Tu for significant protein synthesis , see Fig 2 . Because 1C-1T translation cannot explain the efficient regulation of protein synthesis via EF-Tu as observed in E . coli , we slightly increase the complexity of our computational framework by a second codon and a second tRNA , thereby introducing a two-codon-two-tRNA ( 2C-2T ) translation model . Both tRNAs are assumed to be cognate to one of the two codons , but near-cognate to the other: in particular , tRNA 1 is cognate to codon 1 and near-cognate to codon 2 , and vice versa . Codons 1 and 2 appear with normalized codon usages p1 and p2 , respectively , with p1 + p2 = 1 . The corresponding tRNAs 1 and 2 have molar concentrations X1 and X2 , respectively . As an example , a 2C-2T in-vitro translation system would consist of Phe-tRNAGAA Phe , Cys-tRNAGCA Cys , and mRNAs consisting only of UUC and UGC codons . The system of equations describing 2C-2T translation is given with all details in S1 Text . Because this system of equations has no explicit solution , we numerically solved it for the overall elongation rate as a function of the EF-Tu concentration . For perfectly balanced conditions with X1/X2 = p1/p2 , where the relative abundance of the tRNAs matches the corresponding codon usages , 2C-2T translation is not ultrasensitive to the abundance of EF-Tu . In fact , the overall elongation rate has almost the same dependence on the concentration of EF-Tu as for the 1C-1T system , see Fig 3A ) . However , if the relative tRNA concentrations do not perfectly match the corresponding codon usages , the 2C-2T translation system responds in a much more complex way to variations of the EF-Tu availability: A regime of inhibited translation for small EF-Tu concentrations is followed by a relatively steep increase of the overall elongation rate , that finally saturates at larger EF-Tu concentrations , see Fig 3A ) . Thus , the ultrasensitivity to EF-Tu abundance found for in-vivo translation is also present in the imbalanced 2C-2T system , which renders the latter an appropriate model system to study the influence of EF-Tu abundance on translation . In particular , we can apply the 2C-2T system to find out what determines the onset of translation , i . e . , for which EF-Tu threshold concentration E * the overall elongation rate starts to increase . Fig 3A ) shows that the position of the transition regime from strongly suppressed to physiological protein synthesis is shifted towards higher concentrations of EF-Tu for stronger mismatches of tRNA concentrations and codon usages . We discovered that the EF-Tu threshold concentration E * only depends on the total concentrations X1 and X2 of tRNA species 1 and 2 and the codon usages p1 and p2 . It does not depend on any other parameter of the translation system such as the many transition rates that govern the kinetics of protein synthesis . Instead , this dependence is described by the unexpectedly simple relation E * = X 2 ( X 1 X 2 - p 1 p 2 ) = X 1 + X 2 - X 2 p 2 for X 1 X 2 ≥ p 1 p 2 and p 1 + p 2 = 1 . ( 1 ) The EF-Tu threshold concentration for the parameter regime X1/X2 ≤ p1/p2 is obtained by swapping the indices 1 and 2 in Eq ( 1 ) . A derivation of Eq ( 1 ) can be found in S1 Text . We also confirmed this relation for the singular case in which the ribosome cannot bind any near-cognate tRNAs , see S5 Fig . We conclude that the threshold concentration is determined by an imbalance between tRNA concentrations and cognate codon usages and that this imbalance can be expressed in terms of the total tRNA concentration X1 + X2 and by the ratio X 2 p 2 of tRNA concentration to cognate codon usage for the less abundant tRNA species 2 . The less abundant tRNA is hardly bound by EF-Tu to form ternary complexes if the EF-Tu concentration is below the threshold , see Fig 3B ) , while the concentration of the other ternary complex species remains at a high level . The competition of free aminoacylated tRNAs for free EF-Tu molecules causes the ultrasensitivity of systems with imbalanced tRNA/codon compositions: The oversupplied tRNA acts as a competitive inhibitor for the formation of ternary complexes containing the less abundant tRNA . The loss of one ternary complex species consequently causes a quasi break-down of protein synthesis [33] . A similar phenomenon was observed by Elf et al . when they analyzed the charging levels of isoacceptor tRNAs under amino acid starvation [34] . Elf et al . found that the sensitivities of tRNA charging levels and of translation rates to amino acid starvation depend on codon usages and tRNA concentrations , which is in line with the findings presented here . A direct quantitative comparison of the work by Elf et al . to our results is not feasible because the former is a highly simplified model of translation that neglects ternary complex formation . Still , it is notable that the inhibition of EF-Tu , which acts on protein synthesis in a global manner , and the deprivation of individual amino acids , which affects translation locally at the corresponding codons , have common characteristics . Translation in E . coli is of course much more complex than in the simple 2C-2T system . Surprisingly , Eq ( 1 ) still provides a very good estimate for the in-vivo EF-Tu threshold concentration E * , viv that marks the onset of translation in Fig 2A ) . Analyzing our computational in-vivo-like translation system [28] , we found that for E . coli growing at a specific rate of 2 . 5 h−1 under physiological conditions , ternary complexes containing Lys-tRNALys are most strongly affected by a decrease in available EF-Tu . The total concentration Xall of all tRNAs is 344 μM , the total concentration XLys of tRNALys is 10 . 43 μM , and the combined codon usage pLys of its cognate codons AAA and AAG is 7 . 46% [28 , 35] . Thus , if we replace X1 + X2 in ( 1 ) by Xall and X2/p2 by XLys/pLys , we obtain E * , viv = X all - X Lys p Lys = 204 μ M , ( 2 ) for the EF-Tu threshold concentration of E . coli which corresponds to about 80% of the physiological EF-Tu concentration and is in excellent agreement with the in-vivo EF-Tu threshold concentration predicted by our computational in-vivo-like translation system , see Fig 2A ) . To test the predictions of our computational frame-work of protein synthesis by an independent method , we computed the effect of EF-Tu limitation on the rate of protein synthesis using the “PURE system simulator” developed and published by Matsuura and co-workers [31] . This software is a highly detailed and experimentally well-validated in-silico representation of the E . coli-based reconstituted in-vitro protein synthesis system called PURE [36] . We used a recent version of the PURE system simulator , which was kindly provided by Drs . Matsuura and Shimizu , to simulate the in-vitro synthesis of fMetLysHis tripeptides via translation of short mRNAs consisting of a lysine and a histidine codon enclosed by a start and a stop codon ( AUGAAACACUAA ) . Simulation of fMetLysHis synthesis by the PURE system simulator represents an in-silico prediction of a 2C-2T in-vitro translation experiment . In the simulation , we varied the initial concentration of EF-Tu from 0 to 5 μM and determined for each EF-Tu concentration the rate of tripeptide synthesis at the end of the simulation after 1000 s when the simulated PURE system has long reached a quasi-steady state peptide synthesis rate . Fig 3C ) shows that the quasi-steady state peptide synthesis rate in the simulated PURE system is ultrasensitive to the concentration of EF-Tu if the tRNA concentrations do not perfectly match the codon usages . Again , the onset of translation is well-predicted by Eq ( 1 ) . In addition , we used the PURE system simulator to study the time-dependent response of 2C-2T in-vitro translation to a sudden drop in EF-Tu concentration . We simulated the synthesis of fMetLysHis tripeptides for 100s after which the synthesis rate has just reached a quasi-steady state level , see Fig 3D ) . At 100s , the concentration of total EF-Tu was reduced and the simulation was continued until the peptide synthesis has reached a plateau again , see Methods for details . Fig 3D ) shows that the simulated PURE translation system quickly adjusts to the new quasi-steady state after EF-Tu reduction . We used the PURE system simulator to compare the effects of EF-Tu inhibition as discussed above with the response of fMetLysHis synthesis to ribosome inhibition and the simultaneous inhibition of ribosomes and EF-Tu . We found that , in contrast to an inhibition that affects only EF-Tu , the ( additional ) inhibition of ribosomes causes an increase in the in the simulated peptide synthesis rate of the remaining ribosomes , as long as the inhibition is not too strong and the tRNA concentrations are sufficiently similar , see Fig 4 . This means that the remaining ribosomes proceed faster and that , to some extend , this increase in ribosomal speed can balance the loss of ribosomes , such that in the simulations the total rate of peptide synthesis remains approximately constant under mild inhibiting conditions . However , when the difference between the concentrations of tRNALys and tRNAHis is large enough , the simulated fMetLysHis synthesis is dominated by the availability of EF-Tu and an additional inhibition of ribosomes has a negligible effect on the peptide synthesis rate , see Fig 4A ) .
Efficient delivery of aminoacylated tRNAs to translating ribosomes is crucial for protein synthesis . In bacteria , this process is governed by the elongation factor Tu ( EF-Tu ) , which is one of the most abundant proteins . We applied our computational framework of protein synthesis [28] to investigate the influence of the availability of EF-Tu on the bacterial translation process . Surprisingly , we found a very limited tolerance of the translation system to deviations from physiological EF-Tu concentrations . Even a slight decline in EF-Tu availability causes a strong decrease of the overall translational activity . In turn , when the EF-Tu concentration reaches a certain threshold value , protein synthesis is switched on and the overall translation rate rises from zero to a physiological value within a relatively narrow range of EF-Tu concentrations . This ultrasensitivity is universal because it applies to both complex translation systems like E . coli containing 61 sense codons and 43 different tRNA species as well as simple artificial systems with only two codons and two cognate tRNAs . The onset of translation is determined by the imbalances between codon usages and cognate tRNA abundances . The corresponding EF-Tu threshold concentration can be obtained from an unexpectedly simple expression , see Eq ( 1 ) . Our theoretical predictions were confirmed by experimental data published by van der Meide , see Fig 2 , as well as by independent computer simulations of an in-vitro translation system using the PURE system simulator [31] . The ultrasensitive dependence of protein synthesis on EF-Tu might also be related to an observation made by Škrtić et al . The authors knocked-down mitochondrial initiation factor IF-3 , which facilitates translation initiation , as well as mt-EF-Tu in leukemia cells [37] . Silencing of mitochondrial EF-Tu , but not IF-3 , inhibited mitochondrial protein synthesis , which emphasizes the high regulatory power of EF-Tu . The toxin Doc from the phd/doc toxin-antitoxin system inhibits EF-Tu by phosphorylation [17 , 18] . Our results show that it is not necessary to phosphorylate a major fraction of EF-Tu to achieve a strong suppression of protein synthesis and , thus , cell growth . Instead , in E . coli only 20% of all EF-Tu molecules need to get phosphorylated by Doc to essentially stop protein production , which explains why Doc is an efficient toxin despite the extremely high cellular abundance of its target . Furthermore , the ultrasensitive dependence of protein synthesis on EF-Tu concentration implies that the phd/doc toxin-antitoxin system is an efficient regulator of protein synthesis . Cell growth can be easily regained as soon as the antitoxin Phd inhibits Doc as only relatively few EF-Tu molecules need to be reactivated by dephosphorylation . Consequently , this toxin-antitoxin system may mediate fast transitions from rapidly-growing to dormant phenotypes and vice versa . These transitions enable the effective formation of persister cells that can resist antibiotic treatments . In fact , depending on the specific site of modification , phosphorylation and other modifications of EF-Tu can impact protein synthesis in different ways . For example , in contrast to the toxin Doc , the Ser/Thr kinase YabT simultaneously inhibits both EF-Tu and ribosomes in Bacillus subtilis by stabilizing EF-Tu on translating ribosomes [25] . As a further example , Jakobsson et al . have shown that post-translational modifications of EF-Tu at the N-terminus and at residue Lys55 have an impact on protein synthesis rates [38] . Antibiotics directed against bacterial protein synthesis have a similar plurality in their modes of action and inhibit the ribosome and EF-Tu either individually or simultaneously in various ways , see Table 1 for examples . We found that the efficiencies of these different protein synthesis suppression pathways depend on the tRNA and codon compositions of the translation system . For balanced translation systems , for which the tRNA abundances roughly match the corresponding codon usages , inhibition of ribosomes rather than EF-Tu has a strong impact on the peptide synthesis rate . On the contrary , a decrease in EF-Tu greatly impedes protein synthesis in imbalanced systems , where the additional inhibition of ribosomes has a negligible effect . The ultrasensitive dependence of the overall elongation rate on the EF-Tu concentration is inherent in imbalanced translation systems but not in systems with a balanced tRNA/codon composition . This observation provides a possible explanation for the strain-specific correlation of EF-Tu abundance and growth rate in E . coli that is described in Ref . [30] , see also Fig 2B ) . Moreover , the mechanism studied in this work may help to understand the tissue-specificity of mitochondrial disorders caused by mutations of human mitochondrial EF-Tu ( mt-EF-Tu ) . Valente et al . report that a mutation in mt-EF-Tu that leads to severe inhibition of translation in mitochondria of the central nervous system does not affect other tissues [39 , 40] . This tissue-specificity might be related to differential mitochondrial mRNA and tRNA expression giving rise to differences in the mitochondrial tRNA/codon composition of the various tissues: Protein synthesis in mitochondria with less balanced tRNA/codon compositions should be more susceptible to a decrease in the availability of functional mt-EF-Tu . However , these speculative assumptions about mitochondrial protein synthesis need to be confirmed by quantitative analyses , which require the development of specialized translation models . The theoretical predictions presented in this work could be tested for in-vitro protein synthesis with cell-free expression systems as well as for in-vivo translation using , for example , EF-Tu and ribosome inhibiting drugs . Our finding , that the efficiency of protein synthesis inhibition mediated by EF-Tu depends on the tRNA/codon composition , hints towards a potential use of EF-Tu targeting drugs for tissue- or pathogen-specific treatments . It thus may encourage further studies for the identification of novel compounds directed against EF-Tu [41] .
We performed our analysis within the theoretical framework of translation developed in Refs . [27 , 28] and briefly summarized in S1 Text . The framework incorporates a multitude of factors that influence the speed and fidelity of cellular protein synthesis . Important parameters are concentrations ( of ribosomes , tRNAs , mRNAs , and elongation factors ) , as well as codon usages and predicted in-vivo biochemical rates ( for tRNA charging by aminoacyl-tRNA synthetases , ternary complex formation , and ribosomal kinetics ) . Furthermore , cognate , near-cognate , and non-cognate relations of all tRNAs and codons are taken into account . To capture the stochastic nature of protein synthesis , translation elongation is described as a continuous-time Markov process . For the analysis of protein synthesis in E . coli as shown in Fig 2A ) , we applied our computational framework of in-vivo-like translation as published in Ref . [28] without any changes in parameters ( except for EF-Tu concentration ) nor reaction pathways and , thus , refer the reader to the original publication for details on the method . Adjustments made to the framework to model protein synthesis in simplified 1C-1T and 2C-2T translation systems are described in S1 Text , with all parameters assuming the values given in Ref . [28] . For simulations with the “PURE system simulator” [31] , all parameters , such as kinetic rates and concentrations , except for the initial concentrations of mRNAs , tRNAs and EF-Tu were used as provided by Matsuura et al . to maximize comparability with the experimental PURE system . The concentration of mRNA was set to 10 μM , the total ( summed ) concentration of tRNALys and tRNAHis to 3 . 44 μM , and the concentrations of all other tRNAs were set to zero . No further changes were made to the system . Translation rates were calculated by dividing for each point in time the increment in tripeptide amount by the corresponding increment in time . To simulate the impact of EF-Tu inhibition as a function of time , translation was simulated until the translation rate has reached a quasi-steady state level . At the indicated point in time , the concentrations of all species containing EF-Tu were reduced by a specific amount ( fraction ) . To compensate unintended losses of EF-Tu binding partners , such as tRNAs or ribosomes , the concentrations of these affected species were increased by corresponding amounts . For example , if the concentration of ternary complexes containing EF-Tu and Lys-tRNALys was decreased by a certain amount , the concentration of Lys-tRNALys was increased by the same amount . The original PURE system simulator can be downloaded from the website of Matsuura et al . : https://sites . google . com/view/puresimulator [31] .
|
We predict and analyze the response of differently composed protein synthesis systems to the inhibition of elongation factor EF-Tu and/or ribosomes . The study reveals a strong interdependency of a protein synthesis system’s composition and its susceptibility to inhibition . This interdependency defines a generic mechanism that provides a common basis for a variety of seemingly unrelated phenomena including , for example , persister cell formation and tissue-specificity of certain mitochondrial diseases . The described mechanism applies to simple artificial translation systems as well as to complex protein synthesis in vivo .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"antimicrobials",
"transfer",
"rna",
"medicine",
"and",
"health",
"sciences",
"toxins",
"pathology",
"and",
"laboratory",
"medicine",
"drugs",
"microbiology",
"toxicology",
"toxic",
"agents",
"protein",
"synthesis",
"antibiotics",
"mitochondria",
"pharmacology",
"bioenergetics",
"cellular",
"structures",
"and",
"organelles",
"chemical",
"synthesis",
"research",
"and",
"analysis",
"methods",
"peptide",
"synthesis",
"proteins",
"gene",
"expression",
"biosynthetic",
"techniques",
"ribosomes",
"biochemistry",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"protein",
"translation",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"energy-producing",
"organelles",
"non-coding",
"rna"
] |
2019
|
Efficiency of protein synthesis inhibition depends on tRNA and codon compositions
|
Stomata , microscopic pores in leaf surfaces through which water loss and carbon dioxide uptake occur , are closed in response to drought by the phytohormone abscisic acid ( ABA ) . This process is vital for drought tolerance and has been the topic of extensive experimental investigation in the last decades . Although a core signaling chain has been elucidated consisting of ABA binding to receptors , which alleviates negative regulation by protein phosphatases 2C ( PP2Cs ) of the protein kinase OPEN STOMATA 1 ( OST1 ) and ultimately results in activation of anion channels , osmotic water loss , and stomatal closure , over 70 additional components have been identified , yet their relationships with each other and the core components are poorly elucidated . We integrated and processed hundreds of disparate observations regarding ABA signal transduction responses underlying stomatal closure into a network of 84 nodes and 156 edges and , as a result , established those relationships , including identification of a 36-node , strongly connected ( feedback-rich ) component as well as its in- and out-components . The network’s domination by a feedback-rich component may reflect a general feature of rapid signaling events . We developed a discrete dynamic model of this network and elucidated the effects of ABA plus knockout or constitutive activity of 79 nodes on both the outcome of the system ( closure ) and the status of all internal nodes . The model , with more than 1024 system states , is far from fully determined by the available data , yet model results agree with existing experiments in 82 cases and disagree in only 17 cases , a validation rate of 75% . Our results reveal nodes that could be engineered to impact stomatal closure in a controlled fashion and also provide over 140 novel predictions for which experimental data are currently lacking . Noting the paucity of wet-bench data regarding combinatorial effects of ABA and internal node activation , we experimentally confirmed several predictions of the model with regard to reactive oxygen species , cytosolic Ca2+ ( Ca2+c ) , and heterotrimeric G-protein signaling . We analyzed dynamics-determining positive and negative feedback loops , thereby elucidating the attractor ( dynamic behavior ) repertoire of the system and the groups of nodes that determine each attractor . Based on this analysis , we predict the likely presence of a previously unrecognized feedback mechanism dependent on Ca2+c . This mechanism would provide model agreement with 10 additional experimental observations , for a validation rate of 85% . Our research underscores the importance of feedback regulation in generating robust and adaptable biological responses . The high validation rate of our model illustrates the advantages of discrete dynamic modeling for complex , nonlinear systems common in biology .
The epidermes of leaves and other aerial plant parts have natural openings known as stomata . Stomata are the entry and exit points where gas exchange , particularly CO2 uptake for photosynthesis and O2 and water vapor diffusion from the leaf interior to the atmosphere , takes place . Each stomate is surrounded by a pair of guard cells that modulate stomatal apertures in response to endogenous water status , to multiple phytohormones , and to many environmental signals such as light and CO2 [1–4] . Abscisic acid ( ABA ) is a key phytohormone that functions as an essential signal in plant responses to abiotic and biotic stress [5 , 6] . Drought promotes ABA biosynthesis and accumulation in guard cells [7 , 8] . ABA activates complex intracellular signaling cascades that ultimately open anion channels at the guard cell membrane [9] , leading to anion efflux and thus cellular depolarization , which in turn drives K+ efflux through depolarization-activated outward K+ channels [10 , 11] . Solute loss drives the osmotic efflux of water through aquaporins [12] , resulting in guard cell deflation and stomatal closure [13] . An ultimate goal in modeling a complex system such as guard cell ABA signaling is to create a model in which the output from each element can be quantitatively predicted . This has been accomplished for the small number of “effector” nodes ( i . e . , ion transporters and channels ) of the guard cell network , as provided in the OnGuard software [14] . However , for the intracellular signaling cascade that functions between ABA and effectors , regulation of the secondary messengers is indirect , and little information is available concerning quantitative relationships or response kinetics among the signaling components . In such situations , which are common in complex biological systems , information is simply insufficient to quantitatively parameterize a continuous ( kinetic ) model , and any attempt to do so would require an inordinate number of assumptions . Network modeling then offers the best way forward to synthesize the available information into a logical framework that is consistent with experimental data [15] . Signaling components ( signaling proteins , enzymes , and small molecules ) are represented as nodes , the regulatory relationships between them are represented as edges [16 , 17] , and discrete dynamic modeling is the preferred choice to study the dynamic behavior of the system [18 , 19] . We previously constructed the first discrete model of the ABA signaling network that mediates stomatal closure; this model successfully reproduced many observed knockout phenotypes and predicted many new mutant phenotypes [16] . At the time , our 40-node Boolean model of guard cell ABA signaling was the largest system to have been studied by Boolean modeling with asynchronous update . More recently , we constructed a multilevel dynamic model of light-induced stomatal opening and of the effects of CO2 and ABA on this process , which , among many other predictions , led us to discover and experimentally address an open question concerning ABA inhibition of red light–induced stomatal opening [20] . In recent years , many new aspects of guard cell ABA signaling have been revealed including , at the molecular level , soluble ABA receptors [21 , 22] , anion channel genes such as slow anion channel-associated 1 ( SLAC1 ) [23] , and aquaporin genes [12] . Upon ABA perception , the soluble ABA receptors interact with and inhibit clade-A protein phosphatase 2Cs ( PP2Cs ) : ABA-insensitive 1 ( ABI1 ) , ABA-insensitive 2 ( ABI2 ) , hypersensitive to ABA 1 ( HAB1 ) , and protein phosphatase 2CA ( PP2CA ) ; this interaction relieves inhibition of the serine-threonine kinase OPEN STOMATA 1 ( OST1 ) [24] . Phosphorylation-based channel regulation mediated by OST1 as well as several other classes of protein kinases , including Ca2+-dependent kinases ( CPKs ) and mitogen-activated protein kinases ( MPKs ) , then occurs [25–29] . ABA promotion of cytosolic Ca2+ ( Ca2+c ) increases , reactive oxygen species ( ROS ) production , pH increases , and vacuolar acidification also have been found to play positive roles in stomatal closure [30] . Actin reorganization [31] and microtubule depolymerization [32] observed in response to ABA highlight the importance of cytoskeletal rearrangements . Here , we present a current signaling network and dynamic model for ABA-induced stomatal closure . We extracted information from the published literature and identified 84 signaling components . We employ asynchronous Boolean modeling to study dynamic behaviors of signaling components during signal propagation . We identify long-term states ( attractors ) and their corresponding precursor states in our dynamic model . A novel stable motif analysis provides a particularly compelling prediction regarding Ca2+c regulation of PP2C activation . We also simulate node knockout ( in the in silico presence of ABA ) for all internal nodes and find that the results for over 50 signaling components agree with phenotypes observed previously by wet-bench experimentation . Moreover , we provide new experimental data that verify an unexpected model prediction regarding the impact of heterotrimeric Gα subunit ( GPA1 ) knockout on ABA regulation of Ca2+c . We also comprehensively simulate node constitutive activity ( in the in silico presence and absence of ABA ) and experimentally validate 2 of the resultant predictions regarding ABA hypersensitivity conferred by constitutive activation of Ca2+c or ROS .
Through extensive literature analysis , we identified signaling components and evidence of interactions or causal effects between them from over 120 articles published through 2015 , resulting in compilation of over 190 relationships described in the guard cell literature ( S1 Table ) . We interpreted indirect causal relationships in a parsimonious way , e . g . , by interpreting causal relationships involving 3 nodes as 2 pairwise relationships ( see Materials and methods ) . The resultant network ( Fig 1 ) has ABA as the primary signal and stomatal closure as the only output node . ABA induces stomatal closure mediated by the soluble pyrabactin resistance 1/pyrabactin resistance 1-like/regulatory component of ABA receptor ( PYR1/PYL/RCAR ) ABA receptors ( RCARs node ) . The network also contains a few important ABA-regulated nodes , e . g . , phosphoenolpyruvate carboxylase ( PEPC ) [33] and phosphatidylinositol 3-phosphate 5-kinase ( PI3P5K ) [30] , that are not known to be regulated by these soluble receptors . Twenty-two nodes other than ABA are designated as source nodes because they are not known to be regulated by any other nodes in the network . The network also contains 60 intermediate nodes that are signaling proteins ( e . g . , protein phosphatases , protein kinases ) , metabolic enzymes ( e . g . , NADPH oxidases ) , secondary messengers and small molecules ( e . g . , Ca2+c , cyclic GMP ) , cytoskeletal proteins ( actins and microtubules ) , and effector proteins ( e . g . , anion and K+ channels and aquaporins ) . The network contains 156 edges ( S2 Table ) . There are 103 direct edges ( direct physical interactions or chemical reactions ) , 27 indirect edges , and 26 edges inferred during our interpretation of indirect evidence . S1 Text provides a complete biological description of the network , with the associated literature references that support it . To identify the general information propagation capacity of the network , we determine the network’s largest strongly connected component ( SCC ) ( i . e . , feedback-rich subnetwork ) and the nodes that can reach or can be reached from it ( Materials and methods ) . The ABA signal transduction network contains a surprisingly large SCC composed of 36 nodes ( Fig 2A , S4 Table ) . This SCC contains multiple positive feedback loops , e . g . , Ca2+c→PLC→InsP3→CIS→Ca2+c , which describes a mechanism of Ca2+c-induced Ca2+ release , and negative feedback loops , e . g . , between Ca2+c and the Ca2+ ATPase . Twenty-eight nodes are in the network’s in-component ( i . e . , they can reach the nodes of the SCC through paths ) , including 22 source nodes ( all except CPK6; see S4 Table ) . Nineteen nodes are in the out-component ( can be reached from the SCC ) , including nodes that comprise plasma membrane ion channels and fluxes and the output node , Closure . All the paths from ABA to Closure pass through the SCC . We use 3 measures of each node’s signal-mediating role: in-degree ( the number of incoming edges ) , out-degree ( the number of outgoing edges ) , and betweenness centrality ( the number of paths mediated by the node , see Materials and methods ) . The distribution of the nodes according to these 3 measures is shown in S5 Table . Aggregating all 3 measures , 5 well-represented categories of nodes can be identified ( Table 1 ) : source nodes , transducers , signal integrators , intermediate SCC members , and key nodes ( namely , ABI1 , Ca2+c , phosphatidic acid [PA] , cytosolic pH [pHc] , and ROS ) . Boolean models are a rich , dynamic representation of a signaling system in which each node is characterized by 2 possible states: ON or 1 ( interpreted as a higher-than-threshold abundance or activity ) , and OFF or 0 ( lower-than-threshold abundance and activity ) . For example , the ON state of the OST1 kinase means that there is a sufficient abundance of OST1 proteins in the active , phosphorylated form , such that they can catalyze the phosphorylation of SLAC1 anion channels and other downstream targets . In our dynamic model , we implement the initial states of 54 signaling components as they are known or expected to be in guard cells of open stomata prior to ABA exposure , based on information in the literature , and initialize the remaining 26 randomly ( see Materials and methods , S6 Table ) . The future state of each node is determined by the current state of its direct regulators ( the sources of incoming edges ) and is expressed as a Boolean regulatory function ( see Materials and methods ) . In Table 2 , we give the Boolean regulatory functions for nodes with 3 or more regulators , and associated references . A full list and explanation of the Boolean functions used is in S2 Text . There are 23 unregulated nodes that do not have regulatory functions; they will sustain the state they are initiated with . In Boolean network models , the simulation of successive events ( e . g . , successive activation of receptor proteins , downstream second messengers , and ion channels ) is implemented by update schemes ( see Materials and methods ) . We used a stochastic update algorithm in which the nodes are updated in a randomly selected order for 30 rounds ( time steps ) . In each simulation , the system transitions through several states until it settles down in an attractor ( e . g . , a steady state or oscillation ) . In each setting , we used 2 , 500 replicate simulations , i . e . , 2 , 500 “in silico stomata” and determined at each time step the percentage of simulations in which the node Closure is in state 1 , which we will refer to as the percentage of closure . We started with simulation of the response of wild-type stomata following the sustained presence of ABA ( Fig 3 , filled circles ) . The percentage of closure is initially 0 , corresponding to open stomata . After a time delay of approximately 4 time steps , during which the signal propagates through the network , the percentage of closure increases dramatically , reaching 100% by time step 20 . This indicates that all 2 , 500 simulations , regardless of the random differences in the initial states of 26 nodes and of the random differences in the order of update of the nodes , reach a state that corresponds to stomatal closure . In the converse situation of sustained absence of ABA ( Fig 3 , open circles ) , the percentage of closure remains at zero , indicating that no simulations are able to reach closure in the absence of the ABA signal . These results indicate that we have captured the expected behavior of the wild-type system . For a deeper analysis of the long-term states ( attractors ) reached by the system in the presence and absence of ABA , we performed stable motif analysis ( Materials and methods ) . This analysis is based on finding feedback loops that stabilize in a fixed state and determine the system’s attractors [96] . In the sustained presence of ABA , there are 4 such stable motifs ( Fig 2B ) : positive self-regulation of CPK3/21 , MPK9/12 , and Microtubule Depolymerization , and the positive feedback loop PA→RBOH→ROS→PLDδ→PA , each of which leads to a sustained ON state of the corresponding node ( s ) . In the presence of ABA , any dynamic trajectory leads to the stabilization of each of the motifs of Fig 2B , which yields an attractor of the system ( S6 Table ) in which 70 nodes stabilize ( most , but not all , in opposite states than their initial state ) . Ten nodes oscillate , driven by the negative feedback loop between Ca2+c and Ca2+ ATPase; most of these nodes are well known to be regulated by and/or regulate Ca2+c and indeed Ca2+c is observed to oscillate in response to ABA [59 , 60 , 97] . The model-predicted attractor is in good agreement with the known response of the network elements to the sustained presence of ABA [1 , 3 , 98–101] . In the absence of ABA , there are 2 main stable motifs associated with the state of no stomatal closure . The first represents the OFF state of the positive feedback loop between OST1 and ROS combined with the ON state of 3 PP2Cs , e . g . , ABI2 , HAB1 , and PP2CA , each of which is in a mutually inhibitory relationship with ROS ( Fig 2C ) . Second , a family of 3 related stable motifs ( Fig 2E ) represents the OFF state of 2 interlocking positive feedback loops involving Ca2+ influx through the membrane ( CaIM ) and Ca2+ release from internal stores ( CIS ) , respectively . Stabilization of the first motif leads to stabilization of the second . In addition , the self-regulation of 4 nodes , namely CPK3/21 , MPK9/12 , Microtubule Depolymerization , and Vacuolar Acidification , allows their stabilization in either the ON or the OFF state in the absence of ABA . As a consequence , there are 16 reachable final attractors ( one for each combination of states of these 4 nodes ) associated with the absence of stomatal closure ( S6 Table and S7 Table ) , which nevertheless have identical states for all but 9 nodes . The stabilized state of many nodes is identical to their initial states , which correspond to open stomata , as expected . In addition , there also exists a stable motif associated with closure ( Fig 2D ) . In the absence of ABA , this motif does not stabilize in any of the trajectories that start from our initial condition corresponding to open stomata . Accordingly , our model predicts that there are no trajectories that would lead to closure of open stomata in the absence of ABA or other closing signals , consistent with biological reality . We performed a comprehensive analysis of the effects of single node knockout ( sustained OFF state , see Materials and methods ) and constitutive activity ( sustained ON state ) on ABA-induced stomatal closure . We observed 4 major types of responses: insensitivity , reduced sensitivity , hyposensitivity , and hypersensitivity to ABA ( Materials and methods ) . Fig 3 illustrates a few representative cases . The model predicts that ost1 knockout leads to insensitivity to ABA ( Fig 3A ) , in agreement with experiments [25 , 48] . Two example cases of reduced ABA sensitivity ( rboh and pHc clamp ) are well supported experimentally [52] . A third , phosphatidylcholine ( PC ) depletion , is not amenable to experimental study , as PC is a fundamental component of the lipid bilayer . Fig 3B shows examples of node knockouts that lead to ABA hyposensitivity ( pldα and gapc1/2 ) , both supported by the literature [35 , 67] . Of the 3 examples of ABA hypersensitivity shown ( abi1 , abh1 , and Ca2+ ATPase knockout ) , the first 2 are documented cases of increased ABA sensitivity [44 , 65] , while knockout of Ca2+ pumps has not been experimentally evaluated . Fig 3C shows examples of constitutive node activity that lead to insensitivity to ABA ( abi1 dominant mutant , constitutively active Ca2+c ATPase ) or reduced ABA sensitivity ( constitutive activity of ABI2 or PP2CA ) . Impaired ABA sensitivity of abi1 and abi2 dominant mutants was indeed observed [50 , 102 , 103] . The percentage of closure in the case of ROP11 constitutive activity approaches 100% closure by 30 time steps; thus , this is an example of ABA hyposensitivity , which also agrees with experiments [104] . Fig 3D shows examples of constitutive node activity that lead to ABA hypersensitivity: supply of PA , which agrees with experimental observations [35] , or supply of ROS , which we experimentally assess and confirm . Summarizing the results of individual knockout or constitutive activity of all 79 internal nodes in the presence of ABA ( S8 Table ) , 30% lead to no appreciable difference from a wild-type response to ABA ( i . e . , stomatal closure ) , 30% lead to a more effective response , 14% lead to a somewhat reduced response , and 26% lead to a dramatically reduced response . Where genetic or pharmacological intervention experiments have been performed , comparison with these wet-bench results can serve to evaluate the model . Table 3 summarizes the relationship between the in silico results and the closest comparable experimental results . Agreement is obtained in 57 cases of 158 ( shown in bold font ) or 36% and there are 16 cases of discrepancy ( 10% , shown in italic font ) . In most discrepancies , the model is indicating close to wild-type response to ABA , while experiments found decreased ( 1b ) or increased ( 1a ) sensitivity to ABA . In the remaining 54% of cases , no experimental data are currently available; thus , the model’s result is a novel prediction ( S8 Table ) . Stable motif-based attractor analysis indicates that all the interventions that lead to a final closure probability of 100% ( i . e . hyposensitivity , close to wild-type response , and hypersensitivity ) lead to an attractor identical or very close to the attractor corresponding to wild-type ABA-induced closure ( S6 Table ) . Any differences from this attractor are limited to the node that is knocked out or constitutively activated and up to a few nodes immediately regulated by it , but the stable motifs ( shown in Fig 2B ) are preserved . Interventions that lead to reduced ABA sensitivity lead to an attractor in which AnionEM , H2O efflux , and Closure oscillate , driven by oscillating Ca2+c ( see S9 Table for an example ) . There are 2 types of attractors corresponding to insensitivity to ABA: one type similar to the attractors expressing lack of closure in the absence of ABA ( e . g . , in case of RCARs knockout ) and another type that preserves the stable motifs of Fig 2B but Closure is OFF because one of its required upstream regulators is OFF ( S9 Table ) . The model also can be used to predict how genetic or pharmacological knockout of an internal node affects other internal nodes . S10 Table shows agreement between the model result and experimental observation for 12 of these instances , supporting the internal consistency and predictive power of the model . The most valuable biological models provide new predictions that guide experimental research . One especially intriguing prediction of our model is that GPA1 knockout should not significantly impact Ca2+c oscillations . While gpa1 mutants are known to be impaired in ABA activation of the ROS-activated plasma membrane Ca2+ channels [51] , which our model captures , the simulated loss of GPA1 does not affect stretch-activated Ca2+ channels , nor does it affect Ca2+-induced CIS in our model . We tested this prediction by introducing the Yellow Cameleon 3 . 6 ( YC 3 . 6 ) fluorescent Ca2+ reporter into gpa1 guard cells and measuring Ca2+c following ABA application . As is evident in Fig 4A and 4B , ABA does initiate wild-type Ca2+c transients in gpa1 guard cells , consistent with our dynamic model . Given the paucity of experimental data on the effects of node constitutive activation in the presence of ABA , we decided to test 2 such predictions: constitutive activation of Ca2+c by provision of external Ca2+ [131 , 132] and constitutive activation of ROS by provision of H2O2 . As shown in Fig 4C and 4D , in both cases , the experimental results support the model predictions: in agreement with our definition of ABA hypersensitivity , stomatal closure occurred more rapidly when the internal node was constitutively activated , and it reached the same final closure state . These data illustrate the predictive power of our model . We next considered effects of constitutive activation or external supply of a node in the absence of ABA . According to model predictions , the constitutive activity of any of 67 nodes does not lead to any predicted stomatal closure from the open state , while for 12 nodes , there is a small ( 7 cases ) or significant ( 5 cases ) closure response ( Table 4 , S11 Table ) . Only constitutive activity of ROS leads to a closure probability similar to the response to ABA ( Fig 5A ) , consistent with experimental data [51 , 63] . Comparing to experimental observations , there is agreement in 13 cases ( 16% ) and disagreement in 11 cases . Notably , for the majority of cases ( 70% , see S11 Table ) , there are no comparable experiments , again highlighting the usefulness of modeling in hypothesis generation . Our stable motif analysis explains the effectiveness of ROS in promoting stomatal closure in the absence of ABA: it can lead to the stabilization of the stable motif in Fig 2D . The model indicates that sustained activity of ROS , RBOH , RCARs , or OST1 can induce the stabilization of this stable motif in a fraction of the trajectories in the system , which then leads to stomatal closure in these trajectories . This result is seemingly at odds with the experimental observation that overexpression of OST1 and RCARs does not lead to increased stomatal closure in the absence of ABA [25 , 109] . One reason for apparent inconsistency with experimental results is that overexpression of a protein may not correspond to its constitutive activation , e . g . , if its activation involves posttranslational modification . Experimental observations also indicate more potential nodes whose constitutive activation is able to lead to closure than are predicted by our simulations , namely , the 9 nodes categorized as 1a in Table 4 . We investigated the basis for the above-noted discrepancy between model and experimental results and identified it in the effect of the PP2C family member protein phosphatases . Because , according to experimental observations , these PP2Cs are biochemically active in the absence of ABA [103] , they lead to stabilization of the stable motif in Fig 2C in all the trajectories of the signal-free system and even in the vast majority of trajectories in the presence of internal drivers . Elevated PA or S1P can overcome this outcome only when combined with destabilization of the activity of multiple PP2Cs . To explicitly test in silico the importance of the initial state of the PP2Cs , we reanalyzed the model with a revised initial state , in which the 4 PP2Cs are inactive ( OFF ) . We found that the stable motif in Fig 2D now has a chance of being stabilized , leading to an attractor ( A17 in S7 Table ) close to the attractor corresponding to ABA-induced closure ( S6 Table ) . As indicated in Fig 5B , the simulated system with the 4 PP2Cs initialized as OFF but no other interventions ( a situation that we call the baseline ) reaches closure in around 44% of trajectories . The same result of nonzero percentage of closure is observed for constitutive activity of the majority of nodes ( S12 Table ) . Importantly , the model now predicts robust closure for 20 internal drivers . When combined with an initial down-regulation of the PP2C protein phosphatases , sustained Ca2+c or sustained CaIM [136] leads to a closure response similar to the wild-type response to ABA ( see Fig 5B ) , and external supply of S1P [70] now also leads to an increased frequency of closure , consistent with experimental results . There are no wet-bench experiments that manipulate the initial state of the PP2Cs , but if one uses the experiments that served as comparison in Table 4 , agreement is now found for 10 of the 11 cases of italicized discrepancy ( every case except supply of nitrite ) . The resolved nodes are shown with bold font in S12 Table . Importantly , the consistency with experiments of the boldface nodes in Table 4 is maintained . Stable motif analysis of the simulated system in the absence of ABA with all 4 PP2Cs knocked out ( S13 Table ) indicates that there are 15 attractors: 14 attractors associated with lack of closure ( with a similar node expression pattern as in S6 Table ) as well as an attractor associated with closure ( the same as in S7 Table ) . Importantly , the existence of this closure attractor is supported by the experimental observation that the hab-1 abi-2 pp2ca-1 triple null mutant shows partial stomatal closure in the absence of ABA [45] . The shared observation of the attractor associated with closure for both initial inactivity of the 4 PP2Cs and for their knockout indicates that transient inactivity of the 4 inhibitors of closure can be made permanent by the feedbacks in the system ( specifically , by the stable motif in Fig 2D ) . The assumption that the PP2C phosphatases are present but initially inactive in open stomata has not been evaluated experimentally , and the prediction from this assumption of a significant degree of closure in the absence of ABA ( open circles in Fig 5B ) is not consistent with current knowledge regarding behavior of guard cells with intact PP2Cs . However , the strong effect of a merely transient ( initial ) down-regulation of the PP2Cs ( S12 Table ) suggests that alternative , weaker assumptions may also be explanatory . Reinspecting the stable motifs of Fig 2C and 2D , which correspond to 2 opposite outcomes in the absence of ABA , one can see that both involve mutual inhibition between , on the one hand , PP2Cs and , on the other hand , OST1 , RBOH , and ROS . If ABI2 , PP2CA , and HAB1 are ON , then OST1 , RBOH , and ROS are OFF ( Fig 2C ) . Conversely , destabilization of the activity of ABI1 and ABI2 , coupled with the stabilization of the positive feedback loop PA→RBOH→ROS→PLDδ→PA , can lead to the stabilization of the stable motif in Fig 2D and thus to closure in the absence of ABA . We find that assuming initial inactivity of only ABI1 and ABI2 also resolves the inconsistencies with experimental observations of Table 4 and at the same time leads to a lower baseline probability of closure ( S14 Table ) . Because 7 of the nodes with type 1a discrepancies in Table 4 participate in paths that lead to Ca2+ release from stores ( for example , Nitrite→NO→NOGC1→cGMP→8-nitro-cGMP→ADPRc→cADPR→CIS ) , an alternative specific hypothesis is that Ca2+c increase leads to the inactivation of one or more PP2Cs . Such a mechanism would not affect the system’s responsiveness to ABA , because the PP2Cs are inactivated early by RCARs . It is also consistent with knockout of RCARs , leading to insensitivity to ABA , as our model indicates that the loss of RCARs leads to the absence of an increase in Ca2+c . We find that assuming that Ca2+c increase leads to the inactivation of all 4 PP2Cs resolves 10 of the 11 inconsistencies with experimental observations of Table 4 ( every case except supply of nitrite ) and at the same time leads to a lower baseline probability of closure in the absence of ABA ( S14 Table ) . The thus-augmented model’s experimental validation rate is increased to 85% ( see Materials and methods ) . Consistent with our hypothesis is the observation that dominant negative abi1 and abi2 mutants fail to show stomatal closure induced by external Ca2+ application [138] . In fact , an early study on recombinant ABI1 observed inhibition of phosphatase activity by Ca2+ in vitro , but the result was interpreted to lack biological relevance , as inhibition was only observed at much higher Ca2+ concentrations than those found in living cells [139] . Our model suggests that ABI1 may be more sensitive to Ca2+ inhibition in vivo than in vitro , and it is intriguing to note that the ABI1 protein contains a putative Ca2+-binding EF hand motif [139] .
Our construction of a network of ABA-induced stomatal closure based on the literature provides a means to comprehend the vast number of relationships reported for this complex system . In network construction , we employed causal network inference and binary transitive reduction [140] to facilitate identification of the most parsimonious paths between the more than 80 known signaling components of ABA-induced stomatal closure . We found that the system is strongly nonlinear . The network contains a 36-node SCC ( Fig 2A ) that includes 5 key nodes with high in- and out-degree ( ABI1 , Ca2+c , PA , pHc , and ROS ) , several nodes , particularly CaIM and RBOH , that serve as integrators of information , and several nodes with high out-degree , such as OST1 ( Table 1 ) . The integrators and PA receive information from the network’s in-component ( which includes ABA ) , while Ca2+c , ROS , and pHc relay information to the network’s out-component ( which includes Closure ) , and ABI1 performs both functions ( Fig 2A ) . Indeed , these nodes are known important mediators of ABA-induced stomatal closure [47 , 52 , 77] , supporting the validity of our network reconstruction . The SCC serves as a core of information processing and mediator of all the paths between ABA and closure . Its strong nonlinearity emphasizes how simple epistatic reasoning may not always be valid for biological systems . The SCC contains numerous positive feedback loops , several of which constitute stable motifs . This inclusive relationship between the SCC ( a property of the network ) and the stable motifs ( a property of the dynamic model ) reflects the added insights contributed by dynamic modeling . The SCC identifies the interrelatedness of its 36 nodes; the stable motifs identify which interrelated sets of nodes will stabilize and drive the system’s outcome in the presence or absence of ABA . Given the complexity of the guard cell ABA signaling network combined with lack of information on the quantitative kinetics of changes in internal node status , discrete dynamic modeling offers an optimal approach to provide insights and predictions regarding intracellular signaling mechanisms . The model provided here has been evaluated by new approaches: stable motif analysis , determination of the long-term states ( attractors ) of the system , and simulation of how stomatal apertures in the presence or absence of ABA are affected by in silico constitutive activation of each internal node . Our stable motif analysis indicates that the response to ABA is robust to variability in the timing of individual events and to variations in the initial state of 26 nodes , confirming a similar conclusion of the previous discrete dynamic model of ABA-induced closure [16 , 141] . Our model successfully recapitulates well-characterized examples of impaired ABA sensitivity ( Fig 3A and 3C ) and increased ABA sensitivity ( Fig 3B and 3D ) . In the cases with experimental documentation , 59 out of 76 simulations of individual node knockouts or constitutive activations in the presence of ABA ( boldface nodes in Table 3 and 2 experiments reported here ) and 13 of 24 simulations of constitutive activation of nodes in the absence of ABA ( boldface nodes in Table 4 ) match experimental evidence . Moreover , for the limited number of cases for which experimental data are available , our model accurately captures the effect of knockout of an internal node on the status of a second internal node in the presence of ABA ( S8 Table and the experiment reported here ) . Our model goes beyond the previous discrete dynamic model of ABA-induced closure [16] in a way that preserves the results in which the previous model agrees with experimental observations and improves the agreement of other results ( S3 Text ) . Our model recapitulates the core ABA signaling chain—ABA binding to receptors , which alleviates negative regulation by PP2Cs of the OST1 kinase , ultimately resulting in activation of anion channels—and places it in a broader context . Indeed , the subnetwork from ABA to the 3 anion channels , shown in Fig 6 , is largely dependent on the ABA→RCARs—●PP2Cs—●OST1 chain ( here—● refers to an inhibitory edge ) . This is consistent with the fact that RCARs or OST1 knockout or constitutive activation of the PP2Cs confers ABA insensitivity . This subnetwork also includes the whole SCC of the network ( to which PP2Cs and OST1 belong ) , indicating the importance of feedbacks . Knockout of several other members of this subnetwork has also been observed to lead to decreased ABA sensitivity , indicated as red , orange , or yellow color in Fig 6 . Notably , the process of vacuolar acidification , which , according to current knowledge , is independent of the core ABA signaling chain , is also a significant indirect contributor to anion channel activation . Furthermore , our prediction that Ca2+c increase inhibits the PP2Cs ( which yields an almost perfect model validation rate in the absence of ABA ) suggests the existence of an additional feedback into the core ABA signaling chain . It is important to note that , while our model is based on experimental evidence , this does not ipso facto guarantee the congruencies described above: ( 1 ) the evidence incorporated in the regulatory functions is at the level of specific edges ( i . e . , it only considers the most direct regulators of the target node ) and so provides no control over pathway-level effects; ( 2 ) the network-level outcomes ( the dynamic attractors ) are emergent properties of the system; and ( 3 ) in construction of the regulatory functions , we found numerous instances of insufficiency of information and tested several alternatives before settling on the function that most faithfully recapitulated biological results . Despite the excellent validation rate , some discrepancies with the literature remain . In some cases , this may be due to unresolved inconsistencies within the literature itself . For example , the calcium-dependent kinase CPK6 has been reported to be either active or inactive at resting Ca2+c levels [26 , 142] . Our model uses the assumption that CPK6 is always active , as is CPK23 , while CPK3/21 is Ca2+ activated and then maintains its activity through autophosphorylation . Assuming that CPK6 and CPK23 are Ca2+ activated and thus follow the transient increase and decrease of Ca2+c would yield the result that the anion channels’ activation and ultimately the stomatal aperture also increase and decrease with Ca2+c . This is not consistent with current knowledge . Making the assumption that the activity of all the Ca2+-activated CPKs is maintained through autophosphorylation ( which is a known property of CPK kinases ) would restore agreement and maintain all the current results of the model . Another source of discrepancies between experimental and model results may have to do with cumulative percentage of closure ( CPC ) thresholds , although logically defined , imprecisely capturing the biological reality . For example , had the CPC range for “hyposensitivity” been extended to a slightly greater upper limit ( 23 . 97 instead of 23 . 92 ) , 2 cases of type 1b error ( MPK9/12 knockout and InsP3 knockout; Table 3 ) would have disappeared . Another reason for discrepancies is the fact that in a Boolean framework , either constitutive activity or constitutive inactivity of a source node must have the same meaning as the assumed wild-type unchanged state . In 2 cases of discrepancies regarding source nodes , namely , SCAB1 and GEF1/4/10 , their knockout and their constitutive expression both were observed experimentally to lead to differential sensitivity to ABA [40 , 83 , 105] . This differential sensitivity would be captured by the model if it was assumed that ABA indirectly activates SCAB1 and indirectly inhibits GEF1/4/10 . In general , the source nodes are promising candidates for experimental testing of whether they are in fact regulated by nodes of the ABA signaling network . Incorporation of 3 states also would be able to separate the normal activity of a node from reduced activity due to an intervention and increased activity due to a different intervention . This increased resolution would capture differential sensitivity to the knockout or constitutive activation of a source node and would generally decrease the number of cases in which a simulated intervention leads to a close to wild-type response . While a Boolean regulatory function has a single activity threshold ( between the OFF and ON state ) , a 3-state regulatory function involves 2 activity thresholds . The construction of such functions would require additional node-level , quantitative ( dose-dependent ) experimental information beyond what is currently available , but it is an important area for future work . Many of the model’s predictions are unexpected , illustrating how models can aid human intuition . For example , impairment of ROS-activated Ca2+ channels in null mutants of GPA1 [51] would reasonably lead to the hypothesis that Ca2+c response to ABA would be impaired in gpa1 mutants; however , our model , which integrates all sources contributing to Ca2+c elevation , made the counterintuitive prediction of wild-type Ca2+c response , and this prediction was confirmed upon wet-bench assessment ( Fig 4A and 4B ) . One particularly interesting model prediction is the existence of multiple states of the network ( see S6 Table , S7 Table , S13 Table ) , all consistent with lack of closure in the absence of ABA . Such states may provide entry points for regulation by other stimuli to which multisensory guard cells also respond , including light , CO2 concentrations , and humidity . Indeed , attractor analysis of our previous model of light- and CO2-induced stomatal opening [20 , 143] indicates the possibility of multiple long-term behaviors of nodes involved in K+ efflux in open stomata , consistent with the results reported here . Our comprehensive analysis of the effect of node knockouts and constitutive activity in the presence of ABA allows the evaluation of implicit assumptions frequently encountered in the literature . Summarizing the results of Table 3 ( augmented by our experiments reported here ) in terms of whether an intervention ( knockout or constitutive activation ) yields an increased or decreased sensitivity to ABA yields several patterns ( S15 Table ) . These patterns suggest that observing a differential ABA response for a node intervention ( e . g . , ABA insensitivity upon knockout ) is not sufficient to infer ( without testing ) an opposite effect for the opposite intervention ( e . g . , to infer ABA hypersensitivity upon constitutive activation ) . Current results also show that simply observing state change of an internal node during ABA-induced closure is not sufficient to infer ( without testing ) a key role for this node in the process . Our analysis also emphasizes the absence of data on the extent to which constitutive availability of a signaling metabolite or constitutive activation of a protein influences stomatal apertures either independently of or in combination with a closing signal such as ABA . A main reason is that in many cases , the genetic or pharmacological manipulation that would confer constitutive activation of an enzyme is simply unknown . Our model highlights that protein overexpression does not necessarily suffice to confer constitutive activity if the protein is posttranslationally regulated , e . g . , by metabolites or other proteins . While it is well documented that experimental application of ROS , NO , or Ca2+ can each induce stomatal closure , the effects of activation of these internal nodes in combination with ABA exposure have received little attention . We tested 2 such instances ( Ca2+c and ROS , Fig 4C and 4D ) and found agreement between simulated and experimental results . Our model makes explicit predictions for all such combinatorial effects ( S8 Table ) that can be tested in future experiments . Multiple stimuli are present in the natural environment; for example , both ABA and elevated CO2 induce ROS and NO production in guard cells [144] . Modeling of these and other combinatorial scenarios will provide information that ultimately can be applied to improve crop fitness [145] under the increasing threat of drought [146 , 147] arising from elevated atmospheric concentrations of CO2 and other greenhouse gases . Finally , our approach illustrates the value of discrete dynamic modeling to identify both robustness and vulnerabilities of a complex , nonlinear system , which typifies many biological phenomena .
In the network , components of the ABA signal transduction pathway are represented as nodes and pairwise interactions are denoted by edges . Most edges in this network are directed to indicate the orientation of regulation or information transfer . The exceptions are the edges that represent binding among the subunits of the heterotrimeric G protein ( GPA1 , AGB1 , and AGG1 ) , which are undirected ( symmetrical ) and are considered bidirectional in the network analysis . A positive edge ( terminating in an arrowhead ) indicates positive regulation or information transfer from the starting node ( regulator ) to the end point of the edge ( the target ) . A negative edge ( terminating in a black circle ) indicates the repression or inhibition of the target by the regulator . Nodes in the ABA signal transduction network model originate from the known components reported in the literature through 2015 . From over 120 published articles , we identified evidence of interactions or causal effects between signaling components ( summarized in S1 Table ) . This information comes from 2 major types of experimental evidence . Protein–protein interactions discovered through bimolecular fluorescence complementation ( BiFC ) and/or yeast two-hybrid experiments; assays of phosphorylation status indicative of relationships between kinases , phosphatases , and their substrates; and assays wherein coexpression of an ion channel and an upstream regulator in Xenopus oocytes results in altered ion channel behavior indicate a direct interaction between 2 components . For example , the OST1 kinase activates SLAC1 anion channels by direct phosphorylation [148 , 149]; accordingly , the network contains a direct positive regulatory edge from OST1 to SLAC1 . By contrast , a genetic mutation or pharmacological treatment of a component that alters the status of a downstream component indicates that there is a causal relationship between the 2 components of the biological system , but because it does not assess physical interaction , it does not reveal whether the relationship is direct or indirect . We considered the existence of evidence of physical interaction ( or lack thereof ) to categorize these relationships as direct or indirect . For example , exogenous Ca2+ application indirectly leads to cytosolic alkalinization ( pHc increase ) ; accordingly , the network includes an indirect positive regulatory edge between Ca2+c and pHc . To keep the network parsimonious , indirect causal relationships that could be explained by a path of direct edges ( i . e . , a sequence of direct regulatory relationships ) were considered for elimination . Edge elimination was performed by employing a network reduction technique called binary transitive reduction implemented in the software NetSynthesis [140] . In each case , we consulted the biologically relevant literature to make sure that there is no evidence of a direct interaction between the participants in the causal relationship that was a candidate for elimination . In some cases , the literature indicates the effect of genetic mutation or pharmacological treatment of a component on a process rather than on a target node , which results in a causal relationship among 3 nodes , e . g . , “X promotes the process through which A induces B . ” Following [16] and [140] , we used the most parsimonious interpretation of this relationship . For example , if it is known that A activates X directly or indirectly , we inferred that X activates B . An example of a signed version of this inference is ABI1—● pHc , based on the evidence that ( i ) ABI1 inhibits ABA-induced pHc increase [77] and ( ii ) ABA , via its binding to RCARs , leads to the inhibition of the phosphatase activity of ABI1 [21 , 22] . Conversely , if it is known that X activates B through a direct interaction , we inferred that A activates X . An example of such an inference is ROS→NO , based on the evidence that ( i ) in nia1/2 double knockout plants , ROS fails to induce NO production and ( ii ) NIA1/2 are the enzymes that catalyze the production of NO from NADPH [56] . The network is provided as a graphml file , readable into Cytoscape , in the GitHub repository https://github . com/krhyyme/ABA-Boolean-Network-Model . Network analysis was performed using the Python graph analysis library NetworkX [150] . Node degree quantifies the number of edges that belong to a particular node . The in-degree of a node is the number of edges oriented toward it . The out-degree is the number of edges that originate from it . Undirected edges are interpreted as bidirectional . Nodes with an in-degree of zero ( no incoming edges ) are called source nodes and nodes with an out-degree of zero ( no outgoing edges ) are called sink nodes . A path in a network is a sequence of adjacent edges . The SCC of a network represents the subnetwork whose nodes are connected by paths in both directions . Thus , every pair of nodes A and B in an SCC has both an A→→B path and a B→→A path . The in-component of a network is the set of nodes that can reach the SCC through paths , and the out-component is the set of nodes that can be reached from the SCC through paths . The betweenness centrality of a node i quantifies the node’s participation in directed paths that start and end at nodes other than i . We used the betweenness centrality measure based on shortest paths ( paths with the smallest length ) . Specifically , the betweenness centrality of node i is g ( i ) =∑j≠i≠knjk ( i ) njk , where njk is the number of shortest paths that start from node j and end in node k , both of which are different from node i , and njk ( i ) is the number of shortest paths that start from node j and end in node k and contain node i . In a Boolean model , each node is characterized by 2 possible states: ON or 1 ( interpreted as a higher-than-threshold abundance or activity ) and OFF or 0 ( lower-than-threshold abundance and activity ) . For example , the ON state of the Ca2+ c node indicates that the concentration of Ca2+c is sufficiently higher than the resting level , such that it can activate CPKs and other downstream targets . Biologically known initial states are implemented as ON or OFF . For example , secondary messengers known to be produced in response to ABA ( e . g . , ROS , NO ) and signaling proteins known to be activated in response to ABA ( e . g . , RCARs , OST1 ) are assumed to be OFF in the initial state [1 , 3 , 98–101] . The PP2C protein phosphatases are assumed to be initially ON [103] . The initial states of the 26 nodes with no such information are randomly chosen . The initial state of each node is summarized in S6 Table . The future state of a node is determined by the current state of its directly upstream regulators and is expressed as a Boolean regulatory function . A node with a single direct regulator is characterized by one of 2 types of single variable Boolean function: identity , used for positive regulators and meaning that the target is adopting the state of the regulator , and negation , used for negative regulators and meaning that the target adopts the opposite state as the regulator . Negation is represented by the Boolean operator “not . ” For nodes that have more than one direct regulator , the “or” operator is used if any of these regulators can independently activate the target; the “and” operator is used if all direct regulators are needed for activation . The choice of the most appropriate operator to use is determined by experimental evidence . If there is evidence that knockout of one regulator prevents a node from being activated , then the “and” rule is used to describe the set of regulators to which this behavior applies . For example , as ABA does not induce pHc increase in abi1-dominant mutants , we include “and not ABI1” in the regulatory function of pHc . If it is shown that simultaneous absence of a pair or set of direct regulators is required to prevent activation , then the “or” rule is used . In cases where adequate information is not available , “or” can also be used as the presumptive rule [151] . Below , we indicate the regulatory function of PA as an example . In this function , the node states are represented by the node names and an asterisk is used to denote the future state of a node: PA*=PCand ( PLDδorPLDα ) orDAGandDAGK PC is the substrate required by PLDα or PLDδ for PA production . DAG , a product of PLC activity , can be converted into PA by DAGK-mediated phosphorylation; thus , PC combined with either of the indicated PLDs or the combination of DAG and DAGK is sufficient for PA production and thus for the presence of PA after a time delay ( PA* ) . AGB1 and AGG3 are not included in the dynamic model , as there is no current evidence that they influence ABA-induced stomatal closure other than via GPA1 , with which they share an undirected edge [152 , 153] . ROP10 is also not included in the dynamic model: there is no current evidence that it influences ABA-induced closure; hence , it has no outgoing edges and therefore exerts no control on model outcome . Generally , the regulatory function of a node contains the node itself if and only if there is a self-loop in the network . There are 3 exceptions to this rule in the model . For pHc and S1P/PhytoS1P production , transient increases are observed following ABA exposure [47 , 70] but there is no causal or mechanistic understanding of the negative feedback regulation; hence , the self-regulation is included in the network but not included in the regulatory functions . In the case of vacuolar acidification , we assume the maintenance of an already-initiated state; hence , the regulatory function contains the node itself . Each model simulation starts from an initial condition ( initial state of the network elements ) corresponding to open stomata , specified in S6 Table and transitions through several states until it reaches a set of states in which it settles down , known as an attractor . An attractor can be a fixed point ( steady state ) or a set of states that repeat indefinitely ( a complex attractor ) , such as an oscillation . We simulate successive events by a stochastic update algorithm . Update of a node’s state means that its regulatory function is evaluated and the function’s output is adopted as the new state of the node . Because signaling networks contain a diverse set of components that act at many different timescales , asynchronous update algorithms are most appropriate for modeling signal transduction pathways [154] , particularly when , as in the present case , relative timescales of internal processes are largely unknown . We implemented the model using the Python software library BooleanNet [155] . We used the stochastic asynchronous update algorithm , in which the nodes are updated in a randomly selected order in each time step . Due to the stochasticity introduced by the update method and by the initial node states , we used 2 , 500 replicate simulations . This number of replicates ensures that the margin of error of the percentage of closure is less than 3% , with 95% confidence . In each simulation , we followed the evolution of the model for 30 time steps ( rounds of update ) , which was sufficient to reach an attractor in each modeled scenario . The outcome of the model is summarized as the percentage of simulations in which the node Closure is in the state 1 at each time step , which we refer to as the percentage of closure . A sample script using the BooleanNet library and the list of Boolean regulatory functions in the form readable by the script are provided in the GitHub repository https://github . com/krhyyme/ABA-Boolean-Network-Model . A stable motif is a special kind of SCC that can maintain a specific steady state of its constituent nodes regardless of the state of the rest of the network [96] . The stable motifs of a Boolean network determine its attractors: one can uniquely associate sequences of stable motifs ( stabilized in the order given by the sequence ) to each attractor . We use the method introduced in [96] , which maps the identification of stable motifs into finding SCCs with certain identifiable properties in an expanded representation of the Boolean network ( which includes the network’s Boolean functions as part of the network structure ) . A more detailed explanation of the criteria for identifying stable motifs and of the procedure for creating the expanded representation of the Boolean network can be found in [96] . A Java library that implements the stable motif analysis and attractor determination is available in the GitHub repository https://github . com/jgtz/StableMotifs . We simulate knockout of a node by setting its state as OFF ( 0 ) in the initial condition and maintaining this state throughout the simulation . We simulate constitutive activity of a node by setting its state as ON ( 1 ) in the initial condition and maintaining this state throughout the simulation . Thus the knocked-out or constitutively active nodes are not updated . In the most severe case of defect , the percentage of closure stays exactly or very close to 0% for the entire duration of the simulation , similar to the wild-type simulations in the absence of ABA . It can be concluded that knockouts in this category , e . g . , ost1 , lead to insensitivity to ABA . A less severe defect is indicated by a percentage of closure that stabilizes at a value between 0% and 100%; we call this category “reduced sensitivity” to ABA . Following [16] , we define the cumulative percentage of closure ( CPC ) as the summed fraction of simulations in which Closure = 1 over the 30 time steps . Thus , CPC ranges from 0 ( if the percentage of closure were 0% at every time step ) to 30 ( if the percentage of closure were 100% at every time step ) . Comparison of the CPC corresponding to the knockout or constitutive activity of a node with the CPC of wild type can determine whether the knockout led to a decrease in sensitivity or , possibly , an increase . To determine whether a measured difference in CPC values is significant , especially in light of the fact that 2 , 500 simulations represent a sampling of the system’s total possible trajectories , we benchmark with the fact that certain cases of simulated knockout or constitutive activity are equivalent to a wild-type simulation . Specifically , simulated constitutive activity of any of 20 source ( unregulated ) nodes that are assumed to be ON in the model is equivalent with a wild-type simulation , as is simulated knockout of either of the 2 source nodes that are assumed to be OFF in the model ( see Table 1 and S2 Text for the identity of these nodes ) . We use these 22 ensembles of 2 , 500 simulations to delineate the CPC range of wild-type equivalent simulations . Interventions that lead to a higher CPC value than the highest value in the wild-type equivalent range ( 24 . 08 ) are classified as ABA hypersensitive , and interventions that lead to eventual closure but with a lower CPC value than the lowest value in the wild-type equivalent range ( 23 . 93 ) are classified as ABA hyposensitive . Interventions that lead to a CPC value within the wild-type equivalent range are classified as close to wild type . In the absence of ABA and initial activity of the PP2C protein phosphatases , the threshold between close to wild type and slightly increased response is CPC = 0 . 002 ( Table 4 ) . In the absence of ABA and initial inactivity of the PP2C protein phosphatases , the threshold between slightly decreased response and close to wild type response is CPC = 9 . 66 and the threshold between close to wild type response and increased response is CPC = 10 . 62 ( S12 Table ) . Wild-type Arabidopsis of the Columbia ( Col ) accession and previously described [156] gpa1-3 and gpa1-4 null mutants in the Col background were grown in a growth chamber with 8-h light/16-h dark cycles with a light intensity of 150 μmol/m2s and temperatures of 21°C during the light period and 19°C during the dark period . To measure Ca2+c in response to ABA , epidermal peels from fully expanded leaves from 4-week-old T2 plants stably transformed with a YC3 . 6 Ca2+c reporter construct [157] driven by the GC1 guard cell-specific promoter [158] were imaged using a Zeiss LSM510 confocal microscope as described [159] . Epidermal peels were mounted on a coverslip chamber with medical adhesive ( Hollister , Libertyville , IL , USA ) and incubated with 20 mM KCl , 50 μM CaCl2 , 5 mM Mes-Tris , pH 6 . 15 under illumination ( 150 μmol/m2 s white light ) for 3 hours . The coverslip chamber was placed on the stage of the confocal microscope . The YC3 . 6 fluorescence signal from individual guard cells was observed by a laser scanning confocal microscope ( LSM 510 Meta; Carl Zeiss , Thornwoood , NY , USA ) using a C-Apochromat 40X/1 . 2 W corr water immersion objective with the 458-nm line of the argon laser for excitation of YC3 . 60 and 484–505 nm and 526–536 nm for the detection of CFP and Venus ( for FRET ) of YC3 . 60 , respectively . After 5 minutes , epidermal peels were treated with 10 μM ABA , then images were taken at 10-second intervals for 50 minutes . The ratios of Ca2+-dependent ( FRET/CFP ) fluorescence intensities were calculated using ImageJ software . Ca2+c transients were defined as increases that were at least 0 . 1 above the baseline , following the method of Ye et al . [160] . To measure the effects of Ca2+ and ROS provision on ABA-induced stomatal closure , we modified a previously established protocol [161] . Before initiation of the light cycle , fully expanded leaves from 4–5-week-old plants were excised . Abaxial epidermes were peeled off using forceps and incubated in solution ( 20 mM KCl , 5 mM MES/KOH , pH 6 . 15 ) under white light ( 175 ± 25 μmol m-2 sec-1 ) for 3 h to induce stomatal opening . Different treatments , i . e . , solvent control ethanol ( 0 . 1% ) , ABA ( final concentration at 2 μM ) , CaCl2 ( final concentration at 50 μM ) , and H2O2 ( final concentration at 0 . 1 μM ) were then added to the solutions , as indicated . The abaxial epidermes were imaged at the indicated time points ( 1 h and 2 h ) by light microscopy ( Nikon Diaphot 300 ) and attached camera ( Nikon E990 ) . Stomatal apertures were measured by analysis of the digital images using ImageJ ( National Institutes of Health , USA ) . Experiments were performed blinded and each experiment was repeated 3 times , with 64 stomatal apertures measured for each treatment . In the 97 cases with prior experimental documentation of the effect of a node’s knockout or constitutive activation on the closure response ( Table 3 and Table 4 ) , 57 out of 73 simulations of individual node knockouts or constitutive activations in the presence of ABA ( boldface nodes in Table 3 ) and 13 of 24 simulations of constitutive activation of nodes in the absence of ABA ( boldface nodes in Table 4 ) match experimental evidence . Our model accurately captures the effect of knockout of an internal node on the status of a second internal node in the presence of ABA in all 12 extant experiments reported in S8 Table . Thus , the model’s validation rate with prior results is 82/109 = 0 . 75 . The 3 new experiments reported here increase both the cases of agreement and the denominator by 3 . The hypothesis that Ca2+ increase leads to the inactivation of the four PP2Cs yields 10 new cases of agreement . Thus , the augmented model’s validation rate is 95/112 = 0 . 85 .
|
Guard cells , located in pairs on the surface of plant leaves , circumscribe microscopic pores called stomata , through which plants take up gaseous carbon dioxide that will be fixed by photosynthesis into sugars . However , plants also inevitably lose water vapor to the atmosphere through open stomata . Under drought conditions , the plant hormone abscisic acid ( ABA ) causes volume changes in guard cells that result in stomatal closure , thereby restricting water loss . Given the paramount importance of drought tolerance for plant survival , it is important to understand the cellular mechanisms underlying guard cell response to ABA , and over 100 studies in the literature have addressed this topic . We synthesized this information into a network that contains 84 cellular components and 156 interactions between them and then applied logic-based analyses to predict how these components coordinately transduce the ABA signal . We identified several positive feedback loops and mutual inhibition loops that can lead to sustained activity of their constituent components in the presence , or absence , of ABA . Control of these loops , for example , by other stimuli present in the natural environment , may sensitize the system to ABA . We validated some of the novel predictions from our model with new experiments .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"plant",
"anatomy",
"stem",
"anatomy",
"guard",
"cells",
"enzymes",
"plant",
"cell",
"biology",
"ions",
"enzymology",
"signaling",
"networks",
"anions",
"stomata",
"phosphatases",
"plant",
"science",
"network",
"analysis",
"physical",
"chemistry",
"computer",
"and",
"information",
"sciences",
"network",
"motifs",
"proteins",
"chemistry",
"leaves",
"adenosine",
"triphosphatase",
"centrality",
"biochemistry",
"plant",
"cells",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences"
] |
2017
|
A new discrete dynamic model of ABA-induced stomatal closure predicts key feedback loops
|
Protein-protein interactions ( PPIs ) are of central importance for many areas of biological research . Several complementary high-throughput technologies have been developed to study PPIs . The wealth of information that emerged from these technologies led to the first maps of the protein interactomes of several model organisms . Many changes can occur in protein complexes as a result of genetic and biochemical perturbations . In the absence of a suitable assay , such changes are difficult to identify , and thus have been poorly characterized . In this study , we present a novel genetic approach ( termed “reverse PCA” ) that allows the identification of genes whose products are required for the physical interaction between two given proteins . Our assay starts with a yeast strain in which the interaction between two proteins of interest can be detected by resistance to the drug , methotrexate , in the context of the protein-fragment complementation assay ( PCA ) . Using synthetic genetic array ( SGA ) technology , we can systematically screen mutant libraries of the yeast Saccharomyces cerevisiae to identify those mutations that disrupt the physical interaction of interest . We were able to successfully validate this novel approach by identifying mutants that dissociate the conserved interaction between Cia2 and Mms19 , two proteins involved in Iron-Sulfur protein biogenesis and genome stability . This method will facilitate the study of protein structure-function relationships , and may help in elucidating the mechanisms that regulate PPIs .
Protein-protein interactions ( PPIs ) are critical to virtually all biological processes , from the formation of cellular macromolecular structures and enzymatic complexes to the regulation of signal transduction pathways . Hence , detailed analysis of these interactions is essential for understanding biological phenomena . Many experimental methods have been developed in the past decade for mapping PPI networks . For eukaryotes , the most popular experimental platform for large-scale analysis of PPIs is the yeast , Saccharomyces cerevisiae . Protein complexes have been characterized in yeast using affinity purification followed by mass spectrometry ( AP/MS ) [1] . Other approaches such as high-throughput yeast two-hybrid ( Y2H ) analyses [2] , fluorescence resonance energy transfer ( FRET ) [3] , and protein-fragment complementation assay ( PCA ) [4] have been used to identify binary interactions . The systematic unbiased utilization of these methods led to various maps of the protein interactome of several model organisms [2] , [5] , [6] . There are many changes in protein complexes that occur as a result of genetic and biochemical perturbations . These include changes in protein levels and localization , and posttranslational modifications that may alter the bond between interacting partners ( for more details see Figure S1 ) . One drawback of the above-mentioned studies is that they were performed on a single genetic background , and thus , potential genetic modifications that can dissociate the interaction remain unidentified . The identification of trans-acting mutants that dissociate a particular PPI is valuable for unraveling important regulatory mechanisms , and for defining the biological effect of a specific perturbation . Despite the great importance of such data , none of the available experimental systems allow the systematic detection of such dissociation events . Rather , they are limited to positive selection of protein-protein association events . PCAs are a family of assays for identifying interactions between protein pairs [7] . In this strategy , PPIs are measured by fusing each of the proteins of interest to complementary N- or C-terminal peptides of a reporter protein . Upon interaction of the two fusion proteins , the reporter protein fragments are brought into proximity , thus reconstituting the activity of the reporter , such that it provides a detectable signal [8] . PCAs have been created using many different reporter proteins and thus enable different types of readouts . A PCA based on a mutated version of the murine dihydrofolate reductase enzyme ( mDHFR ) was adapted to the yeast Saccharomyces cerevisiae [5] , [9] . In this case , mDHFR is split into two complementary fragments ( F[1 , 2] , and F[3] ) and inserted at the C-termini of the two genes of interest . The functional copy of the mDHFR confers resistance to the DHFR inhibitor , methotrexate ( MTX ) , which inhibits the native S . cerevisiae DHFR . Thus , the interaction between the two proteins of interest can be detected as cell growth on media in the presence of MTX . This approach has been recently used to systematically identify nearly all possible binary combinations of yeast proteins , and has led to the identification of 2770 interactions that represent an in vivo map of the yeast PPI network [5] . In this report , we describe the “reverse-PCA” ( rPCA ) system , which combines the PCA and the synthetic genetic array ( SGA ) methodologies [10] . In the SGA approach a MATα query strain carrying any genetic element ( or any number of genetic elements ) marked by a selectable marker ( s ) can be crossed to an ordered array of mutants collection ( MATa ) . The resulting array of heterozygous diploids can then be sporulated , and a set of desired MATa haploid meiotic progeny cells can be subsequently selected , exploiting a cleverly designed SGA haploid selection marker ( HSM ) [10] . The combination of these approaches allows us to systematically detect trans-acting proteins that , when mutated , dissociate specific PPIs in S . cerevisiae . We demonstrate the feasibility of this approach by identifying previously characterized and novel proteins that dissociate the conserved interaction between Cia2 and Mms19 , two proteins that play a role in Iron-Sulfur cluster biogenesis , and in genome stability [11] , [12] . Our results validate this method as an efficient and scalable approach that is expected to promote biological discovery .
In the yeast mDHFR PCA , the interaction between two proteins of interest ( designated here as X and Y ) allows yeast cells to grow in the presence of MTX . We designed a system to uncover mutants that impair the growth on MTX containing media . The query strain for the rPCA was chosen from a dataset of 2 , 770 interactions that were recently identified by mDHFR PCA [5] . In this strain , the two complementary fragments of mDHFR ( F[1 , 2] , and F[3] ) are each fused to the C-terminal of X and Y ( X-F[1 , 2] , and Y-F[3] ) . Using SGA methodology [10] , we crossed this strain to ordered arrays of three libraries which together encompass mutations in every yeast gene ( Figure 1A , step-1 ) . The first library was the yeast deletion library [13] , which consists of ∼4500 strains , each harboring a deletion in a single non-essential gene . The second and third were two complementary libraries , which together consist of ∼1000 strains , each expressing a temperature sensitive ( Ts ) allele of an essential gene [14] , [15] . The resulting array of heterozygous diploids was then induced to undergo meiosis ( Figure 1A , step-2 , 3 ) , and the set of desired MATa haploid meiotic progeny cells was subsequently selected , exploiting the SGA haploid selection marker ( HSM ) ( Figure 1A , step-4 ) . These steps allowed the recovery of a library of ∼6000 haploid meiotic progeny , each harboring both X-F[1 , 2] , and Y-F[3] fusion proteins , on the background of a mutation in a single yeast gene . This array was used as a control , and provided samples of mutants that affect growth rate per-se under normal conditions ( Figure 1A , step-5 ( left ) , indicated by a dashed arrow ) . These haploids were also transferred to a second plate to select for MTX resistance . Colony growth was assessed using an automated computer-based scoring system . This system analyzes digital images of colonies to generate an estimate of the relative growth rate based on pixel density [16] . Impaired PPI was scored when the colony size on the MTX containing medium was significantly smaller than that on the control array ( Figure 1A , step-5 ( right ) , indicated by a black arrow , and Figure 1B ) . An initial indication of the feasibility of rPCA could be obtained by demonstrating that factors that alter the formation of a specific PPI would affect resistance to MTX . To test this , we selected a specific PPI from the yeast DHFR network ( Cia2:: F[1 , 2]; Mms19:: F[3] ) [5]; these two proteins have conserved roles in the biosynthesis and delivery of the iron-sulfur ( Fe-S ) cofactors [11] , [12] . We then analyzed the effect of a cis-acting point mutation in Cia2 ( cia2- E208G::F[3] ) , which we recently found ( unpublished data ) to impair its interaction with Mms19 . The results show that while the Cia2-E208G mutant is stably expressed ( Figure S2A ) , there is a clear difference in the resistance of the mutated cells to MTX , relative to the control , and indicate that the mDHFR activity can be eliminated by abrogating the protein-protein interaction ( Figure 2A ) . Furthermore , in order to provide evidence demonstrating the specificity of the effect of this mutant on the interaction with Mms19 , we dissected tetrads originating from a strain heterozygous for a knock out in the essential gene CIA2 ( cia2::KmX/CIA2 ) , and harboring a URA3 marked centromeric plasmid expressing the CIA2-E208G mutant . The results show that the CIA2-E208G mutant can support the growth of haploid spores knocked-out in the endogenous CIA2 ( Figure S2B ) , and suggests that the E208G replacement does not have a general effect on Cia2 function , and that the impaired interaction with Mms19 is likely to be specific . As a further control , we altered the highly reactive cysteine in Cia2 to alanine ( C161A ) , which in contrast to the E208G allele was previously shown to cause lethality [17] . As shown in Figure 2A Cia2-C161A::F[3] construct still interacted with the MMS19::[F3] fusion protein , which supports the idea that E208G mutation is specific for Mms19 . Next , we asked whether our system provides a quantitative readout of the interaction levels of the proteins studied . We compared growth on MTX of haploids carrying the interacting pair Cia2:: F[1 , 2] and Mms19:: F[3] with isogenic strains hetero- or homozygous for the fusion proteins . While in the haploids and homozygous diploids , only the fusion proteins are available , in the heterozygous diploids , the untagged proteins ( Cia2 and Mms19 ) can compete for binding with those that are tagged , and we therefore expected the level of reconstituted mDHFR to be reduced . Indeed , the results show a significant reduction in the ability to grow on MTX of the heterozygote , compared to the homozygous strain ( Figure 2B , left panels ) . Furthermore , since the rPCA procedure is based on the evaluation of colony size , we repeated the same experiment , this time by recording colony sizes subsequent to robotic pinning . The expected fourfold reduction in reconstituted mDHFR in the heterozygous diploid strain was associated with similar fold reduction in colony size ( Figure 2B , right panels ) . Taken together , these results suggest that colony size on MTX containing media provides a quantitative readout indicating the changes in PPIs in-vivo . The range of the signal detectable in rPCA should depend on the quantity of complexes formed , which in turn is determined by the reduction in the abundance of the proteins studied , and their affinity for each other on various genetic backgrounds . We next scaled up our procedures , to test the rPCA method described above for a total of four different screens . The logic behind our choice , to systematically identify mutants that dissociate the PPI between Cia2 and Mms19 , was based on our long-term interest in pathways that play a role in genome stability . These proteins are part of a well-characterized and conserved complex that plays a key role in DNA metabolism , through a defined wide range of known and putative protein interactions ( see below for more details ) . Furthermore , it was shown that a third member of this complex , Cia1 , is able to bridge the complex , and therefore could serve as an important positive control [12] , [18] , [19] . The three other screens were mainly used to identify and eliminate potential sources of false positives ( see below ) . In general , the query strains were screened at least two times against collections of ∼6000 mutants in both the non-essential and essential genes , which together cover almost the entire yeast genome . Screens were carried out in 1536 density format with four replicate copies of each mutant on the array , allowing four repeats for each screen . For the temperature sensitive mutants of essential genes , the experiments were performed at three different temperatures . The degree of interaction perturbation was quantified by calculating the ratio of colony size on medium with and without MTX . We measured the expected variability in colony size by calculating the standard deviation in the Cia2 and Mms19 screen , for each of the genes , based on the four replicate copies of each mutant on the array ( Table S3 ) . The standard deviation of temperature sensitive mutants was calculated for the optimal temperature selected for each particular gene ( for more details see Materials and Methods , under the “Genome-wide rPCA screen , data filtering” ) . The average standard deviation was estimated at 0 . 0425 . The results obtained following data filtering ( see Materials and Methods for more details ) are presented in Figure 2C and 2D . We chose a threshold of 0 . 1 , which represents ∼1% of the assayed genes: All mutants below this threshold are considered to affect the studied PPI . Results were analyzed using the Gene Ontology ( GO ) Term Finder ( http://db . yeastgenome . org ) at the Saccharomyces Genome Database to look for terms enriched among this mutant set . We also accounted for two potential sources of false positives in an rPCA screen: 1 ) mutants in genes that specifically participate in DHFR biogenesis , and thus impair growth on MTX medium even when the mDHFR fragments still associate , and 2 ) mutants that have general effects on the interacting proteins , such as genes that affect protein biosynthesis , trafficking , RNA processing etc . , and are therefore expected to be identified as positive hits in most screens . To control for these factors , we first generated and screened a strain harboring a fused F[1 , 2] and F[3] mDHFR cassette . This cassette provides complete resistance to MTX independently of any PPI . Next , we performed additional rPCA screens using PPIs unrelated to the Fe-S cluster biogenesis pathway . These screens included PPIs between the 26S proteasome subunits Rpn5::F[1 , 2] and Rpn11::F[3] , and the pair of histone Htb2::F[1 , 2] and nucleosome remodeling protein , Nhp6a::F[1 , 2] ( the complete results obtained from these screens are provided in Tables S4 , S5 , S6 ) . To ensure robustness of the false positive selection , we used a threshold of 0 . 2 for Cia2-Mms19 , Htb2-Nhp6a , and Rpn5-Rpn11 screens . For the mDHFR cassette screen , a threshold of 0 . 25 was chosen , since the mDHFR strains are significantly less sensitive to MTX . For example , APN1 , a classic expected false positive ( see below ) , had a score 0 . 09 in the Cia2-Mms19 screen and 0 . 24 in the mDHFR screen . Genes that were considered false positives were those that passed the defined threshold in all three of the screens . As expected , in addition to APN1 , we also identified FOL1; both were previously shown to play a role in the folic acid biosynthesis pathway , and therefore impaired DHFR biogenesis [20] , [21] . Moreover , these screens also allowed us to eliminate 27 additional mutants that passed the threshold , and therefore were considered non-specific ( shown in Table S7 ) . Alongside genes that were categorized as non-specific in the screen with Htb2::F[1 , 2] and Nhp6a::F[1 , 2] , we were also able to identify genes that play a specific role in chromatin biology , and were previously shown to have genetic or physical interactions with Htb2 and Nhp6a ( Table S1 ) . One example is Htz1 , histone variant H2AZ , that is exchanged for histone H2A in nucleosomes by the SWR1 complex . This protein was previously identified as a physical interactor of Htb2 and Nhp6a [22]–[24] . Furthermore , we also found Swr1 , the catalytic subunit of the SWR1 complex and the main scaffold for the assembly of the complex , among the hits from this screen . This protein also physically interacts with Htb2 [24] . When extending the threshold to 0 . 2 , we could identify several more candidates with biological significance . For example , Vps71 and Vps72 are two additional subunits of the SWR1 complex; Rsc8 , Rsc9 , Swi6 and Arp7 are components of two additional remodeling complexes , RSC and SWI/SNF , which show a synthetic growth defect with Nhp6a [25] , [26] , and also facilitate the binding of Nhp6a to nucleosomes [26] . This may be consistent with the fact that temperature sensitive alleles of RSC show a perturbation of the Nhp6a-Htb2 interaction . In contrast to the screen with Htb2 and Nhp6a , in the case of Rpn5::F[1 , 2]-Rpn11::F[3] , we could not detect meaningful biological connections ( Table S5 ) . Apparently , there are cases in which the PPIs are direct , without the need of any modifications . In such screens , we expect to identify genes that affect protein biogenesis in general and not hits with direct relevance to the interacting protein pair . Furthermore , cells can show robustness to the loss of some protein complexes , while being highly sensitive to the loss of others [27] , [28] . Rpn5 and Rpn11 , two essential proteasomal lid subunits , might represent one end of this spectrum . Essential protein complexes such as the proteasome are robust to genetic perturbations because the deletion of a subunit can be buffered by the modification of PPIs by other subunits , particularly by paralogous proteins . This functional compensation would likely lead to a relatively stable and functional alternative configuration . This notion is supported by previous studies showing that essential protein complexes consist of redundant subunits that render them robust to genetic perturbations [27] , [28] , and others showing that abundance of one paralogous protein is increased in response to the deletion of the other , at the levels of both transcript and protein abundances [29] , [30] . As mentioned above , Cia2 and Mms19 are proteins with conserved roles in the biosynthesis and delivery of the iron-sulfur ( Fe-S ) cofactors . Fe-S clusters are small inorganic cofactors found in hundreds of proteins that are required in virtually all phyla of life from bacteria to humans , and serve in electron transfer , enzyme catalysis , regulation of gene expression , and stabilization of protein structures . The biosynthesis of cellular Fe-S clusters is a complex process , starting at the mitochondria , which harbor the iron-sulfur cluster ( ISC ) assembly machinery . Mitochondria contribute to the maturation of cytosolic and nuclear Fe-S proteins as they export to the cytosol a still unidentified , sulfur-containing component which is used to assemble a Fe/S cluster on scaffold proteins . Cia2 and Mms19 form the part of the conserved cytoplasmic iron-sulfur protein assembly ( CIA ) machinery that specifically transfers and inserts the Fe-S cluster into target apoproteins ( for review [31] ) . The rPCA screen between Cia2::F[1 , 2] and Mms19::F[3] identified 56 mutants with scores below the threshold value ( Figure 2D and Table S2 ) . The 56 top hits from the primary screen were reconfirmed by re-arraying on the control plates in 16 replicate copies , and then pinning on the MTX containing media ( Figure S2C ) . We retested the candidates that were selected for further analysis by backcrossing to a wild type strain . Following meiosis and tetrad dissection , we confirmed that ( 1 ) the sensitivity to MTX segregates in a Mendelian manner ( 2∶2 ) , indicating that it depends on a single gene mutation; ( 2 ) MTX sensitivity is linked to URA3 , or kanMX ( which were introduced as markers for the Ts alleles and the deletion mutants respectively ) , and therefore cosegregates with the mutated gene . Indeed , the selected spores that resulted from the backcross , show clear sensitivity to MTX at the semi-permissive temperatures ( Figure 3A ) . Based on Gene Ontology ( GO ) Term Finder annotations , we found functional groups that are likely to play a general role in protein synthesis and maturation , such as transcription and RNA processing , or protein degradation . Based on the rationale for selecting the list of false positives ( see above ) , some of these genes passed the threshold in less than three of the screens , and therefore were retained as candidates that specifically impair the interaction between Mms19 and Cia2 . The genes represented in our list of false positives probably play a general role in protein biosynthesis and modification . Nevertheless , we speculate that the genes that can be categorized as “general findings” in each of our screens , and were not included in our list of false positives , probably play a specific role in the biological pathways that were tested in this study . Moreover , although it was recently shown that the CIA complex mediates the transfer and insertion of the Fe-S cluster into target apoproteins that play roles in DNA metabolism and genome stability ( see below ) , Fe-S clusters are also found in target proteins that participate in other fundamental biological processes such as ribosome biogenesis , enzyme catalysis , regulation of gene expression , stabilization of protein structures , etc . Although we believe that most of the genes are general findings ( see above ) , we cannot rule out the possibility that some are specifically related to iron-sulfur biology . In addition to the general functional groups , CIA1 was one of the top hits which affected the interaction between Cia2::F[1 , 2] and Mms19::F[3] , and served as an important quality control . Data from previous studies revealed that the CIA protein , Mms19 , forms a complex with other late-acting CIA subunits both in yeast and human cells , including Cia1 ( hCiao1 ) , and Cia2 ( hCia2; also termed FAM96B ) [11] , [12] , [18] . Hence , we expected to find that a component of the CIA complex would affect its assembly . We further confirmed this result biochemically by co-IP experiments in yeast and mammalian cells . The potential role of Cia1 in mediating the interaction between Cia2 and Mms19 was examined by depleting the protein in a galactose- regulatable CIA1-FLAG strain , followed by IP . The expression of CIA1-FLAG was induced by growing the cells in galactose ( Gal ) . Then glucose ( Glu ) was added to the media to shut-off the expression , and samples were collected at the indicated time points for immunoblotting with anti-FLAG . The results ( Figure 3B-left ) show that the expression levels of Cia1 were reduced by 91% after 24 hours . An IP experiment on the sample depleted in Cia1 ( Figure 3B-right ) , shows that the depletion of CIA1 did not affect the endogenous levels of Mms19 and Cia2 , however , its depletion reduced the association of Mms19 with Cia2 . This may suggest that in yeast , Cia1 is an adaptor protein for Cia2 and Mms19 . In support of this function , we obtained similar results in a strain harboring Cia2 and Mms19 fused to the Myc and TAP tags , respectively ( Cia-Myc , and Mms19-TAP ) , and the original Ts allele of CIA1 , grown for 24 hrs at the semi-permissive temperature ( cia1-Ts ) . We also FLAG-tagged this allele at the C-terminus , and show that similarly to GAL1-CIA1 , under these conditions expression of Cia1-Ts protein is abolished ( Figure S3A ) . Furthermore siRNA-mediated depletion of CIAO1 in HeLa cells , led to almost four-fold decrease in the endogenous levels of hCia2 ( Figure 3C-left ) . Moreover , only 50% of the available hCia2 co-IPed with hMms19 , in the experimental sample ( siCIAO1 ) ( Figure 3C-right ) . These results suggest that in human cells , Ciao1 is a scaffold protein that also stabilizes hMms19 and/or hCia2 . Collectively these data suggest that Cia1 , like other members of the CIA complex , serves as a molecular scaffold that is required for its proper activity . Another quality control was the discovery that the interaction between Cia2::F[1 , 2] and Mms19::F[3] , is also affected by mutated genes that are involved in DNA metabolism , including components of the DNA replication and repair machinery , or those mediating mitotic chromosome segregation ( Figure 2D and Table S2 ) . Very recent studies have characterized Fe-S protein biogenesis as a key pathway for the maintenance of genomic integrity [11] , [12] . Studies in both yeast and human cells revealed that cytoplasmic Mms19 binds to multiple nuclear Fe-S proteins involved in DNA metabolism [12] , and suggested that the CIA complex targets Fe-S clusters to Fe-S apoproteins that are involved in genome stability . Indeed , components of the replication machinery , such as Pol2 , Pol3 , Pol31 , and Dna2 , that were shown to play an important role in genome integrity [32] , were shown to require Fe-S clusters for their complex formation and activity [33] , [34] . Our screen demonstrated that mutants in these specific Fe-S apoproteins ( Table S2 ) perturb the PPI between Cia2 and Mms19 . This may indicate that , similarly to other catalytic complexes such as the splicesosome [35] , assembly of the CIA complex occurs on its substrates . In order to provide further support for this possibility , we decided to confirm biochemically by co-IP experiments the results showing that the interaction between Cia2::F[1 , 2] and Mms19::F[3] is affected by the temperature sensitive alleles of DNA2 , and POL3 , two previously described Fe-S targets [33] , [34] , and by SPC24 , a protein involved in chromosome stability through its role in kinetochore clustering [36] . The potential role of these proteins in mediating the interaction between Cia2-Myc and Mms19-TAP was examined by growing the cells at the semi-restrictive temperature , followed by IP . The results show that while in all cases , Mms19 and Cia2 were stable , we could identify reduced association of Mms19 with Cia2 ( Figure S3B ) . The question that arises from this result is why depletion of a single substrate would affect the interaction , since there are many other substrates still present in the cell . While revising our manuscript , we became aware of a very recent manuscript by R . Lill and colleagues [37] . In this study , the human CIA2A ( FAM96A ) -CIA1 was identified as a component of the CIA machinery , together with CIA2B ( FAM96B ) -CIA1-MMS19 , which was used in our study . Importantly they show that CIA2B-CIA1-MMS19 specifically binds to and facilitates assembly of Fe-S targets involved in DNA metabolism and protein translation , while CIA2A-CIA1 assists different branches of Fe-S protein assembly . This result suggests that the pool of Fe-S proteins targeted by the CIA2B-CIA1-MMS19 complex is limited to specific functions . Thus , it is expected that specific mutations in a major pathway of the CIA2B-CIA1-MMS19 machinery ( such as the replisome ) would significantly reduce the levels of reconstituted mDHFR in the cells . The impressive list of proteins that were found to be associated with the CIA complex , and the enrichment of genome stability proteins [12] , suggests that numerous other Fe-S cluster containing proteins are required for the maintenance of nuclear genome integrity . It is currently unclear which of these interactions are physiologically meaningful , as their identification and functional characterization are still pending . The identification of known Fe-S proteins in our rPCA screen suggests that the additional proteins identified may represent novel Fe-S cluster-containing proteins , or proteins that play different roles in the assembly of these apoproteins into stable complexes that can be detected by the CIA machinery . Our screen also identified HEM1 , HEM13 , and HEM15 , genes involved in Heme biogenesis of , a prosthetic group that consists of an iron ion contained at the center of a large heterocyclic organic ring [38] . The assembly of both Fe-S clusters and Heme biogenesis are tightly regulated by iron homeostasis . We therefore further explored why heme mutants perturb the PPI between Cia2 and Mms19 . As a first step , we attempted to confirm the result by co-IP . Surprisingly , we found that the dysfunctional allele of HEM1 led to a significant decrease in the endogenous expression levels of both Cia2-F[3] and Mms19-[F3] ( Figure 4A ) . Detailed genetic analysis from previous studies have shown that in yeast cells with reduced heme synthesis , the transcription of selected iron regulon genes is decreased , and leads to lower iron uptake into the cell [39] , [40] . Given the finding that heme deficiency leads to reduced iron levels within the cells , we predicted that the endogenous levels of Cia2 and Mms19 might be regulated by iron availability . To test this , we assayed the expression of Cia2 , and Mms19 fused to the HA and TAP tags , respectively ( Cia-HA , and Mms19-TAP ) at various time intervals under iron deprivation conditions . Cia-HA and Mms19-TAP were degraded in the presence of the Fe ( II ) specific chelator , bathophenantholine disulfonic acid ( BPS ) ( Figure 4B ) . To rule out the possibility that this degradation was the result of cell death , we demonstrated that although their growth rate was delayed in the presence of BPS , cells were still in logarithmic growth at the indicated time points ( Figure 4C ) ; moreover , degradation of Cia2 and Mms19 was reversed by the addition of Fe ( +Fe in Figure 4B ) . Finally , we wished to provide further mechanistic insight into the degradation of Cia2 and Mms19 under Fe deficiency . Previous studies discovered a mechanism that mediates global posttranscriptional control of multiple components of Fe-dependent pathways to respond in a concerted fashion to Fe deficiency . In response to iron depletion , Aft1 , the major iron regulon , induces the transcription of a specific set of genes involved in the activation of iron uptake , mobilization of intracellular stores of iron , and metabolic adaptation to iron limitation [41] . One of the activated genes , CTH2 , coordinates this process by binding and targeting specific mRNA molecules to degradation [42] . We therefore used the approach described in Figure 4B , this time in cells deleted for CTH2 ( Figure 4D ) , or grown in the presence of the proteasome inhibitor MG132 ( Figure 4E ) . Interestingly , in contrast to other proteins such as Isa1 which play a role in the early steps of Fe-S cluster assembly , and were shown to be degraded in a CTH1/2 dependent manner ( used as a control in Figure 4D-bottom ) , Cia2 and Mms19 were clearly stabilized as a result of proteasome inhibition , though not in Δcth1/2 cells ( compare Fig . 4D , and 4E ) . It is well established that iron deficiency in the mitochondria leads to decreased mitochondrial Fe-S protein biogenesis , at the mRNA level [42] . Our rPCA approach reveals for the first time that members of the CIA complex with essential roles at the final steps of Fe-S cluster biogenesis are part of the Fe-dependent pathways which are negatively regulated by Fe levels in a proteasome-dependent manner . This finding suggests that additional genes involved in cellular iron uptake should have been detected in our screen . We believe that since the screen was performed under ample iron availability , many of the genes regulated by Aft1 were not activated , and therefore , single deletions did not affect iron uptake , nor the protein levels of Mms19 and Cia2 . Furthermore , the fact that almost none of these genes is essential [41] , suggests that since the cell is not robust to reduced iron-uptake , redundancy in gene function allowed the cells to tolerate the loss of single genes in our screen . In this case hitting one gene may slightly induce another transporter to ensure proper iron uptake . The PCA system represents an extremely powerful tool for the discovery of PPIs . The rPCA system extends this type of analysis to efficiently analyze trans-acting mutations that dissociate molecular interactions . We show that this technique is simple , and amenable to high throughput testing . Furthermore , PPI perturbations are detected in their endogenous environment in living cells and among proteins that are natively regulated . We validated this approach by demonstrating that previously characterized events leading to PPI dissociation can be reconstituted . Moreover , previously uncharacterized dissociation events could be specifically selected from large libraries using this genetic system . We believe that our study may lay the foundation for future comprehensive studies to study the effect of genetic perturbations on in-vivo PPI networks , and thus , is expected to promote further understanding of the eukaryotic interactome .
All the strains used in this study are isogenic to BY4741 , BY4742 , or BY4743 [43] . The relevant genotypes are presented in Table S8 . Myc , HA , and TAP fusions were generated using one step PCR mediated homologous recombination as previously described [44] . Strains from the PCA collection , containing different F[1 , 2] , and F[3] fusion proteins were provided by Stephen W . Michnick's laboratory ( University of Montreal ) . Galactose-regulatable CIA1 strain was a gift from Roland Lill's laboratory ( University of Marburg ) . The SGA markers were introduced into the query strains containing the fusion proteins of interest by crossing with Y7092 ( MATa can1Δ::STE2pr-his5 lyp1Δ ura3Δ0 leu2Δ0 his3Δ1 met15Δ0 ) [10] . Diploids were sporulated , and tetrads were dissected in order to select for the rPCA starting strain harboring the following drug resistances: X-F[1 , 2] , Y-F[3] ( clonNAT , and HygromycinB respectively ) , can1Δ::STE2pr-his5 ( canavanine ) and lyp1Δ ( thialysine ) . Saccharomyces cerevisiae strains were grown at 30°C , unless specified otherwise . Standard YEP medium ( 1% yeast extract , 2% Bacto Peptone ) supplemented with 2% galactose ( YEPGal ) , or 2% dextrose ( YEPD ) was used for nonselective growth . The media used in the rPCA analysis was a modification of the media used for SGA [10] . Drugs were added to the following final concentrations: canavanine ( 50 µg/ml , Sigma ) ; thialysine ( 50 µg/ml , Sigma ) ; clonNAT ( 100 µg/ml , Werner Bioagents ) ; G418 ( 200 µg/ml , Invitrogen Life Technologies ) ; methotrexate 200 µg/ml ( prepared from a 10 mg/ml methotrexate in DMSO stock solution , Bioshop Canada ) ; and HygromycinB ( 100 µg/ml , Calbiochem ) . Because ammonium sulfate impedes the function of G418 and clonNAT , synthetic medium containing these antibiotics was prepared with monosodium glutamic acid ( MSG , Sigma ) as a nitrogen source . Synthetic medium contained 0 . 1% Yeast nitrogen base w/o aa and Ammonium Sulfate , 0 . 1% Glutamic acid , 2% Dextrose , 0 . 2% amino acid mix , 4% noble agar ( purified Agar , Bioshop ) . The query strain were mated to the DMA on YEPD . Diploids were selected on YEPD supplemented with clonNAT and G418 . Diploids were sporulated on a medium containing 2% agar and 10 gr/L potassium acetate . For selection of MATa meiotic progeny carrying NatMX and , KanMX , and HygB markers , we used SD/MSG lacking histidine ( to select for expression of STE2pr-his5 ) , arginine , and lysine , and containing canavanine ( to select for can1Δ ) , thialysine ( to select for lyp1Δ ) , G418 ( to select for KanMX ) , clonNAT ( to select for NatMX ) , and HygromycinB ( to select for HygB ) [20 gr/L agar , 20 gr/L glucose , 1 . 7 gr/L yeast nitrogen base ( SD/MSG – His/-Arg/-Lys+canavanine/+thialysine/+clonNAT/+G418/+HygromycinB ) ] . This medium was supplemented with methotrexate for the final selection step . Since the average ratio slightly varied from plate to plate , the obtained ratios from each plate were normalized to the common mean ( 0 . 4 ) , which reflected the average impact of MTX on cell viability . Colony size on ( −MTX ) medium depends on the growth rate of the individual mutant strains . Data for mutants with colony size less than 70 pixels were eliminated , in order to remove dead and sick colonies , thereby avoiding experimental noise in data analysis . For temperature sensitive mutants , the experiment was performed at three different temperatures , and the ratio was calculated for each temperature and normalized to the common mean , in a manner similar to the deletion mutants . Given that each temperature sensitive mutant has an optimal temperature for revealing its phenotype , we obtained the optimal ratio for each mutant by choosing the temperature providing the minimal ratio , while only a temperature resulting in a colony size greater than 70 pixels on −MTX media was considered . In order to estimate the variation in colony size , we calculated the standard deviation for each of the genes , based on the four replicate copies of each mutant on the array . The standard deviation of temperature sensitive mutants was calculated for the optimal temperature selected for the particular gene . HeLa cells were cultured in DMEM medium supplemented with 10% fetal bovine serum and 2 mM L-glutamine in a 37°C humidified incubator containing 5% CO2 . siRNA duplexes targeting CIAO1 and non-targeting siRNA control were purchased from Dharmacon . Transient transfection of HeLa cells was performed using DharmaFECT 1 reagent as described by the manufacturer ( Dharmacon ) . HeLa cell pellets were lysed in RIPA buffer ( 20 mM Tris , pH 7 . 5 , 150 mM NaCl , 1 mM EDTA , 1% Nonidet P-40 , 0 . 5% sodium deoxycholate and 2 mM Na2VO4 ) . For immunoprecipitations , lysates were incubated with anti-MMS19 ( EUROMEDEX ) antibody for 16 h at 4°C . Then 25 µl of protein A-G-agarose-conjugated beads were added , and the mixture was incubated for 1 . 5 h at 4°C . Beads were recovered by centrifugation , washed two times with TGET buffer ( 20 mM Tris HCl pH 7 . 5 , 10% glycerol , 0 . 1% Triton X-100 , 1 mM EDTA ) supplemented with 150 mM NaCl and one time with TGET buffer supplemented with 75 mM NaCl . The beads were then boiled in SDS sample buffer for 5 min and briefly pelleted at 13 , 000 rpm before the supernatant was loaded for electrophoresis . A total of 5×108 logarithmically growing cells were washed twice with water and resuspended in 1 . 5 ml of ice-chilled buffer B60 ( 50 mM HEPES-NaOH [pH 7 . 3] , 0 . 1% Triton X-100 , 20 mM β-glycerophosphate , 10% glycerol , protease inhibitor mix ( Roche Biochemicals ) , 60 mM potassium acetate ) , and 1 . 5 g of ice-chilled glass beads was added . The tubes were vortexed eight times for 30 s with 30-s intervals on ice . After 10 min on ice , the lysate was decanted into ice-chilled 15-ml tubes and centrifuged for 20 min at 18 , 000×g at 4°C . A 500-µl volume of clarified lysate was incubated with 25 µl of prewashed protein A-agarose beads at 4°C for 1 h . The beads were pelleted , and 450 µl of the lysate was transferred to a tube containing 7 . 5 µl of anti-myc antibody ( Santa Cruz ) and incubated at 4°C for 2 h . Then 25 µl of prewashed protein A-G-agarose-conjugated beads were added , and the mixture was incubated for 1 h at 4°C . The beads were then washed successively seven times: four times with B60 adjusted to 100 mM potassium acetate and once each with B60 adjusted to 210 , 240 , or 270 mM potassium acetate . The beads were boiled in SDS sample buffer for 5 min and briefly pelleted at 13 , 000 rpm in an Eppendorf centrifuge before the supernatant was loaded for electrophoresis . Image-J software was used to quantitate the western blots data . The protein level was calculated relatively to the loading control used in each experiment . In immunoprecipitation analysis the percentage of the precipitated protein was calculated relatively to the total protein level as indicated by the whole cell extract . Other antibodies used in this study were purchased commercially: anti-TAP ( GenScript ) , anti-HA ( Covance ) , anti-DHFR-[F3] ( Sigma ) , anti-PGK1 ( Molecular probes ) , anti-Ciao1 ( Santa Cruz ) , anti-Cia2 Abcam ) , anti-alpha-Tubulin ( Abcam ) , anti-CPY ( Roche ) , anti-FLAG ( Sigma ) .
|
Protein–protein interactions ( PPI ) occur when two or more proteins bind together to form large molecular machines . The importance of PPIs led to the development of multitude technologies to detect them , and to the first maps of the protein interactomes . One important challenge in biology is to understand how protein complexes respond to genetic perturbations; however , in the absence of a suitable assay , such changes have been poorly characterized . Here , we present a novel systematic genetic approach ( termed “reverse PCA” ) , that demonstrates how the yeast protein complementation assay ( PCA ) , coupled with the synthetic genetic array ( SGA ) technology may be used to study the modulation of protein–protein interactions in-vivo in response to genetic perturbations . Our assay starts with a yeast strain in which the interaction between given proteins can be detected by resistance to the drug , methotrexate . Using the SGA technology , we can systematically identify yeast mutants that reverse this interaction . We were able to successfully validate this approach by identifying mutants that dissociate the conserved interaction between Cia2 and Mms19 , two proteins involved in Iron-Sulfur protein biogenesis and genome stability . This method will facilitate the study of protein structure-function relationships , and elucidate the mechanisms that regulate PPIs .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Reverse PCA, a Systematic Approach for Identifying Genes Important for the Physical Interaction between Protein Pairs
|
The levels of methyl-CpG–binding protein 2 ( MeCP2 ) are critical for normal post-natal development and function of the nervous system . Loss of function of MeCP2 , a transcriptional regulator involved in chromatin remodeling , causes classic Rett syndrome ( RTT ) as well as other related conditions characterized by autism , learning disabilities , or mental retardation . Increased dosage of MeCP2 also leads to clinically similar neurological disorders and mental retardation . To identify molecular mechanisms capable of compensating for altered MeCP2 levels , we generated transgenic Drosophila overexpressing human MeCP2 . We find that MeCP2 associates with chromatin and is phosphorylated at serine 423 in Drosophila , as is found in mammals . MeCP2 overexpression leads to anatomical ( i . e . , disorganized eyes , ectopic wing veins ) and behavioral ( i . e . , motor dysfunction ) abnormalities . We used a candidate gene approach to identify genes that are able to compensate for abnormal phenotypes caused by MeCP2 increased activity . These genetic modifiers include other chromatin remodeling genes ( Additional sex combs , corto , osa , Sex combs on midleg , and trithorax ) , the kinase tricornered , the UBE3A target pebble , and Drosophila homologues of the MeCP2 physical interactors Sin3a , REST , and N-CoR . These findings demonstrate that anatomical and behavioral phenotypes caused by MeCP2 activity can be ameliorated by altering other factors that might be more amenable to manipulation than MeCP2 itself .
Research in the last decade has linked the methyl-CpG-binding protein 2 ( MeCP2 ) with a variety of related neurological disorders [1] . Loss of MeCP2 function causes classic Rett syndrome ( RTT ) , but can also lead to related neurological conditions with symptoms that include autism , mild or severe mental retardation with seizures , or learning disabilities [2] , [3] . Increased dosage of the MECP2 locus also leads to RTT-like features and severe mental retardation [4]–[6] . Similar phenotypes are recapitulated in mice that either lack or overexpress MECP2 , thus underscoring the importance of properly regulating MeCP2 levels [7]–[10] . The MeCP2 protein contains a methyl-CpG-binding domain ( MBD ) and localizes to the heterochromatin where it is believed to regulate gene expression by recruiting histone deacetylases to alter chromatin structure [11] , [12] . While an ortholog for the complete MeCP2 protein does not exist in Drosophila , methyl-CpG-binding domains are conserved from flies to humans [13] . MeCP2 also interacts with other proteins involved in transcriptional repression and chromatin remodeling including Sin3a , REST and Brahma , a core component of the SWI/SNF complex [14]–[16] . These and other previously identified MeCP2 interactors have well conserved orthologs in Drosophila ( Table S1 ) , as do most components of the chromatin remodeling machinery . Examples of evolutionarily conserved chromatin remodeling proteins include members of the Polycomb and trithorax groups ( Pc-G and trx-G , respectively ) , as well as proteins involved in histone tail modification [17] , [18] . Many of these proteins act in protein complexes that function antagonistically to promote either activation or repression of target genes [17] , [19]–[21] . Therefore , we hypothesized that genetic screening in transgenic flies expressing human MeCP2 may permit the identification of genes capable of compensating the phenotypes caused by altered MeCP2 levels . These genetic modifiers may include genes that function antagonistically to MeCP2 in chromatin remodeling , and perhaps other genes modulating MeCP2 functions or interactions . Here we report the identification of such genes .
We generated transgenic flies overexpressing wild-type MeCP2 as well as three mutant alleles using the human MECP2_e2 cDNA ( Figure 1A ) . The RTT R106W allele produces a missense mutation within the MBD that eliminates the protein's ability to bind DNA [22] . The RTT R294X mutation truncates the protein within the transcriptional repression domain ( TRD ) , but maintains the nuclear localization signal . The Δ166 allele completely removes the MBD and N-terminal portion of the protein . Constructs were created inserting each allele into the pUAST vector to utilize the GAL4-UAS system [23] . This system controls expression in specific cell types depending on the Gal4 driver line used , and can be modified by varying the temperature of the fly cultures – increased temperature leads to increased expression . Independent MeCP2 transgenic lines of each allele were generated and tested to ensure that any resulting phenotypes were not caused by the insertion site . Using GMR-Gal4 to drive transgene expression in the eye [24] , we confirmed protein expression by western blot analysis using extracts from whole fly heads ( Figure 1B ) . Furthermore , we found that all three MeCP2 alleles that retain amino acid S423 , which corresponds to murine S421 , produce protein that is specifically phosphorylated at this site ( Figure 1C ) . This specific signal was abolished when the protein extract was treated with alkaline phosphatase ( Figure 1D ) . Phosphorylation at this serine in mammals is brain specific , and it is required by MeCP2 to control dendritic patterning , spine morphogenesis and to regulate the BDNF target gene [25] . Therefore , this key posttranslational modification is conserved when MeCP2 is expressed in Drosophila . Association of MeCP2 with chromatin is a functional property of the mammalian protein [12] , and was evaluated by promoting MeCP2 expression with the ubiquitous Actin5c-Gal4 driver . Immunofluorescent staining of squashed salivary glands demonstrated that MeCP2 localizes to the nucleus and associates with polytene chromosomes along many bands in all four alleles ( Figure 1E-I” ) . While there is widespread association , MeCP2 does not localize to all polytene bands , suggesting target specificity . Association with the polytene chromosomes does not solely depend upon the MBD or C-terminal regions since all four proteins behave similarly . The ability of the R106W and Δ166 mutants to associate with chromosomes implies that the methyl-CpG-binding domain is not required for this activity . Additional factors , possibly functioning in various protein complexes with MeCP2 , may act to recruit MeCP2 to the chromatin . MeCP2 overexpression in the fly eye by GMR-Gal4 was utilized as an assay for rapid genetic screening of modifiers . Overexpression of multiple independent lines of all four alleles resulted in external eye phenotypes of varying degrees ( Figure 2A–E ) . Lines expressing comparable protein levels were selected for each allele ( Figure 1B ) . The full-length wild-type , R106W and Δ166 lines cause a disruption of the external structure of the eye that is recognized as a “glassy” effect on the surface when observed by light microscopy ( Figure 2A–D ) . When evaluated by scanning electron microscopy , these same animals show disorganized ommatidia and partial loss of interommatidial bristles ( Figure 2A'–D' ) . These features were enhanced in flies cultured at a higher temperature ( Figure 2A”–D” ) as a result of elevated expression levels . Of all four alleles , the full-length protein causes the strongest disruption to the external eye . While the R294X allele does not cause an obvious disruption of the external eye structure , it shows a loss of pigmentation phenotype ( Figure 2E , 2E' , and data not shown ) , which had only been seen in one of the most strongly expressing full-length lines . Moreover , expression of the R294X allele at a higher temperature is lethal , possibly a consequence of the leaky expression of the GMR-Gal4 driver into other tissues . We also overexpressed MeCP2 in other fly tissues . Expression of the full-length protein in the wing pouch by C5-Gal4 produces extra vein tissue around the L3 and L5 wing veins ( Figure 2F , G ) . Furthermore , neuronal expression of full-length MeCP2 by the CHA-Gal4 driver [26] leads to impaired motor function in adult flies as measured in a climbing assay ( Figure 2H , Video S1 ) . While external eye phenotypes are most practical for primary screening to identify novel genetic modifiers of MeCP2 , both the wing vein and climbing phenotypes are valuable as secondary screening assays to validate genetic interactions . We rationalized that in vivo genetic modifiers of MeCP2 function might be enriched among known MeCP2 physical interactors . In support of this hypothesis we previously showed that a large proportion of the physical interactors of huntingtin ( the protein that when mutant causes Huntington's disease ) are also genetic modifiers of huntingtin-induced neurodegeneration [27] . To test this hypothesis in the case of MeCP2 , we evaluated Sin3A , Smrter , and crooked legs , the Drosophila homologs of Sin3a , N-CoR , and REST ( Table S1 ) [28] , [29] . We found that heterozygous loss-of-function mutations in each of these three direct interacting partners alter the MeCP2 eye phenotype ( Table 1 , Figure 3 ) . We then tested other candidate modifier genes that were chosen based on their functions . In addition to chromatin remodeling genes , these included a collection of kinases because MeCP2 is phosphorylated [25] , and two genes implicated in Angelman syndrome , a disorder that shares clinical features with Rett syndrome . These last two candidates are the Drosophila homolog of UBE3A , the gene encoding a ubiquitin ligase misregulated in Angelman syndrome , and its target pebble [30] , [31] . When available , both loss-of-function and overexpression mutant Drosophila lines of each candidate were collected . A total of 584 mutant Drosophila lines were obtained and screened against the full-length MeCP2 allele; 392 lines representing 158 individual kinases , 174 lines representing 54 unique chromatin remodeling genes , and 18 lines encompassing UBE3A and pebble mutants . Each mutant line carrying a candidate modifier was crossed to flies expressing the full-length MeCP2 allele from the GMR-Gal4 driver and screened for both enhancers and suppressors . The initial hits in this screen were then re-evaluated with an independent full-length MeCP2 transgenic line . Genes that modify the MeCP2 phenotypes across multiple strains and MeCP2 lines are the chromatin remodeling genes Additional sex combs ( Asx ) , corto , osa , Sex combs on midleg ( Scm ) , and trithorax ( trx ) , the kinase tricornered ( trc ) and the UBE3A target pebble ( pbl ) ( Table 1 ) . Partial loss of function of Asx , corto , osa , pebble , or Scm suppress the eye phenotype induced by full-length MeCP2 , while trc has a similar effect when it is overexpressed ( Figure 4A–H , note improved ommatidial organization relative to MeCP2 control ) . In contrast , enhancement of the eye phenotype was observed in MeCP2 animals with either loss-of-function mutations in trx or overexpression alleles of Scm , osa , and pbl ( Figure 4I-4R , Figure S1 , note greater ommatidial disruption , loss of interommatidial bristles and , in some cases , reduction in eye size and eye depigmentation ) . To exclude the possibility that modifiers of the Gal4-UAS system may simply cause changes in expression of MeCP2 , western blot analysis was performed and demonstrated that the modifiers did not alter the level of MeCP2 protein ( Figure S2 ) . Each modifier line found to alter the full-length MeCP2 phenotype was also investigated in the context of the Δ166 and R294X MeCP2 alleles to determine if the modification was dependent upon a specific MeCP2 domain . For the MeCP2 Δ166 allele , all genetic modifiers behaved similarly to the full-length MeCP2 allele ( Table 1 , Figure S3 ) . Since the MeCP2 R294X allele does not dramatically alter the structure of the eye , suppression was assessed primarily by gain in the amount of eye pigmentation . Enhancement was assessed by increased loss of pigmentation and/or disruption in the external structure of the eye . We found similar phenotype modifications as with full-length MeCP2 with two interesting exceptions ( Table 1 and Figure S4 ) . Partial loss of Sin3A function , which enhances full-length MeCP2 ( compare Figures 3B and 3C ) , suppresses MeCP2 R294X phenotypes ( compare Figures S4D and S4A ) . Partial loss of trx function , which enhances the full-length MeCP2 phenotype ( compare Figures 4B and 4I ) , but , in the case of the trxE2 allele suppresses the R294X phenotype ( compare Figures S4A and S4P ) . The candidate suppressor genes were then further tested against the full-length MeCP2 allele in a second independent assay using the L3 wing vein phenotype ( Figure 5A–E ) . Indeed , alleles of Asx , osa , Scm and trc are able to decrease the penetrance of the L3 wing vein phenotype . Furthermore , loss of function of osa and overexpression of trc improve the climbing phenotype caused by neuronal-specific expression of MeCP2 ( Figure 5F ) . Consistent modification of MeCP2 phenotypes in different tissues , including a behavioral phenotype caused by neural-specific expression , provides additional evidence for the capacity of theses genes to modulate MeCP2 function .
We have used the Drosophila model system to facilitate the identification of genes capable of counterbalancing the consequences of altered levels of the human MeCP2 protein . First , we established anatomical and behavioral assays to assess the effects of expressing human MeCP2 in flies . We used an eye phenotype as a primary assay for the genetic screen , and impaired motor performance and other phenotypes as secondary assays for validating purposes . The eye phenotype has been used successfully in a variety of genetic screens including screens for enhancer/suppressors of other neurological disease models . Although expression of a variety of “toxic” human proteins leads to apparently similar “rough” eye phenotypes , their specificity is demonstrated when comparing the genetic modifiers uncovered in the screens . For example , there is little or no overlap between the MeCP2 modifiers reported here and modifiers of the eye phenotype produced by expression of ataxin-1 [32] , [33] or huntingtin [27] . In contrast , we found that the majority of the modifier genes modulating the eye phenotype caused by wild-type MeCP2 similarly modulate the phenotypes caused by the R294X and Δ166 MeCP2 mutations . Two exceptions are Sin3A and trx , which have opposite effects on wild-type and R294X MeCP2 ( Table 1 , Figures 3B , C versus Figures S4A , D , and Figures 4B , I versus Figures S4A , P ) . MeCP2 associates with a co-repressor complex containing Sin3A through the TRD domain [14] , which is partially deleted in the truncated R294X protein . This mutant also lacks the MeCP2 C-terminal region that is important for interactions with chromatin in vitro [34] . The TRD domain and/or C-terminal region may thus be involved in the observed genetic interaction between MeCP2 and trx . It is important to note that both Sin3A and trx do modify the eye phenotype of R294X MeCP2 animals , albeit in the opposite way from the wild-type MeCP2 . Thus , the TRD/C-terminal domains may play a modulating role rather than being required for the interaction . A commonly accepted model of MeCP2 function postulates that MeCP2 binds to methylated CpG islands in promoters where it recruits histone deacetylases and other co-repressors to silence gene transcription [14] , [35] . However , accumulating evidence suggests that this may be too simple a view of MeCP2 function . For example , MeCP2 binds to unmethylated DNA with affinity only 3 times weaker than to methylated DNA [36] , and MeCP2 also binds [37] or requires AT sequences for binding [38] . Moreover , MeCP2 interacts with both methylated and unmethylated chromatin and leads to alterations in the secondary structure of both types of chromatin [34] , [39] . In addition , large-scale mapping of MeCP2 binding sites in chromosomal regions containing candidate MeCP2 target genes revealed that: 1 ) MeCP2 is absent from highly methylated promoters , 2 ) only ∼6% of MeCP2 binding sites are in CpG islands , and 3 ) many MeCP2-bound promoters are actively expressed [40] . Furthermore , a recent study of gene expression patterns in mice that either lack or overexpress MeCP2 suggests that many genes are activated by MeCP2 [41] . Here we show that the methyl-CpG-binding domain is not necessary for association of the MeCP2 protein with chromatin in polytene chromosomes ( Figures 1H-H” ) , nor is it required to produce an eye phenotype in Drosophila ( Figures 2D-D” ) . In this context it is interesting to note that unlike mammals , bacteria , plants , and other insects , the levels of DNA methylation are very low in Drosophila [42] . Together these data suggest that MeCP2 function may be more complex than previously thought . MeCP2 may regulate both methylated and unmethylated target genes in vivo , possibly as part of large protein complex ( es ) of chromatin remodelling proteins regulating gene expression both positively and negatively . Using a candidate gene approach , we provide proof of principle that modulating the activity of modifier genes can amend MeCP2 function in vivo . Among this group of genes is the kinase trc , a member of the NDR ( nuclear Dbf-related ) family . We could not detect alterations in the phosphorylation of MeCP2 in trc mutants ( data not shown ) . However , there is evidence that both trc and one of its mammalian homologs , NDR2 , are involved in dendritic formation [43] , [44] , a feature also found to be affected by mutations in MeCP2 . Also , modification of the MeCP2 phenotype by the E3 ligase UBE3A target pbl [31] is noteworthy due to the similarities between Rett and Angelman syndromes . Patients with Angelman-like features have been identified with MeCP2 mutations [45] , [46] and , while still controversial , some studies have demonstrated a decrease of UBE3A in Rett patients and Mecp2 null mice [47]–[49] . The data presented here suggest that shared pathways may be involved in Rett and Angelman syndromes . Misregulation of neuronal genes caused by alterations in MeCP2 activity is thought to cause Rett and Rett-like syndromes [50] , [51] . One possible avenue for therapy is to identify the MeCP2 target genes misregulated during disease and to restore their normal regulation . This approach may prove impractical if the targets are numerous or difficult to identify due to subtle variations in expression levels in response to MeCP2 activity [52]–[57] . A possible future treatment based on gene therapy to restore normal levels of MeCP2 also seems improbable . The nervous systems of Rett patients are mosaic due to random X-chromosome inactivation causing some neurons expressing the normal while others expressing the mutant allele . Therefore , in the context of neurons expressing the wild-type allele , gene therapy is not possible because doubling of MeCP2 also leads to disease [5] , [6] , [58] . An alternative approach is to identify molecular mechanisms capable of compensating for the misregulation of target genes caused by MeCP2 altered levels . This study provides support for the validity of this approach . We identified specific chromatin remodeling genes of the Pc-G and Trx-G ( i . e . , Asx , corto , osa , and Scm ) that suppress the phenotypes caused by MeCP2 overexpression in Drosophila . Interestingly , both in Drosophila and mammals , mutations in genes of either Pc-G or Trx-G also suppress the body patterning abnormalities caused by mutations in members of the other group [17] , [19] . In conclusion , human MeCP2 protein expressed in Drosophila maintains important features observed in mammals such as phosphorylation and association with the chromatin . The novel modifiers identified in this model system point to potential therapeutic targets that might be more amenable to manipulation than MeCP2 , and thus they provide new opportunities to develop therapies for Rett syndrome and related neurological disorders .
Each of the MeCP2 alleles described was cloned into the pUAST vector in order to utilizing the GAL4-UAS system ( Figure 1A ) . The full-length human cDNA of the MECP2_e2 isoform ( 1461 nucleotides , 486 amino acids ) was subcloned into the EcoRI site of the pUAST vector . The remaining three alleles were generated by PCR mutagenesis of this initial construct . MeCP2 R294X was amplified with primers that attached a stop codon and Kpn I site to the C-terminal end . This PCR fragment was digested with EcoRI and Kpn I and then ligated between these restriction sites in pUAST . Primers amplifying the MeCP2 Δ166 fragment added an EcoRI site , a conserved Drosophila consensus sequence ( TCGAC ) , and an ATG start site to the N-terminal side of the protein . Transgenic Drosophila lines were generated by injection of these constructs in embryos following standard methods . We generated eleven MeCP2 full-length lines , ten MeCP2 R106W lines , three MeCP2 R294X lines and ten MeCP2 Δ166 lines . Additional Drosophila lines were obtained from the Bloomington Drosophila Stock Center , the Harvard Medical School Exelixis Drosophila Stock Collection , and private investigators ( see acknowledgements ) . Protein was collected from Drosophila heads in a solution of 5% β-mercaptoethanol in Laemmli Sample Buffer ( Bio-Rad ) . For the alkaline phosphatase treatment , Drosophila heads were collected in protein extraction buffer ( PBS with 0 . 1% Nonident P40 and protease inhibitors ) , samples were smashed and kept on ice for 1 hour , vortexing each 10 minutes in order to facilitate protein extraction . Samples were then mixed 1∶1 with the calf intestinal alkaline phosphatase ( CIP ) treatment , 10% CIP enzyme ( New England Biolabs ) , 30% CIP buffer in water , and incubated for 50 minutes at 37°C . Laemmli Sample Buffer was then added to these reactions . Proteins were run on SDS-PAGE gels with eight Drosophila heads per lane ( except for the alkaline phosphatase experiment which had ten heads per lane ) . Proteins were then transferred to nitrocellulose membrane ( Optitran ) using 10mM CAPS with 10% methanol . Membranes were blocked in BLOTTO 5% Non-Fat Dry Milk ( Bio-Rad ) in TBS-T ( 100mM Tris-Cl pH 7 . 5 , 150mM NaCl , 0 . 1% Tween 20 ) . The following antibodies were used diluted in BLOTTO: anti-MeCP2 antibodies ( 1∶1000 , Upstate , #07-013 , and Affinity , #PA1-887 ) , anti-lamin C ( 1∶1000 , Developmental Studies Hybridoma Bank , #LC28 . 26 ) , and anti-tubulin ( 1∶5000 , Developmental Studies Hybridoma Bank , #E7 ) . Anti-phosphorylated MeCP2 S423 was diluted in 5% BSA ( 1∶1000 ) [25] . Anti-rabbit and anti-mouse horseradish peroxidase-conjugated secondary antibodies ( Bio-Rad ) were diluted 1∶5000 in BLOTTO and membranes were developed using chemiluminescence with either the ECL kit ( Amersham Biosciences ) or the SuperSignal West Dura kit ( Pierce ) . Quantification of western blots was performed on a densitometer ( Molecular Dynamics ) using the ImagQuant program . Experimental and control lines were crossed to flies with the eye specific GMR-Gal4 driver . Offspring were sorted by genotype and whole adult flies were sequentially dehydrated in ethanol , critically-point dried and placed on aluminum mounts . Samples were coated with a platinum alloy for a thickness of 50 nm and flash carbon coated . Drosophila heads were then analyzed with a JEOL JSM-5900 scanning electron microscope . Experimental and control lines were crossed to flies with the C5-Gal4 driver and cultured at 25°C . Once the offspring had eclosed , flies were sorted by genotype and each individual wing was scored under a light microscope for extra vein tissue near or attached to the L3 wing vein . Wings were removed from flies and mounted in DPX Mounting Medium ( Electron Microscopy Sciences ) . Experimental and control lines were crossed to flies with the CHA-Gal4 driver and cultured at 25°C . Virgins were collected of each genotype and sorted into batches of 20 flies . Flies were enclosed inside two clean , unused 9 . 25 cm culturing tubes that had been taped together , for a total height of 18 . 5 cm . Flies were tapped down to the bottom of the vial and permitted 18 seconds to climb within both tubes to the top . At the end of 18 seconds , flies were scored as to whether their final position was either above or below 7 cm . Each group was trained in this procedure for 10 trials and then tested for 10 trials . Trials were performed between 3–6 pm . Experimental and control lines were crossed to flies with the ubiquitous Actin5c-Gal4 driver and cultured at 25°C . Salivary glands were dissected from third instar larvae , fixed with formaldehyde , and squashed according to standard protocols . Samples were blocked with PBT with 0 . 2% BSA and 5% horse serum to reduce background . Primary antibodies for MeCP2 were used ( 1∶100 , Affinity and 1∶200 , Upstate ) . The secondary immunofluorescence goat anti-rabbit Cy3 antibody was used at a 1∶200 dilution . The slides were also treated with an RNase cocktail ( 1∶1000 , Ambion ) and then TOTO-3 ( 1∶2000 , Molecular Probes ) to stain the DNA for confocal microscopy . Slides were then mounted with a drop of Vectashield containing DAPI in order to visualize the DNA by eye . Images were collected by confocal microscopy using the AxioVision and ImageJ programs .
|
Rett syndrome ( RTT ) is a progressive neurodevelopmental disorder that affects girls early in childhood and is caused by mutations in the MECP2 gene . Loss of MeCP2 function can also lead to clinically distinct conditions characterized by autism , learning disability , and mental retardation . Remarkably , increased levels of MeCP2 leads to related neurological disorders and mental retardation as well . These data emphasize the critical importance of regulating MeCP2 protein levels for normal post-natal development and function of the nervous system . MeCP2 is a protein that associates with chromatin and is thought to modulate gene expression . We have generated Drosophila that overexpress human MeCP2 to investigate the possibility that adjusting the activity of other genes may compensate for altered levels of MeCP2 . In support of this hypothesis , we found a variety of modifier genes , including chromatin remodeling genes , that are able to ameliorate and/or aggravate the consequences of MeCP2 overexpression . These findings open the possibility of therapeutic avenues for RTT and related neuropsychiatric disorders by targeting proteins that are possibly easier to manipulate than MeCP2 itself .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"neurological",
"disorders"
] |
2008
|
Genetic Modifiers of MeCP2 Function in Drosophila
|
Accurate and reliable predictions of infectious disease dynamics can be valuable to public health organizations that plan interventions to decrease or prevent disease transmission . A great variety of models have been developed for this task , using different model structures , covariates , and targets for prediction . Experience has shown that the performance of these models varies; some tend to do better or worse in different seasons or at different points within a season . Ensemble methods combine multiple models to obtain a single prediction that leverages the strengths of each model . We considered a range of ensemble methods that each form a predictive density for a target of interest as a weighted sum of the predictive densities from component models . In the simplest case , equal weight is assigned to each component model; in the most complex case , the weights vary with the region , prediction target , week of the season when the predictions are made , a measure of component model uncertainty , and recent observations of disease incidence . We applied these methods to predict measures of influenza season timing and severity in the United States , both at the national and regional levels , using three component models . We trained the models on retrospective predictions from 14 seasons ( 1997/1998–2010/2011 ) and evaluated each model’s prospective , out-of-sample performance in the five subsequent influenza seasons . In this test phase , the ensemble methods showed average performance that was similar to the best of the component models , but offered more consistent performance across seasons than the component models . Ensemble methods offer the potential to deliver more reliable predictions to public health decision makers .
The practice of combining predictions from different models has been used for decades by climatologists and geophysical scientists . These methods have subsequently been adapted and extended by statisticians and computer scientists in diverse areas of scientific inquiry . In recent years , these “ensemble” forecasting approaches frequently have been among the top methods used in prediction challenges across a wide range of applications . Ensembles are a natural choice for noisy , complex , and interdependent systems that evolve over time . In these settings , no one model is likely to be able to capture and predict the full set of complex relationships that drive future observations from a particular system of interest . Instead “specialist” or “component” models can be relied on to capture distinct features or signals from a system and , when combined , represent a nearly complete range of possible outcomes . In this work , we develop and compare a collection of ensemble methods for combining predictive densities . This enables us to quantify the improvement in predictions achieved by using ensemble methods with varying levels of complexity . To illustrate these ensemble methods , we present time-series forecasts for infectious disease , specifically for influenza in the United States . The international significance of emerging epidemic threats in recent decades has highlighted the importance of understanding and being able to predict infectious disease dynamics . With the revolution in science driven by the promise of “big” and real-time data , there is an increased focus on and hope for using statistics to inform public health policy and decision-making in ways that could mitigate the impact of future outbreaks . Some of the largest public health agencies in the world , including the US Centers for Disease Control and Prevention ( CDC ) have openly endorsed using models to inform decision making , saying “with models , decision-makers can look to the future with confidence in their ability to respond to outbreaks and public health emergencies” [1] . There is a large literature on prediction methods for influenza . We will give a brief overview of this literature here , and refer the reader to Chretien et al . [2] and Nsoesie et al . [3] for more comprehensive reviews; additionally , [4] present the results of a recent influenza prediction challenge run by the CDC where many of these models were employed . Infectious disease prediction methods can be broadly grouped into three categories: agent-based models , compartmental models [5–10] , and regression-based time series models that may include auto-regressive and seasonal terms [11] . Additionally , these models may use a variety of different data sources and covariates to inform their predictions , including historical values of the disease incidence time series [5–11]; data derived from internet sources such as web searches , wikipedia page views , and twitter [5 , 6 , 8 , 11–14]; and climatological variables [5 , 6 , 8 , 10 , 11] , among others . These models may generate either point predictions , possibly along with associated predictive intervals , or full predictive distributions . The ensemble methods that we explore in the present work are designed to combine predictions from multiple models , which could use a variety of different model structures and covariates to generate predictions . Development of the methods presented in this manuscript was motivated by the observation that certain prediction models for infectious disease consistently performed better than other models at certain times of year . We observed in previous research that early in the influenza season , simple models of historical incidence often outperformed more standard time-series prediction models such as a seasonal auto-regressive integrated moving average ( SARIMA ) model [15] . However , in the middle of the season , the time-series models showed improved accuracy . We set out to determine whether ensemble methods could use this information about past model performance to improve predictions . A large number of ensemble methods have been developed for a diverse array of tasks including regression , classification , and density estimation . These methods are broadly similar in that they combine results from multiple component models . However , details differ between ensemble methods . We suggest Polikar [16] for a review of ensemble methods; many of these are also discussed in detail in Hastie et al . [17] . While there are many different methods for combining models , all ensemble models discussed in this paper use an approach called stacking [18] . In this approach , each of the component models is trained separately in a first stage , and cross-validated measures of performance of those component models are obtained . Then , in a second stage , a stacking model is trained using the cross-validated performance measures to learn how to optimally combine predictive densities from the component models . The specific implementations of stacking that we use obtain the final predictive density as a weighted sum of the component predictive densities , where the weights may depend on covariates . We refer to this approach generally as a “weighted density ensemble” approach to prediction . Several variations on this strategy have been explored in the literature previously [19–21] . However , other ensemble methods for density estimation have also been developed . For example , Rosset and Segal [22] develop a boosting method in which the component models are estimated sequentially , with results from earlier models affecting estimation of later models . In structured prediction settings such as time series forecasting , ensemble methods may benefit from taking advantage of the data structure . For example , it may be the case that different models offer a better representation of the data at different points in time . A common idea in these settings is to use model weights that change over time . For instance , model weights may vary as a function of how well each model did in recent predictions [23] or by using a more formal graphical structure such as a hidden Markov model to track which component model is most likely to have generated new observations as they arise over time [24 , 25] . It is also possible to combine the component models with weights that depend on observed covariates or features [26] . For example , in an ensemble for a user recommendation system , Jahrer et al . [27] allowed model weights to depend on a variety of features including the time that a user submitted a rating . Using component models that generate predictive densities for outcomes of interest , we have implemented a series of ensembles using different methods for choosing the weights for each model . Specifically , we compare three different approaches . The first approach simply takes an equally weighted average of all models . The second approach estimates constant but not necessarily equal weights for each model . The third approach is a novel method for determining model weights based on features of the system at the time predictions are made . The overarching goal of this study is to create a systematic comparison between ensemble methods to study the benefits of increasing complexity in ensemble weighting schemes . We are aware of two previous articles that developed ensemble methods for infectious disease prediction . Yamana et al . [28] and Chakraborty et al . [11] both developed model stacking frameworks that are similar to the second approach outlined above using a constant weight for each component model . The present article is differentiated from this previous work in that we explore and compare a range of more flexible ensemble methods where the weights depend on observed features . This paper presents a novel ensemble method that determines optimal model combinations based on ( a ) observed data at the time predictions are made and ( b ) aspects of the predictive distributions obtained from the component models . We refer to models built using this approach as “feature-weighted” ensembles . This approach fuses aspects of different ensemble methods: it uses model stacking [18] and estimates model weights based on features of the system [26] using gradient tree boosting [29] . Using seasonal influenza outbreaks in the US health regions as a case-study , we developed and applied our ensemble models to predict several attributes of the influenza season at each week during the season . By illustrating the utility of these approaches to ensemble forecasting in a setting with complex population dynamics , this work highlights the importance of continued innovation in ensemble methodology .
We obtained publicly available data on seasonal influenza activity in the United States between 1997 and 2016 from the US Centers for Disease Control and Prevention ( CDC ) ( Fig 1 ) . For each of the 10 Health and Human Services regions in the country in addition to the nation as a whole , the CDC calculates and publishes each week a measure called the weighted influenza-like illness ( wILI ) index . The wILI for a particular region is calculated as the average proportion of doctor visits with influenza-like illness for each state in the region , weighted by state population . During the CDC-defined influenza season ( between Morbidity and Mortality Weekly Report week 40 of one year and 20 of the next year ) , the CDC publishes updated influenza data on a weekly basis . This includes “current” wILI data from two weeks prior to the reporting date , as well as updates to previously reported numbers as new data becomes available . For this analysis , we use only the final reported wILI measures to train and predict from our models . In the early seasons , data were not recorded during the off-season . Additionally , there were 52 observations in which the reported wILI was zero; these generally occurred near the off-season in early years , and occurred in weeks when only small numbers of health care providers submitted reports to the CDC . We treated these reported zeros as missing data throughout the analysis . The CDC defines the influenza season onset as the first of three consecutive weeks of the season for which wILI is greater than or equal to a threshold that is specific to the region and season . This threshold is the mean percent of patient visits where the patient had ILI during low incidence weeks for that region in the past three seasons , plus two standard deviations [30] . The CDC provides historical threshold values for each region going back to the 2007/2008 season [31] . Additionally , we define two other metrics specific to a region-season . The peak incidence is the maximum observed wILI measured in a season . The peak week is the week at which the maximum wILI for the season is observed . Each predictive distribution was represented by probabilities assigned to bins associated with different possible outcomes . For onset week , the bins are represented by integer values for each possible season week plus a bin for “no onset” . For peak week , the bins are represented by integer values for each possible season week . For peak incidence , the bins capture incidence rounded to a single decimal place , with a single bin to capture all incidence over 12 . 95 . Formally , the incidence bins are as follows: [0 , 0 . 05 ) , [0 . 05 , 0 . 15 ) , … , [12 . 85 , 12 . 95 ) , [12 . 95 , 100] . These bins were used in the 2016-2017 influenza prediction contest run by the CDC [32] . We measure the accuracy of predictive distributions using the log score . The log score is a proper scoring rule [33] , calculated in our setting as the natural log of the probability assigned to the bin containing the true observation . Proper scoring rules are preferred for measuring the quality of predictive distributions because the expected score is optimized by the true probabilty distribution . We note that for peak week , in some region-seasons the same peak incidence was achieved in multiple weeks ( after rounding to one decimal place ) . In those cases , we calculated the log score as the log of the sum of the probabilities assigned to those weeks; this is consistent with scoring procedures used in the 2016-2017 flu prediction contest run by the CDC [32] . However , the log score is not directly comparable with the score used by the CDC in the prediction contest . The CDC calculates the score of a prediction as the log of the combined probability assigned to several bins surrounding the realized outcome; this has some benefits , but has the disadvantage that it is not a proper score . We have opted to use the log score in this work because it is a proper score . We used three component models to generate probabilistic predictions of the three prediction targets . The first model was a seasonal average model that utilized kernel density estimation ( KDE ) to estimate a predictive distribution for each target . The second model utilized kernel conditional density estimation ( KCDE ) and copulas to create a joint predictive distribution for incidence in all remaining weeks of the season , conditional on recent observations of incidence [15] . By calculating appropriate integrals of this joint distribution , we constructed predictive distributions for each of the seasonal targets . The third model used a standard seasonal auto-regressive integrated moving average ( SARIMA ) implementation . All models were fit independently on data within each region . All of the ensemble models we consider in this article work by averaging predictions from the component models to obtain the ensemble prediction . Additionally , these methods are stacked model ensembles because they use leave-one-season-out predictions from the independently estimated component models as inputs to estimate the model weights [18] . We begin our discussion of ensemble methods with a general overview , introducing a common set of notation and giving a broad outline of the ensemble models we will use in this article . We then describe our proposed weighted density ensemble model specifications in more detail . In this section we introduce the particular specification of the parameter weight functions πm ( xt ) that we use for the FW-wu , FW-reg-w , FW-reg-wu , and FW-reg-wui models and discuss estimation . In order to ensure that the the πm are non-negative and sum to 1 for all values of xt , we parameterize them in terms of the softmax transformation of real-valued latent functions ρm: π m ( x t ) = exp { ρ m ( x t ) } ∑ m ′ = 1 M exp { ρ m ′ ( x t ) } . ( 2 ) For a pair of models l , m ∈ {1 , … , M} , ρl ( xt ) > ρm ( xt ) indicates that model l has more weight than model m for predictions at the given value of xt . The functions ρm ( xt ) could be parameterized and estimated using many different techniques , such as a linear specification in the features , splines , or so on . We chose to estimate the functions ρm ( x ) using gradient tree boosting . Gradient tree boosting uses a forward stagewise additive modeling algorithm to iteratively and incrementally construct a series of regression trees that , when added together , create a function designed to minimize a given loss function . In our application , the algorithm builds up the ρm ( xt ) that minimize the negative log-score of the stacked predictions f ( yt|xt ) across all times t: L {ρ ( x t ) } = - ∑ t log { f ( y t | x t ) } = - ∑ t log [ ∑ m = 1 M exp { ρ m ( x t ) } ∑ m ′ = 1 M exp { ρ m ′ ( x t ) } f m ( y t | x t ( m ) ) ] , ( 3 ) where f m ( y t | x t ( m ) ) is the cross-validated predictive density from the mth model evaluated at the realized outcome yt . Specifically , we define a single tree as T ( x t ;θ ) = ∑ j = 1 J γ j I R j ( ψ ) ( x t ) , ( 4 ) where the Rj ( ψ ) are a set of disjoint regions that comprise a partition of the space X of feature values xt , and I is the indicator function taking the value 1 if xt ∈ Rj ( ψ ) and 0 otherwise . The parameters θ = ( ψ , γ ) for the tree are the split points ψ partitioning X into the regions Rj ( ψ ) and the regression constants γ associated with each region . The function ρm ( xt ) is obtained as the sum of B trees: ρ m ( x t ; Θ m ) = ∑ b = 1 B T ( x t ; θ m , b ) . ( 5 ) In each iteration b of the boosting process , we estimate M new regression trees , one for each component model . These trees are estimated so as to minimize a local approximation to the loss function around the weight functions that were obtained after the previous boosting iteration . Our approach builds on the xgb . train function in the xgboost package for R to perform this estimation [39] . The functionality in that package assumes that the loss function is convex , and optimizes a quadratic approximation to the loss in each boosting iteration . The loss function in Eq ( 3 ) is not guaranteed to be convex , so a direct application of this optimization method fails in our setting . We have modified the implementation in the xgboost package to use a gradient descent step in cases where the loss is locally nonconvex ( concave or indeterminate ) . Gradient tree boosting is appealing as a method for estimating the functions ρm because it offers a great deal of flexibility in how the weights can vary as a function of the features xt . On the other hand , this flexibility can lead to overfitting the training data . In order to limit the chances of overfitting , we have explored the use of three regularization parameters: We selected values for these regularization parameters using a grid search optimizing leave-one-season-out cross-validated model performance . We used R version 3 . 2 . 2 ( 2015-08-14 ) for all analyses [35] . All data and code used for this analysis is freely available in an R package online at https://github . com/reichlab/adaptively-weighted-ensemble and may be installed in R directly . Predictions generated in real-time with early development versions of this model during the 2016/2017 influenza season may be viewed at https://reichlab . io/flusight/ . To maximize reproducibility of our work , we have set seeds prior to running code that relies on stochastic simulations using the rstream package [40] . Additionally , the manuscript itself was dynamically generated using RMarkdown .
Fig 3 displays variation in leave-one-season-out log scores from the three component models over the course of the training phase seasons , along with the corresponding model weight estimates from the CW and FW-reg-w models . Performance of the SARIMA and KCDE models is similar , with mean log scores from those models starting out near or slightly below the mean performance of KDE , but with performance improving as more data become available . Near the beginning of some seasons , predictions from the SARIMA model are quite a bit worse than predictions from the other two component models . S4 Fig illustrates that these patterns are consistent across the other regions . S5 Fig shows that performance of the component models also varies with the model’s uncertainty as measured by the number of bins required to cover 90% in the predictive distribution , and S6 Fig shows that performance varies with the observed wILI in the week when predictions are made . The model weights assigned by the feature weighted ensemble models generally track these trends in relative model performance ( Fig 3 , S7 Fig ) . For all three targets , at the national level the weight assigned to the SARIMA model increases and the weight assigned to KDE decreases as the season progresses . However , the magnitude of shifts in model weights as the weighting features vary is different for the three prediction targets . Aggregating across all combinations of prediction target , region , and season in the test phase , the best component models and the best ensemble models had similar performance ( Fig 4 ) . The CW ensemble had the highest average log scores across all three prediction targets , but a permutation test ( described in S1 Text ) was unable to distinguish its performance from the KCDE , SARIMA , or FW-reg-w models . However , these four methods all outperformed the KDE model in terms of mean log scores by a wide margin , as well as the EW and FW-wu , FW-reg-wu , and FW-reg-wui ensembles by narrower margins . These general trends in model performance were similar for each of the three prediction targets individually; for example , S8 Fig demonstrates that average performance of the FW-reg-w and SARIMA models is similar for all three prediction targets . As noted above , our test set included only 5 seasons , and the effective sample size for model comparison is smaller than the 165 combinations of prediction target , region , and test phase season due to correlations in predictive performance across regions and seasons . This may have contributed to our inability to detect statistically significant differences between the best models , and may limit the generalizability of these results; we will return to this point in the discussion . Although the aggregate performance of these models is quite similar , some differences between the methods begin to emerge when we examine performance in more detail . Predictions that are used in setting public policy must be of consistent quality across all regions and seasons . We observed that the component models showed more variability and lower worst-case performance than the ensemble methods . The discussion in this subsection presents results of an exploratory analysis of the results , and all p-values are from post-hoc hypothesis tests . To examine consistency of predictive performance , for each combination of prediction target , region , and test phase season we calculated the difference in mean log scores between each method and the method with median performance for that target , region , and season . This measure of model performance relative to the median can be compared across prediction targets , regions , and seasons that may be predicted with varying levels of difficulty . Fig 5 displays these differences in performance relative to the median for just the KCDE , SARIMA , CW , and FW-reg-w methods . This comparison demonstrates that while these methods all had similar average performance , the CW and FW-reg-w ensemble methods had more consistent performance than the component models did , as is observed by the heavier distributional tails below zero on the horizontal axis . We can quantify this observation by comparing the minimum performance relative to the median across all prediction targets , regions , and seasons for each method ( Fig 6 ) . This comparison reveals that the FW-reg-w ensemble had better worst-case performance than all of the component models , and the CW ensemble had better worst-case performance than the KDE and KCDE component models . These differences were both statistically and practically significant . The differences between the ensemble and component models become more marked if we use the 10th percentile of performance differences relative to the median as a more stable measure of the lower tail of this distribution than the minimum ( S9 Fig ) . Additionally , the FW-reg-w model had a higher 10th percentile difference in performance from the median model than all other methods . Across all three prediction targets and all test phase seasons , the FW-reg-w ensemble had the most consistent performance of all methods we considered ( S10 Fig ) . The regularization of feature-weighted ensembles improved early-season prediction accuracy . A comparison of the FW-wu and FW-reg-wu models shows improvements in both mean performance and worst-case performance when regularization was used to create smoother functions of model weights as a function of season week and model uncertainty ( Figs 4 and 6 ) .
In this work we have examined the potential for ensemble methods to improve infectious disease predictions . We explored a nested series of ensemble methods , focusing on methods that computed weighted averages of predictive distributions for seasonal targets of public health interest , such as the peak intensity of the outbreak and the timing of both season onset and peak . The methods we examined ranged from using equal model weights to more complex schemes with weights that varied as functions of multiple covariates . The best of these ensemble methods achieved overall performance that was about as good as the best of the individual component models , with increased stability in model performance across different regions and seasons . Increased stability in predictive accuracy can provide decision makers with more confidence when using predictions as inputs to set policy . For example , if a single model does well in most seasons but occasionally fails badly , planning decisions may be negatively impacted in those failing years . This may be particularly important in a public health setting where the events that are most important to get right are those relatively rare cases when incidence is much larger than usual or the season timing is earlier or later than usual . This reduction in variability of model performance achieved by ensemble methods is therefore important for ensuring that our predictions are reliable under a variety of conditions . Among the different ensemble specifications we considered , the CW and FW-reg-w models had slightly better average performance during the test phase than the three other ensemble methods that included some form of regularization on the model weighting functions , and much better performance than an ensemble with unregularized weighting function . The FW-reg-wui and FW-reg-wu ensembles did not outperform the simpler FW-reg-w ensemble , indicating that including model uncertainty and recent observations of disease incidence did not add much more information about relative model performance than was available from the week of the season in which predictions were generated . Analysis of worst-case performance suggests that the FW-reg-w ensemble had more stable performance across different regions and seasons than the other ensemble specifications . However , whether or not this difference was statistically significant depended on the measure of worst-case performance used . Overall , the FW-reg-w method had good average and worst-case performance across all test phase seasons and prediction targets; the CW ensemble had similar average performance , but its worst-case performance was not as good as that of the FW-reg-w method . All hypothesis tests we conducted related to worst-case performance were post-hoc tests conducted after an exploratory analysis of relative model performance , and these results should be confirmed in future studies . Additionally , the permutation test we used accounts for serial autocorrelation in model performance within a region-season , but does not account for correlation across region or seasons; thus the p-values discussed throughout this work should be regarded as only approximate indicators of statistical significance . The feature-weighted ensemble models presented in this article use a novel scheme to estimate feature-dependent model weights that sum to 1 and are therefore suitable for use in combining predictive distributions . This general method could be applied to combine distribution estimates in any context , and is not limited to time-series or infectious disease applications . Furthermore , comparing an implementation of the feature-weighting that smoothed the model weights to one that did not showed consistent improvements in model performance . This result suggests that future work on feature-weighted ensemble implementations should consider regularized estimation . Infectious disease predictions are only useful to public health officials if they are communicated effectively in real time . Predictions from an early version of the FW-reg-w model were updated weekly during the 2016/2017 influenza season and disseminated through an interactive website at https://reichlab . io/flusight/ . While we have successfully deployed the methods discussed here in a real-time setting , in this article we have ignored the important issue of reporting delays that occur with real-time data . All models were trained using the finalized value of the incidence measure , and these finalized values were used to make the cross-validated predictions that were inputs to the ensemble estimation as well as the predictions for the test set evaluation . Some component models may be more or less sensitive to reporting delays than other models , and this could lead to inappropriate estimates for the ensemble weighting functions if finalized data were used for the cross-validated predictions but the methods were then used in real time . Ideally , the cross-validated model log scores used to estimate the ensemble weighting functions should be obtained using the same sort of “non-finalized” data that the models will encounter when making real-time predictions . A central challenge of working with infectious disease data sets is the limited number of years of data available for model estimation and evaluation . We have used approximately one fourth of our data set for model evaluation , which left us with only 14 seasons of training data and 5 seasons of testing data . Additionally , we had fewer than 14 seasons of leave-one-season-out predictions to use in estimating the model weighting functions for the FW-wu ensemble methods because the SARIMA model required unobserved seasonally lagged incidence to make predictions for the first few seasons in the training phase . This small sample size may have negatively impacted our ability to estimate the weighting functions . Altogether the test phase included 55 combinations of region and season , with a total of 2469 predictions from each method made across all three prediction targets before the test phase season onset or peak occurred . Nevertheless , because of the high degree of correlation in model log scores for the same prediction target in different weeks and regions within the same season we have a smaller effective sample size for detecting differences in average model performance in the test phase . The findings in this work should be confirmed with additional data sets . Another possible avenue would be to obtain pseudo out-of-sample results by performing cross validation within the training phase . Another limitation of this work is the small selection of component models used . Theoretical results and applications have demonstrated that ensemble methods are most effective when using a diverse set of component models [16] . In our study , the KCDE and SARIMA component models are similar in that they both use seasonal terms and observations of recent incidence to inform their predictions ( though we note that these two models tended to perform well in different seasons , as illustrated in S10 Fig ) . Increased component model diversity could yield improved ensemble performance; this could be achieved either through inclusion of different model structures ( for example , agent-based or mechanistic models such as those explored in [5–10] ) or different covariates ( such as information about the circulating strains of a disease , spatial effects , weather , or social media data , as used by [5 , 6 , 8 , 10–14] ) . Thus , the current work should not be viewed as a competitor to the models developed in previous work , but rather as a method for integrating and unifying the diverse array of methods that have been developed in the literature . The methods presented here are suitable for combining predictions from any collection of component models that each output a full predictive distribution , regardless of model structure . Our exploration of feature-weighted ensembles is also limited by the relatively restricted feature sets we used for the weighting functions . We selected a few features based on exploratory analysis of the training phase results , and set all ensemble model formulations before obtaining any predictions for the test phase . It is possible that other weighting features not considered in this work may be more informative than those we have used . Some ideas for weighting covariates to use in future work include the largest incidence so far this season; the onset threshold; alternative summaries of the predictive distributions from the component models such as the probability at the mode or the modal value; the predominant flu strain; or the distribution of incidence in age groups . The performance of the ensemble methods might be improved by subsetting the training data for the ensembles to the most important observations . The discrepancy in this work between the times used to train the ensembles ( all leave-one-season-out predictions ) and the times used for model comparison ( only predictions made before the season onset or peak ) may have led to an artificial decline in performance for the ensembles; this may be especially so for the relatively inflexible CW method . This work provides a rigorous and comprehensive evaluation of ensemble methods for averaging probabilistic predictions for features of infectious disease outbreaks . A range of models , both single component models and ensemble models that combined component model predictions , demonstrated the ability to make more accurate predictions than a seasonal average baseline model . Additionally , systematic comparisons of simple and complex prediction models highlight a crucial added value of ensemble modeling , namely increased stability and consistency of model performance relative to the component models . Continued investigation , application , and innovation is necessary to strengthen our understanding of how to best leverage combinations of models to assist decision makers in fields , such as public health and infectious disease surveillance , that require data-driven rapid response .
|
Public health agencies such as the US Centers for Disease Control and Prevention would like to have as much information as possible when planning interventions intended to reduce and prevent the spread of infectious disease . For instance , accurate and reliable predictions of the timing and severity of the influenza season could help with planning how many influenza vaccine doses to produce and by what date they will be needed . Many different mathematical and statistical models have been proposed to model influenza and other infectious diseases , and these models have different strengths and weaknesses . In particular , one or another of these model specifications is often better than the others in different seasons , at different times within the season , and for different prediction targets ( such as different measures of the timing or severity of the influenza season ) . In this article , we explore ensemble methods that combine predictions from multiple “component” models . We find that these ensemble methods do about as well as the best of the component models in terms of aggregate performance across multiple seasons , but that the ensemble methods have more consistent performance across different seasons . This improved consistency is valuable for planners who need predictions that can be trusted under all circumstances .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"influenza",
"seasons",
"probability",
"distribution",
"mathematics",
"forecasting",
"statistics",
"(mathematics)",
"discrete",
"mathematics",
"combinatorics",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"epidemiology",
"mathematical",
"and",
"statistical",
"techniques",
"probability",
"theory",
"permutation",
"earth",
"sciences",
"viral",
"diseases",
"physical",
"sciences",
"statistical",
"methods"
] |
2018
|
Prediction of infectious disease epidemics via weighted density ensembles
|
Late gene transcription in herpesviruses is dependent on viral DNA replication in cis but the mechanistic basis for this linkage remains unknown . DNA replication results in demethylated DNA , topological changes , removal of proteins and recruitment of proteins to promoters . One or more of these effects of DNA replication may facilitate late gene transcription . Using 5-azacytidine to promote demethylation of DNA , we demonstrate that late gene transcription cannot be rescued by DNA demethylation . Late gene transcription precedes significant increases in DNA copy number , indicating that increased template numbers also do not contribute to the linkage between replication and late gene transcription . By using serial , timed blockade of DNA replication and measurement of late gene mRNA accumulation , we demonstrate that late gene transcription requires ongoing DNA replication . Consistent with these findings , blocking DNA replication led to dissolution of DNA replication complexes which also contain RNA polymerase II and BGLF4 , an EBV protein required for transcription of several late genes . These data indicate that ongoing DNA replication maintains integrity of a replication-transcription complex which is required for recruitment and retention of factors necessary for late gene transcription .
The human gammaherpesviruses Kaposi's sarcoma-associated herpesvirus ( KSHV ) and Epstein-Barr virus ( EBV ) establish lifelong persistent infection in B lymphocytes and intermittently reactivate , producing infectious virions transmitted by the oral route [1 , 2] . KSHV is associated with Kaposi’s sarcoma , Multicentric Castleman’s disease and primary effusion lymphoma , whereas EBV is linked to nasopharyngeal carcinoma and a variety of lymphoproliferative syndromes and lymphomas [1 , 2] . In common with all herpesviruses , both KSHV and EBV sequentially express immediate-early ( IE ) , early ( E ) and late ( L ) genes during the lytic phase of replication . Immediate-early genes in the γ-herpesviruses encode transcriptional activators that initiate the lytic cycle and are necessary for virus reactivation from latency [3] . Many early gene products are required for lytic viral DNA replication , which is carried out by virus-encoded homologs of the cellular DNA replication machinery [3] . Late genes encode the structural virion components that comprise capsid , tegument and glycoproteins [1] . In addition to these essential aspects of virion production , lytic gene products are increasingly recognized as contributing to tumorigenesis [4] . Intermittent lytic replication and virion production may also contribute to maintenance of the latently infected reservoir in vivo [4] . Understanding the molecular mechanisms by which sequential gene expression is controlled during the lytic phase of replication is therefore important for identifying molecules and pathways that can be therapeutically targeted . Whereas expression of IE and E genes is unaffected by inhibition of DNA replication [5] , most EBV and KSHV late genes are strictly dependent on DNA replication from the lytic origin of replication , and inhibitors of viral DNA polymerase drastically inhibit late gene expression [6 , 7] . The basis for lytic DNA replication-dependent late gene expression has been a subject of investigation for many years but remains incompletely characterized . Whether DNA replication from an origin in cis to the late gene promoter is required for late gene transcription has been controversial . Two studies in EBV indicated that late promoter driven gene expression from plasmids could occur in the absence of plasmid DNA replication [8 , 9] . However , several studies using plasmid reporters and late promoters have concluded that the presence of DNA replication in cis is necessary to allow efficient late gene expression in KSHV , MHV-68 and EBV although virally produced proteins acting in trans are also necessary [10–13] . Similar conclusions were also drawn regarding HSV late gene expression from plasmids and from the viral genome [14 , 15] . Recent findings have revealed fundamental differences between late promoters and IE , E and cellular promoters in β and γ herpesviruses that provide a starting point for understanding the regulatory mechanisms of late genes in β and γ herpesviruses [8 , 13 , 16–21] . KSHV , EBV , murine gammaherpesvirus 68 and human cytomegalovirus all express a complex of six proteins which form a viral pre-initiation complex ( vPIC ) specific for late gene promoters and which are essential for late promoter function [8 , 16–21] . Homologous genes have also been identified in human HHV6 [22] . One of the proteins in the EBV vPIC , the BcRF1 gene product , functionally substitutes for TBP and binds to atypical TATT boxes in late promoters [18] . The vPIC has been demonstrated to bind RNAP at late promoters at which the vPIC forms [8 , 18] . A recent elegant study used recombinant EBV bacmids to demonstrate that each of the six components is individually required for late gene expression but do not affect DNA replication [23] . Further , this study confirmed that late gene expression from the EBV genome requires DNA replication in cis . Thus , it appears that the unique virus-specified late gene transcriptional machinery is dependent on DNA replication of the template on which the late gene promoter resides . Despite the discovery of the specific vPIC involved in β and γ herpesvirus late gene expression , the basis of its unique dependence on DNA replication in cis remains to be explained . Several possible mechanisms , which are not mutually exclusive , may be considered . First , late gene transcription may require chromatin modifications that occur concurrent with lytic genome replication , such as removal of methylation marks and histone occupancy [24 , 25] . The “naked” chromatin of newly replicated genomes may thereby allow more efficient access of the vPIC to late gene promoters . Alternatively , the translocating DNA replication complex may bring or recruit factors or components of the vPIC required for late gene transcription to the relevant promoters . Third , the physical localization of replicating genomes may facilitate or be required for co-localization of transcription factors needed for late gene transcription . For example , EBV lytic DNA replication occurs at discrete sites , called replication compartments , in which viral transcription also takes place . These include domains designated BMRF1 cores , in which newly synthesized viral DNA genomes are organized around and then stored inside cores composed of BMRF1 , the DNA polymerase processivity factor [26] . It has also recently been shown that KSHV assembles an "all-in one" factory for both gene transcription and DNA replication [27] . Finally , DNA replication increases template copy number and may contribute to late transcript abundance . In this study , we have investigated the involvement of several of these possible mechanisms , including demethylation , template abundance and the role of ongoing DNA replication and replication compartment formation . Our findings suggest that demethylation and template abundance are not significant factors in the linkage between late gene expression and DNA replication in KSHV and EBV . However , continuous , ongoing DNA replication appears to be necessary for late gene expression and maintenance of replication compartments .
Early experiments demonstrated that promoter hypomethylation correlated with transcriptional activity of EBV lytic genes but that demethylation alone was insufficient to induce lytic gene expression [28 , 29] . It is also known that DNA hypomethylation occurs during lytic replication and that both EBV and KSHV virion DNA are essentially devoid of DNA methylation and histones [25 , 29] . It was therefore possible that DNA hypomethylation that occurs coincident with lytic replication and production of linear DNA templates might be required for efficient transcription of late genes . In order to determine whether demethylation of KSHV latent DNA could rescue late gene transcription when DNA replication was blocked , we used the demethylase inhibitor 5-azacytidine ( 5-Aza ) in combination with phosphonoacetate ( PAA ) a selective inhibitor of herpesvirus DNA polymerases [30] . BAC16/iSLK cells are epithelial cells that inducibly express RTA ( ORF50 ) upon treatment with doxycycline and stably carry the eGFP-expressing JSC-1 KSHV bacmid BAC16 [31 , 32] . RTA expression is necessary and sufficient to induce lytic KSHV replication [33] , and treatment of BAC16/iSLK with doxycycline leads to highly efficient induction of the KSHV lytic cycle and production of infectious KSHV [19] . BAC 16/iSLK cells were cultivated in media supplemented with 2μM 5-Aza for two days to demethylate DNA before induction of lytic replication . Lytic viral replication was induced with doxycycline and DNA replication was simultaneously inhibited with PAA . Two days post induction , RNA and DNA were harvested and qPCR was performed to quantitate individual gene expression and viral genome copy number . As expected , RNA transcription of early genes ( ORF6 and ORF57 ) was highly induced by doxycycline but was unaffected by PAA treatment ( Fig 1A ) . There was a further increase when iSLK cells were treated with 5-Aza , indicating that methylation inhibits early gene promoter activity . Demethylation also upregulated expression of late genes , ( ORF25 , ORF26 and K8 . 1 , Fig 1B ) . However , consistent with the dependence of late gene transcription on DNA replication , PAA completely inhibited transcription of K8 . 1 , ORF25 and ORF26 . Importantly , even when viral genomes were demethylated by 5-Aza , transcription was strongly inhibited by PAA , demonstrating that removal of methylation is not the reason that DNA replication is required for late lytic gene transcription . In order to determine whether the dependence of late gene transcription on DNA replication was similarly independent of the methylation status of template DNA during EBV replication , we repeated the same experiment in AGSiZ cells . AGSiZ is an EBV bacmid infected gastric carcinoma-derived cell line in which a doxycycline-inducible EBV transactivator Zta has been stably introduced by lentivirus transduction [34] . As expected , EBV early genes ( BMRF1 and SM ) were highly expressed without being inhibited by PAA treatment ( Fig 1C ) and a further increase in BMRF1 and SM mRNA levels was observed upon treatment with 5-Aza . Expression of late genes ( BDLF1 , BcLF1 and BCRF1 ) was highly inhibited by PAA treatment ( Fig 1D ) . However , demethylation did not affect the dependence of late gene transcription on DNA replication as PAA was equally effective in inhibiting late gene expression despite the presence of 5-Aza . These data therefore indicate that DNA demethylation occurring during DNA replication is not the basis of the linkage between late transcription and genome replication in either KSHV or EBV . We confirmed that PAA had effectively inhibited KSHV and EBV lytic replication by measuring the KSHV and EBV copy number in cells used for the above experiment . Increases in intracellular EBV and KSHV DNA are maximal by 4–5 days post induction . As shown in Fig 2A and 2B , PAA completely blocked the increase in KSHV and EBV genome copy number seen 5 days after induction of lytic replication . Another possible basis for replication dependent late gene transcription is the increase in template number that occurs with lytic DNA replication . To ask whether there was a significant increase in KSHV template number at 48 hours when 500–1 , 000-fold increases in late transcript abundance were observed ( Fig 1 ) , we measured KSHV DNA copy number at 48 hours after induction . KSHV genome copy number was measured by qPCR of DNA extracted from the same cells that were used for measurement of mRNA transcript levels . As shown in Fig 2C , DNA copy number had only increased by 20% at 48 hours , in the absence of PAA , demonstrating that the large increases in late gene transcript levels precede those in DNA copy number . Increased KSHV template numbers are therefore unlikely to play a role in replication dependent increases in KSHV late gene transcription . When similar measurements were performed to measure EBV copy numbers in AGSiZ cells , the kinetics were different , EBV replication occurring somewhat earlier , with increases in copy number at 48 hr ( approximately 7-fold ) ( Fig 2D ) . We therefore repeated the experiment in AGSiZ cells and harvested cells for RNA and DNA 24 hr post induction to determine whether late gene transcription could be demonstrated to precede increases in genome copy number . Although as expected , late gene transcription was not as pronounced as at 48 hr , all patterns at 24 hr were similar to those observed at 48 hr: Early gene transcription was insensitive to PAA ( Fig 1E ) and 5-Aza treatment did not rescue late gene transcription when DNA replication was inhibited ( Fig 1F ) . At this time point , DNA copy number however , had not increased more than 10% from levels prior to induction ( Fig 2E ) . Therefore , increases in template numbers did not play a role in replication dependent stimulation of EBV or KSHV late gene transcription . Although the increases in both early and late gene transcription seen with 5-Aza were consistent with promoter demethylation , we performed direct analyses to confirm CpG demethylation by 5-Aza . We analyzed a known methylated region in the cellular genome ( LINE-1 , L1-PKP4 [35] ) by COBRA ( combined bisulfite restriction analysis ) ( Fig 3 ) . DNA was amplified with primers specific for bisulfite converted DNA ( in which unmethylated , but not methylated C , is converted to T ) and then digested with TaqI . As shown in Fig 3B and 3C , specific amplification in bisulfite treated samples yielded a 392 bp PCR product , but not in bisulfite untreated samples from iSLK and AGSiZ cells . Bisulfite treatment is predicted to create 3 TaqI cut sites ( TCGA ) in the PKP4 PCR product , but only from methylated CCGAs ( Fig 3A , left panel ) . Bisulfite-treated PCR products demonstrated almost complete TaqI cleavage , confirming the methylated status of the amplified region ( Fig 3B and 3C , lanes 9 and 10 ) . In 5-Aza-treated samples , analysis of the PCR products revealed a mixture of partially cleaved and un-cleaved bands ( Fig 3B and 3C , lanes 11 and 12 ) . These data therefore demonstrate that 5-Aza led to CpG demethylation and prevented production of TaqI sites from methylated CCGA sites . In addition to demethylation of cellular DNA , we wished to confirm that viral genomes were also demethylated by 5-Aza . The methylation status of the ORF33 promoter in KSHV genome , which has been previously demonstrated to be methylated [25] , was therefore assessed by pyrosequencing . Five CpG sites in the ORF33 promoter showed changes in the level of methylation upon 5-Aza treatment ( Table 1 ) . At 2 days post induction of lytic replication ( 4 days after 5-Aza treatment ) , the average percentage change of methylation had decreased in all 5-Aza treated samples . Lytic replication led to demethylation ( 15 . 7% change ) and when lytic replication was blocked by PAA there was no change in methylation . However , 5-Aza treatment in the presence of PAA led to a degree of demethylation similar to that of lytic replication ( 15 . 7% change ) . This experiment demonstrated that the KSHV viral genome DNA was demethylated by 5-Aza treatment , consistent with its observed effects on gene expression ( Fig 1 ) . Since neither removal of DNA methylation nor increasing template number appeared to be the reason for the linkage of late gene expression to DNA replication , we considered the possibility that DNA replication removed other protein factors from latent episomes that are chromatinized during latency . If such a mechanism were involved , initiation of DNA replication could be sufficient to enable late gene transcription . Alternatively , if DNA replication were required to bring proteins required for late gene transcription to the late gene promoters , or were necessary to maintain a replication factory with the requisite transcription factors , continuing ( ongoing ) DNA replication would be necessary to maintain late gene transcription . To distinguish between these alternatives , we performed the following experiment to assess the effect of blocking DNA replication at various times after induction of lytic replication . We induced lytic KSHV replication in iSLK cells and added PAA at 0 hr , 12 hr , 24hr , 36hr post-induction , and then harvested RNA at 48 hr post induction . At each time point when we add PAA , we also harvested untreated cells in parallel to compare the temporal pattern of RNA accumulation . The results showed that KSHV late gene ( ORF25 and ORF26 ) accumulation at 48 hr was inhibited compared to no PAA treatment when we added PAA at any time after induction ( Fig 4A and 4B] . On the other hand , early gene ( ORF59 and ORF57 ) accumulation did not change ( Fig 4C and 4D ) . These data suggest that KSHV late gene expression requires ongoing DNA replication to maintain transcription . If initiation of DNA replication merely established a permissive condition for late gene transcription , accumulation should continue to increase even after PAA was added . We then further extended the time course to 96 hr post-induction and added PAA up to 72 hours post-induction to maximize the opportunity for genomes potentially licensed by DNA replication to transcribe late genes . As shown in Fig 4E and 4F , late gene RNA accumulation at 96 hr was still inhibited when we added PAA at any time point compared to no PAA treatment . Addition of PAA at any time point essentially led to cessation of further RNA accumulation . These results demonstrated that continued KSHV late gene mRNA accumulation requires maintaining ongoing viral DNA replication . We next asked whether EBV late gene expression also required DNA replication for continued late mRNA transcription . We performed an analogous experiment with EBV in AGSiZ cells . Since EBV replication and increases in DNA copy number occur earlier post-induction in this system than in iSLK cells ( Fig 2 ) , we measured RNA accumulation at 24 hr post-induction and added PAA at serial times prior to 48 hr ( 0 , 6 , 12 , and 18 hr ) . Similar to KSHV , EBV late gene ( BDLF1 and BcRF1 ) mRNA accumulation at 24 hr was inhibited when we added PAA at any time point ( Fig 5A and 5B ) while early gene ( BMRF1and SM ) accumulation was unaffected ( Fig 5C and 5D ) . These results confirmed that late gene transcription requires ongoing DNA replication in both KSHV and EBV . A potential explanation for the dependence of late gene transcription on continued DNA replication is that the process of lytic replication , which involves formation of nuclear replication factories , that include recruited transcription factors [23 , 26] , is essential for maintenance of the transcriptional milieu required for late gene transcription . We therefore wished to ask whether inhibiting DNA replication could affect the maintenance of EBV replication factories . The EBV DNA polymerase processivity factor BMRF1 ( EA ) concentrates in EBV replication factories and serves as a marker of these compartments in immunofluorescence studies [26 , 36] . Identification and visualization of these nuclear replication compartments is easily performed in EBV infected 293 cells by staining for BMRF1 . We first performed preliminary experiments in which we followed the formation of the replication compartments after induction of lytic replication . We induced EBV replication in EBV-infected 293 cells by transfection of lytic activator Zta expression plasmid . We then fixed and stained the cells with anti-BMRF1 antibody at various times post-induction . As shown in Fig 6 , formation of replication factories is clearly evident by 24 hr post-induction , and continues through 96 hr . Consistent with previous reports [26] , concentration and co-localization of RNA pol II in these nuclear replication foci is also evident ( but not in cells in which EBV replication was not induced ) . We then asked what would happen to these replication compartments if DNA replication were blocked after they had been allowed to form . We added PAA after the formation of replication complexes , at 24 hr and 48 hr post replication induction , and stained the cells at 96 hr . Addition of PAA at either time point led to loss of the large , discrete foci typical of cells in which DNA replication was not blocked . Rather , BMRF1 became dispersed throughout the nucleus in fine speckles . RNA pol II remained associated with these residual speckles , also assuming a diffuse nuclear distribution . These data suggest that continuing DNA replication is required to maintain a nuclear replication/transcription factory . In order to confirm the observations shown above , we measured the changes in replication complex structure by counting the percentage of cells that contained the large clusters and smaller clusters ( spots ) versus the fine speckles more commonly seen after PAA treatment . As shown in Fig 7A , the number of cells with clearly formed larger replication foci ( spots/clusters ) increased over time with over 65% of BMRF1-positive cells containing such complexes . However , when PAA was added at 24 or 48 hr , by 96 hr , the number of cells with large foci had decreased to 29% and 38% respectively . While cells with replication foci do not completely disappear upon replication inhibition , they undergo decreases that are highly significant when compared to their initial prevalence at the time of PAA addition and at 96 hr ( Fig 7B , left panel ) . As expected , the percentage of cells with speckles increased commensurately after PAA addition as shown in Fig 7B ( right panel ) . Representative cells of each category are shown in Fig 7C . In order to confirm that these effects were not unique to PAA , we used two additional inhibitors of EBV DNA replication , acyclovir and ganciclovir to assess their effect on replication complex maintenance . AGSiZ/EBV cells were induced to permit EBV replication and the replication inhibitor was added either 24 or 48 hr later and cells were fixed at 96 hr post-induction . Cells untreated with either drug were induced and fixed at 96 hr post induction exactly as performed previously with PAA . Fixed cells were then stained with anti-BMRF1 and anti-RNAP antibodies and examined by fluorescence microscopy . The results , shown in Fig 8 , confirm that inhibiting EBV DNA replication with either ganciclovir or acyclovir leads to loss of replication complex integrity . As with PAA , RNA pol II appears to maintain association with BMRF1 in the finer speckles resulting from blocking DNA replication . Disintegration of replication complexes upon blockade of DNA replication suggested that complex integrity might be necessary for recruitment of transcription factors . Since RNA pol II appeared to remain associated with BMRF1 even after complex dispersal , we wished to ask whether inhibition of DNA replication might lead to loss of other factors needed for late gene transcription . BGLF4 , an EBV encoded kinase with multiple functions in viral packaging and egress [37 , 38] , has also recently been shown to be essential for transcription of several late genes [39] and to be present in replication complexes [40] . We therefore examined the location of BGLF4 after inducing EBV replication and after inhibiting DNA replication . In cells undergoing replication , BMRF1 was localized in mostly large structures . BGLF4 was found in these structures as well , although the co-localization with BMRF1 was not 100% . Whereas the majority of BMRF1 complexes contained BGLF4 , nuclear foci of BGLF4 without BMRF1 were also present ( Fig 9 , 48h + Z ) . After treatment with PAA , the BMRF1 foci exhibited dispersal as expected . In addition , BGLF4 was no longer highly co-localized with BMRF1 . In order to quantify and confirm this effect , we counted nuclei that were BMRF1 positive and categorized the nuclei as either co-localized or not co-localized . Cells where greater than 75% of the BMRF1 foci also contained BGLF4 were counted as co-localized and vice versa ( Fig 10 ) . The number of cells exhibiting co-localization decreased significantly after PAA treatment . These data indicate that when the replication factories undergo dispersal , there is a concomitant loss of BGLF4 recruitment .
We have investigated the factors that link late gene transcription to lytic DNA replication in the γ herpesviruses EBV and KSHV . We have shown the DNA demethylation that occurs consequent to lytic replication of KSHV and EBV genome by viral DNA polymerases is unlikely to be the mechanism by which late gene transcription is linked to DNA replication in cis . Increases in late gene expression also preceded increases in DNA template copy number , confirming that template amplification does not contribute significantly to late transcript accumulation . Temporal analysis of the relationship between DNA replication blockade and late mRNA accumulation indicated that KSHV and EBV late gene expression require ongoing DNA replication . Further , RNA pol II and the EBV kinase BGLF4 colocalized with DNA polymerase processivity factor in EBV replication factories and blockade of DNA replication led to loss of replication compartment integrity . DNA methylation has an important regulatory role in gene expression during the KSHV and EBV life cycle . The EBV and KSHV genomes are highly methylated during latency , and become hypomethylated or unmethylated during lytic infection [29 , 41] . Unlike cellular DNA or latent herpesvirus genomes , newly produced herpesviral genomes are not remethylated after lytic replication , and virion DNA is unmethylated [42–44] . DNA demethylation has been shown to upregulate viral gene expression and induce EBV and KSHV lytic replication [29 , 41 , 45] . Thus , it was plausible that replication-dependent demethylation was required for late gene promoter activity . We found that viral DNA demethylation by 5-Aza treatment potentiated both EBV and KSHV early and late lytic gene expression ( Figs 2 , 3 and 4D ) . Although demethylation increased both early and late lytic promoter activity , it could not rescue late gene transcription from DNA replication blockade . These data therefore indicate that removal of CpG promoter methylation is not the mechanism of late gene expression dependence on DNA replication . Previous studies using plasmids with lytic herpesviral origins and late gene promoters have suggested that increases in DNA template number are not required to support late gene expression in cis [11] . In order to confirm these findings in the context of entire viral genomes , we measured the viral DNA content of lytically induced KSHV and EBV infected cells over time and compared it to the temporal accumulation of late gene transcripts . The increase in transcript levels occurred prior to increases in DNA copy number , confirming that replication induced template abundance does not play a significant role in enhancing late gene transcription . Thus , DNA replication has an effect on late gene transcription very early after it begins and before template number is significantly increased . It therefore appeared that late gene transcription occurs from newly replicated genomes , and some property of the newly replicated DNA makes it supportive of late gene transcription . Two alternative models for the permissiveness of nascent herpesvirus genomes can be envisioned with slightly different implications . First , a topological or steric property of newly replicated DNA allows access of the late gene vPIC and the RNA polymerase to the relevant promoters . Such a mechanism would suggest that once a linear or partly linear , non-supercoiled genome were produced , it would be permissive for continued late gene transcription , and subsequent blockade of DNA replication would not shut off persistent late gene expression from the “licensed” genomes . Alternatively , a mechanism that requires continuing operation of the DNA polymerase might be required . The DNA polymerase or associated factors might bring or recruit essential transcription cofactors to the late gene promoters as has been demonstrated to occur during replication of bacteriophage T4 [46] . A recent report of interaction between ORF59 , the KSHV DNA polymerase processivity factor and RNA pol II make this an attractive hypothesis [47] . In this alternative scenario , blocking DNA replication at any point would strongly inhibit , if not curtail , further late gene expression as continued delivery of proteins essential for transcription would cease . A related model is one in which the replication factory: a complex of the vPIC , late gene transcription factors , RNAP , the replicating genome and DNA polymerase , requires ongoing DNA replication to be maintained . Finally , if linear genomes enter a non-permissive state for late gene transcription shortly after replication , either by temporary chromatinization or encapsidation , continued production of new templates would be required . We first asked whether DNA replication licensed templates for continued late gene transcription , that is , whether stopping DNA replication would still allow late gene transcription from linear genomes that had previously been synthesized . We inhibited DNA replication at various times after induction of lytic transcription , and either harvested RNA or allowed potential transcription to proceed and measured RNA accumulation at the end of a fixed period allowed for transcription . A similar analysis of HSV late gene transcription concluded that continued DNA replication was not required for additional synthesis of late gene mRNAs [48] . However , in the case of both KSHV and EBV , we found that inhibiting DNA replication essentially led to a cessation of further RNA accumulation beyond that point . Thus , it appears that in the absence of continuing DNA replication , the newly replicated genomes do not remain competent for transcription . Possible explanations for this phenomenon stem from the basic concept that changes occur to the nascent genomes shortly after their production that render them no longer accessible for transcription . One mechanism by which this could occur is the deposition of proteins on linear genomes that prevent further transcription unless removed by additional rounds of replication as depicted in Fig 11A . While it is generally accepted that herpes virus genomes prior to packaging are devoid of histones or other chromatin marks , it is possible that linear genomes undergo a period of rechromatinization prior to the removal of histones upon encapsidation . However , histones have been reported to be excluded from EBV replication factories during late stages of lytic replication [49] . Similarly , a recent report shows that newly replicated KSHV genomes do not associate with histones [50] . Thus , we consider this model less likely as an explanation of DNA replication and late transcription linkage . Nevertheless , it remains possible that other proteins may bind to the newly replicated genomes after synthesis [50] , and such proteins may prevent continued access of the late gene transcription complex to the DNA . An alternative model is that topological changes in the newly replicated genomes allow access to late gene transcription complexes . There may be a limited period of time shortly after synthesis when the linear genomes sterically allow access of the RNA polymerase complex but then assume conformations , perhaps due to encapsidation , that render them no longer accessible to the RNA polymerase machinery ( Fig 11B ) . Finally , replication may also be necessary to continually bring or recruit factors necessary for late gene transcription to the late gene promoters ( Fig 11C ) . A distinct , although not mutually exclusive , mechanism is suggested by our finding that there is a physical dissolution of the replication complexes upon cessation of EBV DNA replication . The process of ongoing DNA replication thus seems to be required to maintain the integrity of the replication factory . Consistent with previous findings , we have shown that RNA pol II co-localizes with BMRF1 during EBV lytic replication in large nuclear structures [26] . The late gene vPIC complex is recruited to noncanonical TATT boxes of true late gene promoters; components of the vPIC have been shown to interact with RNA polymerase II; and late gene mRNAs localize to the inner cores of replication factories [8 , 18 , 26 , 51] . BGLF4 , which is required for transcription of late genes encoding structural proteins [39] , also is found in the replication complex [52] . The fact that these replication/transcription factories , in which late lytic gene transcription and DNA replication occur , undergo breakdown when DNA replication stops , suggests that a functioning complex of DNA polymerase and viral genome is required to maintain the physical and functional integrity of the structure . When DNA replication stops , the replication/transcription factory begins to disintegrate ( Fig 11D ) . It appears that while some association of DNA replication proteins with transcription machinery , such as with RNAP , may persist , others such as with BGLF4 , may not . Ultimately it appears that late gene transcription is dependent on a on a very tight physical and temporal association with ongoing lytic DNA replication . Such a mechanism would help to ensure that late gene products are synthesized concurrently and at the appropriate time to be available for packaging of newly replicated genomes .
iSLK cells [32] , kind gift of Don Ganem , were maintained in DMEM containing 10% charcoal stripped FBS ( Sigma ) and 1% glutamine with 250 μg/ml G-418 and 1 μg/ml puromycin . iSLK cells were infected with WT KSHV derived from bacmid BAC16 , expressing eGFP and hygromycin resistance ( kind gift of Jae Jung ) [31] . KSHV infected iSLK cells were maintained in 1 . 2 mg/ml hygromycin , 250 μg/ml G-418 and 1 μg/ml puromycin . We generated an EBV positive gastric carcinoma cell line AGSiZ by transducing AGS/BX1 cells [53] ( gift of Lindsey Hutt-Fletcher ) , with a lentivirus expressing the EBV transactivator protein Zta ( BZLF1 ) under control of a doxycycline inducible promoter [34 , 54] . Cells were grown in F-12 media containing 10% charcoal stripped FBS ( Clontech ) and 1% glutamine with 500 μg/mL G-418 and 0 . 5 μg/mL puromycin . For demethylation of iSLK cells or AGSiZ cells , iSLK or AGSiZ cells were cultivated in media supplemented with 2 μM 5-Aza for two days . Viral lytic replication was induced by treatment with 1 μg/ml doxycycline . For inhibition of viral DNA replication , 500nM PAA was added simultaneously with 5-Aza . 293 cells infected with 2089 B95-8 bacmid ( kind gift of Henri-Jacques Delecluse ) [55] were grown in DMEM with 10% FBS and 100 μg/ml hygromycin .
800 ng genomic DNA was treated with sodium bisulfite using an EpiTech Bisulfite Kit ( Qiagen ) according to the manufacturer’s protocol . Two microliters of bisulfite converted DNA were subjected to 39 cycles of PCR with the following primers to amplify the PKP4 locus [35]: PKP4 1F: GGTATGATTTTAAAAAAAGAGAT PKP4 1R: GTAAAACCCTCCGAACCAAATATAAA . PKP4 PCR product ( 3ul ) was digested with TaqI ( NEB ) at 65°C for 2 . 5 hrs . The undigested and digested samples were separated in a 2% agarose/ethidium bromide gel to assess the demethylation effect of 5-Aza . Cleavage only occurs if the cytosines in the restriction sites are preserved as cytosines during the bisulfite modification as a result of methylation . Methylation status of CpG sites in the KSHV ORF33 promoter region was examined by a pyrosequencing-based methylation assay using the PyroMark Q24 instrument ( Qiagen ) . Briefly , primers for the polymerase chain reaction ( PCR ) and pyrosequencing were designed using PyroMark Assay Design Software 2 . 0 ( Qiagen ) . Bisulfite-converted DNA samples were amplified using the PyroMark PCR kit ( Qiagen ) with following PCR primer sets: ORF33 BS1F: TGTTGGATGGAGGTGTTAGGATTATG ORF33 BS1R: ACCTTTAATAACAAAACCCCCAAAT ( Biotinylated ) . ORF33 seq1: GTGTTAGGATTATGGGAAA was used for pyrosequencing in the PyroMark Q24 instrument . Data analysis was carried out with PyroMark Q24 Software ( Qiagen ) . 293 cells infected with 2089 B95-8 bacmid [55] were grown on glass coverslips plated at 250 , 000 cells per well in six-well dishes and transfected with Zta expression plasmid ( +Z ) to induce EBV lytic replication by using TransIT293 ( Mirus ) according to the manufacturer’s protocol . PAA ( 500nM final concentration ) was added to block viral replication at different time points as indicated in the text . 96 hours after transfection , cells were washed with 1x PBS , fixed and permeabilized with PBS containing 4% paraformaldehyde and 0 . 2% Triton X-100 for 15–20 minutes at room temperature , and then washed two times with 1x PBS followed by incubating with blocking buffer ( 20% goat serum in PBS ) for 30 mins at room temperature . Finally , the cells were incubated with anti-BMRF1 monoclonal antibody ( Capricorn ) and polyclonal anti-RNAP antibody at 37°C for 1 hour . For BGLF4 visualization , rabbit anti-BGLF4 antibody , kind gift of Ayman El-Guindy [39] , was used at a dilution of 1:1000 . The slides were washed three times with 1x PBS and incubated with secondary antibody Alexa Fluor 594 goat anti-mouse IgG , or Alexa Fluor 647 goat anti-rabbit IgG for 1 hour at 37°C ( in the dark ) . Nuclear staining was performed with 4’ , 6- diamidino-2-phenylindole ( DAPI ) ( Invitrogen ) . Images were collected and analyzed with a ZEISS Imager M2 microscope system . Statistical testing for comparison of proportions and p values was performed using MedCalc software , which uses the "N-1" Chi-squared test [56] .
|
Herpesviruses exhibit both latent and lytic replication cycles . Gammaherpesviruses such as Kaposi’s sarcoma-associated herpesvirus and Epstein Barr virus undergo lytic replication when they reactivate from latency . During this process , when infectious virions are produced , an orderly cascade of gene expression occurs . Late lytic genes , which primarily encode structural components of the virion , are only transcribed after replication of the DNA genome has occurred . Unlike early lytic genes , late gene transcription is tightly linked to viral DNA replication; if viral DNA replication is blocked , late gene mRNA accumulation is severely inhibited . The mechanism by which late gene transcription is linked to DNA replication has remained elusive . In this paper we show that a process of continuous DNA replication is required . If one blocks DNA replication , further transcription also ceases , indicating that concurrent DNA replication is required to maintain late transcription . We also show that when DNA replication is blocked , the nuclear complexes in which herpesviruses are replicating dissociate . These replication complexes also serve as factories of viral transcription . When the complexes disperse , proteins required for transcription dissociate from the DNA replication machinery . These data indicate that ongoing DNA replication is necessary to maintain the physical and functional integrity of these structures . Our study provides new insight into this linkage that ensures coordination between viral replication and late gene expression .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods",
"DNA",
"and",
"RNA",
"isolation",
"and",
"analysis"
] |
[
"medicine",
"and",
"health",
"sciences",
"functional",
"genomics",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"dna",
"transcription",
"viruses",
"dna",
"replication",
"dna",
"viruses",
"dna",
"epigenetics",
"dna",
"methylation",
"chromatin",
"herpesviruses",
"chromosome",
"biology",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"chromatin",
"modification",
"kaposi's",
"sarcoma-associated",
"herpesvirus",
"dna",
"modification",
"viral",
"replication",
"biochemistry",
"cell",
"biology",
"nucleic",
"acids",
"virology",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"organisms"
] |
2018
|
Continuous DNA replication is required for late gene transcription and maintenance of replication compartments in gammaherpesviruses
|
The evolution of proteins is one of the fundamental processes that has delivered the diversity and complexity of life we see around ourselves today . While we tend to define protein evolution in terms of sequence level mutations , insertions and deletions , it is hard to translate these processes to a more complete picture incorporating a polypeptide's structure and function . By considering how protein structures change over time we can gain an entirely new appreciation of their long-term evolutionary dynamics . In this work we seek to identify how populations of proteins at different stages of evolution explore their possible structure space . We use an annotation of superfamily age to this space and explore the relationship between these ages and a diverse set of properties pertaining to a superfamily's sequence , structure and function . We note several marked differences between the populations of newly evolved and ancient structures , such as in their length distributions , secondary structure content and tertiary packing arrangements . In particular , many of these differences suggest a less elaborate structure for newly evolved superfamilies when compared with their ancient counterparts . We show that the structural preferences we report are not a residual effect of a more fundamental relationship with function . Furthermore , we demonstrate the robustness of our results , using significant variation in the algorithm used to estimate the ages . We present these age estimates as a useful tool to analyse protein populations . In particularly , we apply this in a comparison of domains containing greek key or jelly roll motifs .
The current wealth of freely available genetic sequences offers the potential to uncover the evolutionary history of genes and their products , proteins . While there exist no remains of primitive proteins , extant protein information can be used to estimate a protein family's history . This approach is particularly well suited to structural information . Protein structures are far more conserved than their sequences and thus preserve a deep phylogenetic signal [1] . Furthermore , for the majority of globular proteins , a stable three-dimensional structure is thought to be a requirement for many aspects of its function . By maintaining the precise positioning of functional residues while also minimising other undesirable interactions a protein's structure is intimately linked to the role it plays within the cell [2] . Moreover , phylogenetic trees built using the structural content of species' proteomes have been shown to produce more reliable topologies than trees constructed using their protein sequences [3] . These observations support the use of structure as a fundamental molecular unit when studying the evolution of proteins . Furthermore , they suggest that any conversation on the evolution of proteins must first understand the major driving forces behind such changes from a structural perspective . In order to visualise the landscape and diversity of structure space protein structures have been clustered within a hierarchical taxonomy [4] , [5] . The SCOP database is one such manual classification scheme which , at the superfamily level , attempts to cluster together protein domains with a common evolutionary origin , based primarily on strong functional and structural similarity [6] . The superfamily classification lies in between the family level , largely defined by a domain's amino acid sequence , and the fold , a structural consensus of a domain's topology . In this work we primarily consider sets of structural domains classified as superfamilies in SCOP 1 . 75 . Despite the potential for rich diversity within the structural universe it is surprising how sparse this space remains [7] . The current repertoire of proteins with known structure fall into less than 1 , 200 unique SCOP folds and the majority of these contain only one sequence family [8] . While this is unlikely to represent the true diversity of naturally occurring proteins and current projections for the size of protein fold space range from around 2 , 000 [9] to over 10 , 000 [10] , it is thought that the vast majority of extant proteins will fall into only around 1 , 000 common folds [11] . Furthermore , the landscape of this core fold space is highly heterogeneous , with a few so called ‘superfolds’ densely populated by sequence families [12] . The unique composition of this space is a consequence of protein evolution through neutral drift and active selection together with a complex interplay of other factors such as genome structure , mutational mechanisms , function and the need for interactions , all of which close off portions of the configuration space . However , little is known about the exact nature by which the range of protein structures we see today have evolved [2] . One way in which we can seek to explore the forces behind such a history is to consider annotating the protein structure universe with an estimate of its evolutionary age [7] , [13] , [14] . The age of a population of proteins is the estimated node age of its first ancestor across a phylogeny of completely sequenced genomes . This method has been implemented for both structural superfamilies [13] , [15] and sequence families [7] , [14] , although the latter tend to involve a much reduced phylogenetic tree and evolutionary scale . Methods for predicting the internal node of the ancestor for a given family or superfamily also vary . A maximum parsimony model for superfamily evolution has been largely adopted for this step [13] , [15] , [16] , although alternatives include Dollo parsimony: taking the most recent common ancestor [7] , [13] . These parsimony models take as input a phylogenetic species tree and the occurrence profile of each structural superfamily across this set of species . The occurrence profile for a superfamily is simply its presence or absence on each of the genomes [16] . Parsimony attempts to reconstruct the most likely series of gain and loss events at internal nodes of the tree which explain the occurrence profile at its leaves . The likelihood of these events is based on simple assumptions relating to the evolution of protein domains . The principle underlying all types of parsimony is that the scenario of events involving the least evolutionary change is preferred . Gain events can represent de novo superfamily gain , lateral gene transfer of a superfamily between genomes , and a false positive assignment of a superfamily to a genome . Loss events can represent the loss of a superfamily and also false negative assignments to a genome . Maximum parsimony methods allow for a weighting of the likelihood of loss events relative to gain events , while Dollo parsimony considers a gain event to be so rare it is most likely to have occurred only once in the evolution of a superfamily . Since lateral gene transfer is rare between Eukaryotes but may be quite common among Prokaryotes it has been suggested that maximum parsimony is an appropriate model for Prokaryotic genomes while Dollo parsimony should be used for Eukaryotes [16] , [17] . Previous studies have shown a significant positive correlation between the age of a domain's structure and its length [7] , [14] . These results remain pronounced over different methods for calculating the age of a superfamily or protein sequence . This seemingly fundamental relationship between the age of a structure and its length has supported the idea that the primitive protein universe was populated mainly by small folds [7] . In fact , the recent success in using structural fragments to predict protein structures ( see , for example [18] ) has further stimulated debate as to whether the evolutionary origins of the current fold space are in fact short peptide fragments that have combined to form larger folds [19] . It has also been reported that class domains tend to be significantly older than superfamilies belonging to other classes [13] . domains also tend to be significantly longer than other classes but they are also distinguishable in other respects [20] . They are unique among the classes in containing a majority of parallel -strands as opposed to the antiparallel structure which characterise all- and classes . folds also contain a large number of the so-called ‘superfolds’: folds containing large numbers of different superfamilies and a high proportion of all determined structures [12] . Such superfolds include P-loop NTPases , Rossmann folds and TIM barrels [11] . In this work we present phylogenetic profiles and evolutionary ages for superfamilies representing the current known structural universe . We show that these age estimates are largely robust to different evolutionary models , datasets and phylogenetic trees . We compare the structural characteristics of two protein populations: new-borns , with biologically recent structural ancestors , and ancients , with ancestors at the root of the tree of life . Our results identify several characteristics that differ between the two populations . These differences support known relationships , such as the propensity of and longer superfamilies to be ancient , and also postulate several previously unseen characteristics which correlate with age . While these structural relationships are marked we considered the possibility that they were the result of an asymmetry in the functional annotation of fold space . Here we show that our structural partitions result in far more dramatic age differences than functional groupings and as such the relationships between structure and age are not a residual effect of functional preferences .
Superfamily ages are sensitive to the phylogenetic tree of life used , the prediction of superfamilies on genome sequences for the occurrence profiles , and to the parsimony method and parameters used to estimate events . In order to investigate the robustness of our age estimates to these assumptions we undertook our analysis across several phylogenetic trees and multiple parsimony models . We also explored the effect on our results of using different datasets: changing both the occurrence profiles and the set of genomes considered . Representative domains for these superfamilies were taken from the ASTRAL database [22] . A number of different properties pertaining to the sequences , structures and functions of these domains were then used to compare the ancient and new-born populations . The enrichment of disulphide bonds among new-born superfamilies indicated a potential over-representation of cysteine residues among these superfamilies . We investigated whether there were further relationships with other amino acids . Very little is known about the evolution of early life but it is a common theory that the twenty amino acids we see today did not appear simultaneously . It is likely therefore that the earliest peptides consisted of only a subset of these amino acids: the first to evolve . Trifonov suggests a chronological order for the evolution of these amino acids: Gly , Ala , Asp , Val , Pro , Ser , Glu , Leu , Thr , Arg , Ile , Gln , Asn , His , Lys , Cys , Phe , Tyr , Met , Trp [27] . We looked here at the sequence composition of different domains and the propensity for different amino acids for ancient or new-born superfamilies . Since sequence change is rapid compared to structural change it is unlikely that the composition of the earliest peptides could be detected from their extant descendants . However , the propensities calculated here may still hold some signal of preference for certain amino acids . Propensities were calculated for all 20 amino acids across the two age groups and are shown in Table 1 . While amino acids predicted by Trifonov to occur early during protein evolution were more likely to be enriched in ancient superfamilies this relationship was by no means strict . Amino acids significantly over-represented in ancient superfamilies are Arg , Gly , and Val , which are hydrophobic , non-polar residues , with the exception of Arg , which is polar and positively charged . Residues over-represented in new-born superfamilies are Asn , Cys , Gln , Ser , Thr , Trp and Tyr . These residues are mostly polar and uncharged . Trp and Tyr also contain large , aromatic side chains . The propensities in new-born superfamilies for polar residues further supports our previous observation that newly evolving structures may have a larger surface area to volume ratio . In this study we have primarily focussed on the structural properties characterising superfamilies rather than on their functional roles . We performed enrichment analysis of GO functions for populations of superfamilies in the different age groups . We compared three different age groups: new-born , ancient and middle-aged superfamilies ( those superfamilies in neither the new-born or ancient groups ) . A list of all terms which were significantly enriched can be found in Table S1 . It has been observed in a study of the protein interaction network of yeast that older proteins tend to have more interaction partners than either middle-aged or young proteins [28] . This would appear to indicate that older superfamilies will tend to have more enriched functional terms than younger superfamilies , since partners in the interaction network will tend to share functional annotations . Indeed we find this to be the case . Of 189 GO terms found to be enriched in any one of the three age groups ( ancient , middle-aged or new-born ) , none were enriched in new-born superfamilies , 8 in middle-aged superfamilies and the remaining 181 were enriched in ancient superfamilies . The terms enriched in middle-aged superfamilies refer mostly to the regulation of developmental growth unique to Eukaryotes . The majority of terms enriched in ancient superfamilies correspond to fundamental cellular processes common to the vast majority of the tree of life . Interestingly , while RNA synthesis is enriched in ancient superfamilies , terms relating specifically to DNA synthesis are not . This supports the RNA world hypothesis , that during early evolution genetic material was stored as RNA as opposed to DNA [29] . For full details of the functional terms enriched in our age groups see Table S1 . We considered the possibility that the structural biases of ancient and new-born superfamilies we report here might be a residual effect of a more fundamental relationship with function . For example , we observe a strong relationship between ancient superfamilies and parallel strands . But , as mentioned before , folds are often superfolds , and are known to be associated with a large repertoire of fundamental functions . Perhaps it is the enrichment of these functions in the class that drives the preference for ancient superfamilies to have parallel strands . We compared our structural ages ( Figure 2c ) with ages for populations of superfamilies annotated with functional terms enriched in either parallel or antiparallel superfamilies . In order to do this we constructed lists of parallel/antiparallel functions: GO terms significantly enriched in the subset of parallel/antiparallel superfamilies . We then compared the ages of the superfamilies annotated with these terms . The results of this comparison are shown in Figure S4 . We found that the structural partition resulted in a much more dramatic age difference than the functional groupings . In particular , the functional annotations failed to divide the space efficiently , with many superfamilies annotated with both ‘parallel’ terms and ‘antiparallel’ terms . Even when considering superfamilies unique to a directional functional annotation , there was a less marked distinction than seen in superfamilies distinguished by structural features alone . Not only can these ages be related to general properties of proteins but they also provide a framework for examining more specific questions . For example , we present here a case study for analysing the evolutionary dynamics of certain structural motifs common in domains in a number of different folds . As was discussed earlier , antiparallel -sheet structures appear to be significantly younger than parallel sheets . Antiparallel topologies are , however , more common and more varied than parallel motifs . The most common topology in antiparallel sheets is the hairpin meander where neighbouring strands in a sheet are consecutive in the amino acid sequence . Apart from the simple meander the next two most common topological motifs are the greek key and the jelly roll . Around of all- folds in SCOP are annotated as containing either a greek key or a jelly roll and these motifs form a considerable role in their classification . Proteins containing these motifs rarely share either sequence similarity or a common function [30] . The topological architecture of these two common motifs is very similar , with the jelly roll containing a greek key at its core . While some papers treat the jelly roll motif as a special case of the greek key [31] , others argue that they occupy a unique portion of fold space [32] . In this study the age distributions of superfamilies classified as containing a greek key or a jelly roll were compared . Greek keys were significantly older than jelly rolls ( , Figure 3 ) . Moreover , we could find no other disparity ( for example , in the lengths of these populations ) that helped explain this difference .
In this work we estimate the evolutionary age of structural superfamilies . Our results are highly robust to different evolutionary assumptions in estimating ages , as well as alternative topologies and a smaller number of species in the phylogenetic trees . The results presented here indicate that newly evolving superfamilies tend to be , in general , shorter and structurally more simple than ancient structures . They appear , on average , to have a less hydrophobic core and a greater surface area to volume ratio . They differ from ancient superfamilies in terms of their amino acid composition , containing more polar residues , and tend to contain more additional stabilising features such as disulphide bonds and aromatic residues . Ancient superfamilies on the other hand are dominated by superfamilies and are enriched for many fundamental cellular functions . In particular , the still extant LUCA folds contain a comprehensive repertoire of proteins relating to RNA synthesis and maintenance rather than those used in DNA synthesis , and thus LUCA may have contained a ribosome mechanism for protein synthesis . The age of a superfamily could also be described as the depth at which it can be traced back through evolution . As such , there are several interpretations of our results , in particular in the case of what we have termed new-born superfamilies . Firstly , it could be that an entirely new domain was formed at some point in evolution . This could indicate that the evolution of a new superfamily as a transition from an already existing structure is a rare event , or that evolutionary transitions through fold space , when they occur , are more often reductive . It could also suggest that , through evolutionary drift , there is a tendency towards an increasingly elaborate structure . Secondly , a superfamily with a low age estimate might have originated earlier in evolution but the family recognition profiles have failed to identify homologues in distantly related species . In this case , such a superfamily may lack a representative deposition of solved structures , or be rapidly evolving and highly divergent . Certainly , characteristics such as a high solvent accessibility are correlated with the rate of sequence evolution [33] . Nevertheless , by using multiple profiles to build their Hidden Markov Models , SUPERFAMILY improves detection of sequence-divergent families compared to pairwise comparison and single profile searches [34] . As a greater coverage of proteins in such superfamilies are solved structurally , the likelihood of an incorrect low age estimate will decrease . Thirdly , a young superfamily may be the result of an unfound evolutionary link between superfamilies . As such the structural ancestor of these superfamilies may be earlier than their given age estimates . In order to address this possibility we have shown that the preferences are preserved at both the superfamily and fold level of the SCOP hierarchy . Finally , what appears to be a young superfamily may actually be ancient but has been lost at several more internal nodes than a parsimonious scenario suggests . This could be the result of functional specialisation within a superfamily . At present our understanding of the evolutionary history of individual superfamilies is not advanced enough to alter the evolutionary model behind age estimation for each superfamily . Our work concerning the robustness of the dataset overall to differing gain weights suggest that our results will be upheld within a moderate level of variation between different superfamilies . In this study we consider the structural universe of proteins and show that the age preferences of structural characteristics are not a residual effect derived from functional preferences . This result alone justifies the use of protein structures as a fundamental evolutionary unit . Using our age estimates we examined the specific case of greek key and jelly roll motifs , and identified a significant difference between their ages of origin . Given their similarity in topology it is possible that some superfamilies containing these motifs were involved in evolutionary transitions , where a greek key acted as a scaffold during the innovation of a jelly roll topology . This example demonstrates that these ages can be used to examine specific properties or motifs of interest , as well as explore more general fold space preferences for proteins at different stages in their evolution .
Occurrence profiles of superfamilies across whole genome trees were analysed using the principles of parsimony to estimate when their structural ancestor first evolved . The method described here is based on the the formulation developed by Winstanley et al . [13] . In subsequent sections we outline the process as it is used in this work . The data we use in this study were taken primarily from the SUPERFAMILY ( v1 . 75 ) database . SUPERFAMILY uses families of HMMs to identify homologues of 2 , 019 SCOP superfamilies . The database comprises protein sequences taken from completely sequenced and annotated genomes and assignments of these sequences to SCOP superfamilies . We downloaded predicted superfamilies for all 1 , 496 species available in the SUPERFAMILY database on September 11th 2012 . This set was then filtered as follows: This left 649 Bacteria , 265 Eukaryotes and 100 Archaea . We called this set the ALLgenomes and it was intended to represent the diversity in the currently known tree of life as accurately as possible . A second set ( MULTIgenomes ) was created that contained 211 multi-cellular Eukaryotes , a subset of ALLgenomes . The list of all these species including those removed from the original data are included in Table S2 . These predictions of superfamilies on genome sequences were then collapsed to a binary occurrence matrix where each element represents the presence or absence of a superfamily on a genome . A similar occurrence matrix was constructed at the fold level of the SCOP hierarchy . Multiple species trees were considered as the underlying phylogeny for the completely sequenced genomes . Using numerous trees helps to ensure that the results presented here are robust to inaccuracies in estimating the tree of life . We considered both the NCBI common taxonomy tree [37] , [38] as well as phylogenies constructed using the superfamily and fold occurrence profiles calculated above . For completeness the constructed trees were estimated using both parsimony and distance-based algorithms . All the trees were inferred using the PHYLIP package [39] . A total of 8 different trees were constructed for each of the genome sets ( ALLgenomes , MULTIgenomes ) . For each tree , the age of a superfamily is the result of a parsimony analysis on potential gain and loss events of the superfamily . Structural properties of 1 , 279 superfamilies were obtained using domains from the ASTRAL ( 1 . 75 ) database [22] with an aerospaci score and filtered to sequence identity . This set of 5 , 493 domains will be referred to as the ASTRAL40 set . The number of representative ASTRAL40 domains for each superfamily is included in Table S3 . Comparisons between the properties of new-born and ancient superfamilies were carried out using the Mann-Whitney U test [40] . Since multiple superfamilies shared the same age and therefore tied in rank the standard deviation of the distribution for the test statistic was appropriately adjusted [41] . While age distributions from all trees were considered in the analysis , for simplicity the p-values reported in the Results section derive from the ages calculated by maximum parsimony on the NCBI tree with branch lengths added using superfamily annotations . However , the results are only reported as significant if they gave significant p-values on ages from all the trees .
|
Proteins are the molecular workers of the cell . They are formed from a string of amino acids which folds into an elaborate three-dimensional structure . While there is a relationship between a protein's sequence and its structure this relationship is highly complex and not fully understood . Protein structures tend to evolve differently to their sequences . They are far more conserved so tend to change slower . The aim of this paper was to identify trends in the way that protein structures evolve , rather than adapting models of sequence evolution . To do this we have provided a database of ages for structural superfamilies . These ages are robust to drastic differences in the evolutionary assumptions underlying their estimation and can be used to study differences between populations of proteins . For example , we have compared newly evolved structures against those with a long evolutionary history and found that , overall , a shorter evolutionary history corresponds to a less elaborate structure . We have also demonstrated here how these ages can be used to compare particular structural motifs present in a large number of protein structures and have shown that the jelly roll motif is significantly younger than the greek key .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2013
|
Exploring Fold Space Preferences of New-born and Ancient Protein Superfamilies
|
The T cell immunoglobulin mucin 3 ( Tim-3 ) receptor is highly expressed on HIV-1-specific T cells , rendering them partially “exhausted” and unable to contribute to the effective immune mediated control of viral replication . To elucidate novel mechanisms contributing to the HTLV-1 neurological complex and its classic neurological presentation called HAM/TSP ( HTLV-1 associated myelopathy/tropical spastic paraparesis ) , we investigated the expression of the Tim-3 receptor on CD8+ T cells from a cohort of HTLV-1 seropositive asymptomatic and symptomatic patients . Patients diagnosed with HAM/TSP down-regulated Tim-3 expression on both CD8+ and CD4+ T cells compared to asymptomatic patients and HTLV-1 seronegative controls . HTLV-1 Tax-specific , HLA-A*02 restricted CD8+ T cells among HAM/TSP individuals expressed markedly lower levels of Tim-3 . We observed Tax expressing cells in both Tim-3+ and Tim-3− fractions . Taken together , these data indicate that there is a systematic downregulation of Tim-3 levels on T cells in HTLV-1 infection , sustaining a profoundly highly active population of potentially pathogenic T cells that may allow for the development of HTLV-1 complications .
The vast majority of HTLV-1-infected individuals with low and stable HTLV-1 proviral load levels are clinically asymptomatic for life [1] . However , 1–3% of subjects develop progressive neurological complications related to HTLV-1 infection , classically denominated as HTLV-1 associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) [2] , [3] , [4] . The infection can also lead to a debilitating malignancy , known as HTLV-1 associated adult T cell leukemia ( ATL ) in approximately 2–5% of infected individuals [4] , [5] , [6] , [7] . The immune response , and in particular the cellular immune response , plays an important role in the control of HTLV-1 infection [8] , [9] , [10] , [11] , [12] . In vitro studies further demonstrate that CD8+ T cell responses are able to directly lyse HTLV-1-infected CD4+ T cells [9] , [11] , [13] . In patients with HAM/TSP , CD8+ T cells are capable of producing multi-cytokine responses and are able to release cytotoxic molecules [14] , [15] . Recent studies have selected out patients with HLA-A*02 and HLA-Cw08 genes as being associated with lower HTLV-1 proviral load and a reduced risk of progression to HAM/TSP [16] , [17] . While these data support an important protective role for the CD8+ T cell immune response with the potential for viral control , other studies suggest that HTLV-1-specific CD8+ T cells may paradoxically contribute to the neuromuscular immunopathology through autoimmune mechanisms , leading to the clinical manifestation of HAM/TSP [18] . Furthermore , patients with HAM/TSP also present with high numbers of HTLV-1 Tax-specific CD8+ T cells in the cerebrospinal fluid [15] , [19] , [20] , [21] , [22] that are thought to play a immunopathogenic role , either by release of neurotoxic cytokines , such as TNF-α and IFN-γ [23] , [24] , or by direct cytotoxicity . It is evident from these studies that the precise role of CD8+ T cells in the control or pathogenesis of HTLV-1 disease progression remain unclear . Further knowledge of the mechanisms leading to T cell induced immunopathology in HTLV-1 infection will be important in determining successful immune-based therapies and provide insights for effective vaccine designs . During chronic viral infections , virus-specific CD8+ T cells undergo an altered pattern of differentiation and can become exhausted [25] , [26] . CD8+ T cell exhaustion is a transcriptionally altered state of T cell differentiation distinct from functional effector or memory CD8+ T cells [27] . CD8+ T cell exhaustion leads to profound T cell dysfunction and the inability of the T cells to control retroviral replication [28] , [29] , [30] . Conversely , downregulation of exhaustion markers could lead to a highly functional population of T cells . T cell immunoglobulin and mucin domain-containing protein 3 ( Tim-3 ) , is upregulated on CD8+ T cells during chronic viral infections [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] . Programmed death receptor-1 ( PD-1 ) is also known as another immune exhaustion biomarker expressed in chronic viral infections [28] , [37] , [38] , [39] , [40] , [41] , [42] , [43] . High levels of PD-1 and Tim-3 on virus-specific T cells have been shown to lead to poor proliferative capacity and , in some cases , ineffective Th1 cytokine production [29] , [39] , [44] . A sustained downregulation of these receptors would lead to an exacerbated constitutively active T cell population . The phenotypic profile of immune exhaustion markers on T cells is unknown in seropositive HTLV-1 individuals . In this study , we show for the first time that HTLV-1 associated complications may be related to the highly responsive inflammatory Tax-specific T cells in HTLV-1-infected individuals . These results support the idea that HTLV-1 infection induces mechanisms resulting in a limited T cell exhaustion profile , leading potentially to neuro-immunopathology and disease complications .
The research involving human participants reported in this study was approved by the institutional review board of the University of Sao Paulo ( IRB #0855/08 ) Sao Paulo , Brazil . Informed consent was obtained for all subjects . All clinical investigation were conducted according to the principles expressed in the Declaration of Helsinki ( http://www . wma . net/en/30publications/10policies/b3/index . html ) . Patients were serially recruited in the HTLV-1 Outpatient Clinic at the University of Sao Paulo , Brazil in two stages with written informed consent approved by the University of Sao Paulo's Institutional Review Board ( #0855/08 ) . The diagnosis of HAM/TSP based on criteria outlined by the WHO [45] ( Table 1 ) . The majority of the patients were female ( 63% ) with a median age of 48 ( IQR: 22–66 ) years . We enrolled age and sex matched healthy uninfected volunteers without clinical and laboratory evidence of HTLV-1-associated disease , from the same demographics as the infected subjects . All HTLV -1 seropositive subjects tested negative for Hepatitis B , Hepatitis C , and HIV infections . No other inflammatory diseases or disorders were present in any of the participants . Blood samples were processed with Ficoll-Paque PLUS ( Amersham Pharmacia Biotech , Uppsala , Sweden ) gradient centrifugation , and peripheral-blood mononuclear cells ( PBMC ) were isolated and cyropreserved in fetal bovine serum ( FBS ) containing 10% DMSO in liquid nitrogen . Conjugated Pentamers were obtained commercially from Proimmune ( Oxford , UK ) . The HLA-A*02 restricted HTLV-1 Tax ( LLFGYPVYV ) and CMV ( NLVPMVATV ) peptides were obtained from New England Peptide ( Gardner , MA ) . In some experiments rIL-2 [80 IU/ml] ( Roche Diagnostics , Mannheim , Germany ) and rIL-15 [50 ng/ml] ( R&D Systems , Minneapolis , MN ) were used during in vitro culture studies . Cryopreserved PBMC were rapidly thawed in warm RPMI 1640 with 10% FBS , washed in FACS buffer ( PBS , with 0 . 5% bovine serum albumin , 2 mM EDTA ( Sigma-Aldrich , St . Louis , MO ) ) . For staining , 5×105 cells were incubated with conjugated antibodies against Tim-3 ( R&D Systems , Minneapolis , MN ) , PD-1 ( Biolegend , San Diego , CA ) , CD4 , CD8 , CD3 ( all from BD Biosciences , San Jose , CA ) for 30 min on ice . In some experiments , PMBC were then fixed and permeabilized prior to staining with conjugated anti-Tax ( clone Lt-4 ) antibodies [46] or a control labeled IgG . Fluorescence minus one ( FMO ) samples were prepared for each fluorochrome to facilitate gating as well as conjugated isotype control antibodies . Anti-mouse IgG-coated beads were stained with each fluorochrome separately and used for software-based compensation . Analysis was performed using a FACSCanto instrument ( BD Biosciences ) and at least 100 , 000 events were collected and analyzed with FlowJo software ( TreeStar , Ashland , OR ) . To define pentamer positive cells: staining was initially performed immediately after thawing with biotin-labeled HLA-A2 Tax or CMV epitope specific pentamer fluorotags followed a secondary staining step with fluorophore conjugated antibodies against CD8 ( BD ) , Tim-3 ( R&D Systems ) , PD-1 ( Biolegend ) and CD3 ( BD ) , and with labeled streptavidin . Cells were washed twice with PBS containing 1% FBS , then fixed in 2% paraformaldehyde and run on a customized BD FACSCanto within 12 hours . HTLV-1 proviral DNA was extracted from PBMC using a commercial kit ( Qiagen GmbH , Hilden Germany ) and according to the manufacturer's instructions . The extracted DNA was used as a template to amplify a fragment of 158 bp from the viral tax region using previously published primers[47] . The SYBR green real-time PCR assay was carried out in 25 µl PCR mixture containing 10× Tris ( pH 8 . 3; Invitrogen , Brazil ) , 1 . 5 mM MgCl2 , 0 . 2 µM of each primer , 0 . 2 mM of each dNTPs , SYBR Green ( 18 . 75 Units/r×n; Cambrex Bio Science , Rockland , ME ) and 1 unit of platinum Taq polymerase ( Invitrogen , Brazil ) . The amplification was performed in the Bio-Rad iCycler iQ system using an initial denaturation step at 95°C for 2 minutes , followed by 50 cycles of 95°C for 30 seconds , 57°C for 30 seconds and 72°C for 30 seconds . The human housekeeping β globin gene primers GH20 and PC04[48] were used as an internal control calibrator . For each run , standard curves for the value of HTLV-1 tax were generated from MT-2 cells of log10 dilutions ( from 105 to 100 copies ) . The threshold cycle for each clinical sample was calculated by defining the point at which the fluorescence exceeded a threshold limit . Each sample was assayed in duplicate and the mean of the two values was considered as the copy number of the sample . The amount of HTLV-1 proviral load was calculated as follows: copy number of HTLV-1 ( tax ) per 1 , 000 cells = ( copy number of HTLV-1 tax ) / ( copy number of β globin/2 ) ×1 , 000 cells . The method could detect 1 copy per 103 PBMC . MAIPS4510 Elispot plates ( Millipore , Danvers , MA ) were coated with anti-IFN-γ ( 10 µg/ml ) ( Mabtech , Nacka Strand , Sweden ) in PBS , 50 µl/well , either overnight at 4°C or for one hour at room temperature . After three washes with PBS , PBMC ( 1×105 cells/well ) and the appropriate antigens were added ( Tax peptide and CMV peptide ) , with a final volume of 200 µl/well . Plates were incubated at 37°C in 5% CO2 for 16–20 hours . After washing with phosphate-buffered saline ( PBS ) plus 0 . 1% Tween 20 ( PBST ) , biotinylated anti-IFN-γ 1 µg/ml ) ( Mabtech ) , antibodies were added to the appropriate wells in PBS 0 . 1% tween 1% BSA ( PBSTB ) for 30 minutes at room temperature . Plates were washed again three times with PBST , and alkaline phosphatase-conjugated streptavidin ( Jackson Immunoresearch , West Grove , PA ) was added ( 50 µl of 1∶1 , 000 dilution in PBSTB ) and incubated for 30 min at room temperature . Plates were washed in PBSTB , soaked for 1 hour in PBSTB and incubated with blue substrate ( Vector Labs , Burlingame , CA ) until spots were clearly visible , then rinsed with tap water . When plates were dry , spots were counted using an automated ELISPOT reader . Statistical analysis was performed by using GraphPad Prism statistical software ( GraphPad Software , San Diego , CA ) . Non-parametric statistical tests were used . The Mann-Whitney U was used for comparison tests and the Spearman rank test were used for correlation analyses .
Peripheral venous blood was drawn from 22 HTLV-1 seropositive patients and 7 HTLV-1 seronegative matched donors , all screened for the presence of HLA-A*02 alleles , and peripheral blood mononuclear cells ( PBMC ) were extracted and cryopreserved . Tim-3 and PD-1 are two cellular molecules expressed on T cells implicated in immune exhaustion . We evaluated the expression and co-expression of Tim-3 and PD-1 on T cells derived from HTLV-1 seropositive ( both asymptomatic carriers and patients with the diagnosis of HAM/TSP ) and seronegative controls to determine whether they were modulated in HTLV-1 infection . We observed a significant decrease in the frequency of Tim-3+ PD-1− expressing CD8+ and CD4+ T cells among HTLV-1 seropositive subjects ( CD8+: median 8 . 01% , IQR 5 . 42–10 . 50; CD4+: median 4 . 3% , IQR 3 . 50–5 . 99 ) compared to HTLV-1 seronegative controls ( CD8+ median 15 . 10% , IQR 10 . 50–17 . 60; CD4+: median 6 . 84% , IQR 5 . 74–7 . 85 ) ( Figure 1A and B ) . Patients with HAM/TSP ( red circles ) had significantly lower levels of Tim-3+ PD-1− expressing CD8+ ( p = 0 . 002 ) and CD4+ ( p = 0 . 004 ) T cells compared to healthy uninfected controls ( open circles ) . In contrast , the frequency of Tim-3− PD1+ T cells trended to an increase in subjects with HTLV-1 infection ( CD8+: median 18 . 80% , IQR 10 . 42–24 . 90; CD4+: median 20 . 70% , IQR 13 . 6–25 . 35 ) compared to healthy uninfected controls ( CD8+: median 9 . 22% , IQR 8 . 97–15 . 50; CD4+: median 13 . 60% , IQR 12 . 7–18 . 6 ) ( Figure 1A and B ) . Only a few T cells co-expressed both Tim-3 and PD-1 , and no differences were observed between uninfected subjects and those with HTLV-1 asymptomatic infection or HAM/TSP patients . Using linear regression analysis we observed no association between the frequency of Tim-3 or PD-1 expression on CD8+ T cells in HTLV-1 infected subjects and proviral load . ( p = 0 . 68; r = 0 . 1043; or p = 0 . 89; r = −0 . 03202 , respectively ) . HLA- A*02 positive HTLV-1-infected patients have high amounts of circulating CD8+ T cells specific for an immunodominant HLA- A*02 -restricted epitope , HTLV-1 Tax 11–19 [20] , [49] , [50] . In HAM/TSP patients , these HTLV-1's Tax-specific CD8+ T cells correlate with HTLV-1 proviral load [23] . Among this cohort , we identified 15 HLA-A2 positive subjects ( asymptomatic carriers , n = 9 and HAM/TSP , n = 6; Table 1 ) , and evaluated the Tim-3 and PD-1 receptor expression on Tax-specific CD8+ T cells . Eight patients had Tax-specific CD8+ T cells ( median 2 . 45% , IQR 1 . 11–5 . 31 ) as determined by specific pentamers . Among these patients we also observed HLA-A*02 -restricted CMVpp65 CD8+ T cells ( median 2 . 49% , IQR 1 . 87–11 . 37 ) . Interestingly , Tim-3 levels were dramatically reduced on CD8+ Tax 11–19-specific T cells ( median 24 . 77% , IQR 15 . 2–39 . 54 ) compared to the expression of PD-1 ( median 48 . 06% , IQR 36 . 81–65 ) ( Figure 2A , C and E ) . We also evaluated Tim-3 expression on HLA- A*02 CMV specific T cells and found a similar pattern of expression with Tim-3 levels reduced on CD8+ CMV-specific T cells ( median 27 . 62% , IQR 21 . 48–43 . 19 ) compared to PD-1 ( median 47 . 70% , IQR 40 . 45–51 . 16 ) ( Figure 2B , D and F ) . To determine whether there was an association with Tim-3 or PD-1 levels on Tax 11–19-specific CD8+ T cells and their functionality , we evaluated the production of IFN-γ in response to the HLA-A*02-restricted Tax 11–19 immuno-dominant epitope and in comparison , the CMVpp65 epitope by an ELISPOT assay derived from PBMCs derived from 8 HLA- A*02 restricted infected individuals with Tax 11–19- and CMVpp65 specific CD8+ T cells ( Figure 3 ) . We saw no correlation between IFN-γ secretion and global PD-1 or Tim-3 expression on either the CD4+ or CD8+ T cells , irrespective of disease status ( data not shown ) . The frequency of PD-1 expression on Tax-specific or CMV-specific CD8+ T cells also did not associate with the amount of IFN-γ secreted ( r = 0 . 1317; P = 0 . 7520 and r = 0 . 2245; P = 0 . 594 , respectively ) ( Figure 3B ) . However , we observed a statistically significant inverse correlation between the frequency of Tim-3 on both Tax-specific as well as CMV-specific CD8+ T cells and the amount of IFN-γ secreted ( r = −0 . 8982; P = 0 . 0046; r = 0 . 9710; P = 0 . 0028; Figure 3A ) . Tax expression marks HTLV-1 viral replication in both CD4+ and CD8+ infected T cells . We aimed to determine whether the downregulation of Tim-3 we had observed was occurring only among infected cells , or in bystander cells as well . We therefore co-stained for Tax and Tim-3 expression on T cells from HTLV-1 infected subjects . We also stained for PD-1 expression as a control . The culture of PBMC overnight did not alter Tim-3 or PD-1 expression levels on the HTLV-1-infected T cells ( data not shown ) . We observed that Tax was expressed on PBMC from some subjects following 24 hours of culture and was detected on both Tim-3+ as well as Tim-3− CD4+ T cells ( Figure 4A ) . Similarly , Tax was present on both PD-1+ and PD-1− T cells . We further identified a unique subset of Tax expressing CD4+ T cells that were Tim-3hi and lacked PD-1 in most of the subjects expressing Tax ( Fig . 4A ) . No difference in the pattern of co-expression between HTLV-1 seropositive asymptomatic patients and those diagnosed with HAM/TSP was observed . An increase in Tim-3 levels on T cells would potentially lead to a downregulation of T cell functionality . We therefore tested several gamma-chain associated cytokine mediators that could potentially modulate Tim-3 expression . We observed that IL-2 , and especially IL-15 , led to a prominent increase in the frequency of Tim-3 levels , specifically on the CD8+ T cell population after only 12 hours in culture ( Figure 4B , C ) . No change in the levels of PD-1 expression were observed on both CD8+ and CD4+ T cells ( Figure 4B , C ) .
CD8+ T cell dysfunction and/or exhaustion are common features of many chronic viral infections , including HIV-1 and HCV infections [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] . The mechanisms of T cell dysfunction are complex , but are in part mediated by a distinct set of inhibitory receptors [27] , [51] . A high , and sustained , expression of Tim-3 and PD-1 , have emerged as hallmarks of T cell exhaustion in human viral infections , and blockade of these pathways can reinvigorate immune responses during persisting viral infections [29] , [30] , [33] , [34] , [36] . In this study , we report that CD8+ and CD4+ T cells in HTLV-1 infection express lower levels of Tim-3 , and this was more pronounced in patients with HAM/TSP . Phenotypically , we observed that Tax HTLV-1-specific , HLA-A*02 -restricted CD8+ T cells consistently retain a lower frequency of Tim-3 . We propose that this low expression of Tim-3 on HTLV-1 Tax-specific T cells may lead to a persistent and deleterious effector T cell pool leading to more inflammation . The pattern of expression of PD-1 in HTLV-1 infection has recently been shown to be elevated on T cells in HTLV-1 carriers and also on CMV and EBV specific T cells in asymptomatic carriers compared to healthy controls [52] . This opposing relationship of PD-1 and Tim-3 expression on T cells in patients with HTLV-1 infection suggests that the downregulation of Tim-3 expression potentially leads to more vigorous T cell activity in the HTLV-1-infected individual , whereas PD-1 may not fully reflect T cell dysfunction , but rather an activated status of the T cell response to infection . Indeed the association between the frequency of Tim-3 and PD-1 levels with IFN-γ secretion in response to either Tax or CMVpp65 epitopes show remarkably different correlations . In a study by Petrovas and colleagues , it was apparent that PD-1 expressing T cells are able to secrete cytokines in response to viral peptides [39] . Our data suggests that PD-1 and Tim-3 on antigen specific CD8+ T cells are functionally different , and this may reflect a distinct stage of differentiation . PD-1 appears to mark early T-cell activation and exhaustion , while Tim-3 represents a more terminal stage of impairment . The positive association between the frequency of HTLV-1's Tax-specific CD8+ T cells and HTLV-1's Tax mRNA load and proviral load is well documented [8] , [53] , [54] . Studies evaluating the phenotype of CD8+ T cells in HTLV-1 infection have been largely limited to characterizing the expression of T cell maturation and differentiation markers ( CD28 , CD45RO ) [14] . Our data suggest that downregulation of Tim-3 , rather than PD-1 , marks global and Tax-specific CD8+ T cells , which are hyperfunctional . This contrasts with HIV-1 and HCV infections , where the expression of Tim-3 is increased , leading to a population of CD8+ T cells that are rendered dysfunctional both in terms of proliferative capacity and cytokine release as well as release of cytolytic granules [29] , [36] . Surface receptors known to regulate T cell function like CD244 and PD-1 have been shown to be upregulated either directly due to Tax or indirectly due to the cytokine milieu [52] , [55] . We postulate that either direct HTLV-1 viral components led to a downregulation of Tim-3 , or as yet to be defined cytokine ( s ) , suppress Tim-3 expression . In several human and murine studies , the manifestation of autoimmune diseases such as multiple sclerosis , have been attributed as a result of downregulated Tim-3 expression on T cells [56] . It still remains unclear how HTLV-1 infection sustains low levels of Tim-3 on T cells in infected patients and whether this is a cause or a consequence of disease progression . Multilayered mechanisms for this regulation may be occurring in the context of HTLV-1 infection . One strategy to reduce the T cells response would be through enhancement of the Tim-3 receptor for engagement with its cognate ligand . This could serve as a novel strategy to dampen the inflammatory inducing T cells . From our results , PD-1 engagement may not be as effective since both PD-1− and PD-1+ cells retain the potential for CD8+ T cell lytic function . A novel strategy to reverse or prevent the onset of neurological complications would be through dampening effector T cell functions . From our results , it appears the γ-chain cytokines elicited higher levels of Tim-3 on specifically on CD8+ T cells , and such a strategy could be harnessed to dampen T cell function in the HTLV-1 infected individual . Further work to understand the mechanisms for HTLV-1 disease progression and devise strategies to effectively prevent neurological complications will be needed . Targeted modulation of the Tim-3 pathway provides a viable model for this intervention .
|
The retrovirus , Human T lymphotropic virus type 1 ( HTLV-1 ) infects 10–20 million people worldwide . The majority of infected individuals are asymptomatic; however , approximately 3% develop the debilitating neurological disease , HTLV-1-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) . There is also currently no cure , vaccine or effective therapy for HTLV-1 infection . The precise role of CD8+ killer T cells in the control or contribution of HTLV-1 disease progression remains unclear . The T-cell immunoglobulin mucin domain-containing ( Tim ) proteins are type 1 transmembrane proteins . Three human Tim proteins ( Tim-1 , -3 , and -4 ) exist and display markedly diverse expression patterns and functions . Tim-3 is upregulated on CD8+ T cells during chronic viral infections leading to a population of poorly functioning T cells . We investigated the expression of Tim-3 on T cells from patients with asymptomatic and symptomatic HTLV-1 infection and compared this with HTLV-1 uninfected donors . Patients diagnosed with HAM/TSP down-regulated Tim-3 expression on T cells when compared to asymptomatic patients and uninfected controls . Our study provides evidence of a novel mechanism for the persistent inflammation observed in HTLV-1 infected patients with neurological deficits and significantly advances our understanding of how the Tim-3 pathway functions .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"immune",
"cells",
"clinical",
"immunology",
"immunity",
"t",
"cells",
"immunity",
"to",
"infections",
"immunology",
"viral",
"diseases",
"immune",
"response"
] |
2011
|
HTLV-1 Tax Specific CD8+ T Cells Express Low Levels of Tim-3 in
HTLV-1 Infection: Implications for Progression to Neurological
Complications
|
Mutations in the human survival motor neuron 1 ( SMN ) gene are the primary cause of spinal muscular atrophy ( SMA ) , a devastating neuromuscular disorder . SMN protein has a well-characterized role in the biogenesis of small nuclear ribonucleoproteins ( snRNPs ) , core components of the spliceosome . Additional tissue-specific and global functions have been ascribed to SMN; however , their relevance to SMA pathology is poorly understood and controversial . Using Drosophila as a model system , we created an allelic series of twelve Smn missense mutations , originally identified in human SMA patients . We show that animals expressing these SMA-causing mutations display a broad range of phenotypic severities , similar to the human disease . Furthermore , specific interactions with other proteins known to be important for SMN's role in RNP assembly are conserved . Intragenic complementation analyses revealed that the three most severe mutations , all of which map to the YG box self-oligomerization domain of SMN , display a stronger phenotype than the null allele and behave in a dominant fashion . In support of this finding , the severe YG box mutants are defective in self-interaction assays , yet maintain their ability to heterodimerize with wild-type SMN . When expressed at high levels , wild-type SMN is able to suppress the activity of the mutant protein . These results suggest that certain SMN mutants can sequester the wild-type protein into inactive complexes . Molecular modeling of the SMN YG box dimer provides a structural basis for this dominant phenotype . These data demonstrate that important structural and functional features of the SMN YG box are conserved between vertebrates and invertebrates , emphasizing the importance of self-interaction to the proper functioning of SMN .
Proximal spinal muscular atrophy ( SMA ) is a common neuromuscular disorder , recognized as the most prevalent genetic cause of early childhood mortality [1] . SMA is characterized by degeneration of motor neurons in the anterior horn of the lower spinal cord , and progressive symmetrical paralysis . Coupled with this loss of motor function , SMA patients display severe atrophy of the proximal muscles . The onset of symptoms and their severity can vary , leading to an historical classification of SMA into three distinct subtypes [1] . More recently , clinicians have recognized that SMA is better characterized as a continuous spectrum disorder , ranging from severe ( prenatal onset ) to nearly asymptomatic [2] . Almost two decades ago , mutations in the survival motor neuron 1 ( SMN1 ) gene were shown to be causative for SMA [3] . The disease typically results from homozygous deletion of SMN1; however , a small fraction of SMA patients have lost one copy of SMN1 and the remaining copy contains a point mutation [4] . The best characterized function for the ubiquitously expressed SMN protein is in the biogenesis of Sm-class small nuclear ribonucleoproteins ( snRNPs ) , core factors of the spliceosome [5] , [6] . In addition , SMN has been implicated in numerous other cellular activities , including axonal transport , neuronal pathfinding , formation and function of neuromuscular junctions , myoblast fusion and maintenance of muscle architecture [4] , [7]–[11] . Despite this multitude of putative functions attributed to SMN , or perhaps because of it , the precise pathophysiological mechanisms that give rise to SMA are the subject of considerable debate . Seemingly straightforward questions of disease pathogenesis have yet to be definitively answered . For example , is SMA caused by a cell-autonomous reduction of SMN protein levels in motor neurons [12]–[16] or is it a more systemic defect involving other cell types [17]–[24] ? Irrespective of the question of cellular autonomy , the molecular etiology of SMA also remains unclear . Is the neuromuscular dysfunction seen in SMA patients caused by a loss of Sm-class spliceosomal snRNPs , ultimately leading to defects in pre-mRNA splicing ? Or is it due to some non-canonical function of SMN and/or snRNPs ? Experiments using animal models of severe SMA suggest that the pre-mRNA splicing defects observed in late-stage SMA animals are indeed tissue-specific [25] , [26] . However , such deficits are only detectable later in the disease course , after the onset of neuromuscular dysfunction [27]–[29] . Complicating matters , SMN-deficient animals are developmentally delayed or arrested [22] , [29]–[34] , making the selection of properly staged control animals critical to the comparison of phenotypes . Moreover , alternative splicing of the SMN gene duplicate ( SMN2 ) in humans or mouse models is another variable , creating a feedback loop [35] , [36] that can negatively regulate SMN expression . In the fruitfly , the vast majority of Smn pre-mRNA transcripts are intronless ( Flybase; [37] ) . Therefore , we set out to create an allelic series of Drosophila SMA models wherein we could specifically focus on SMN protein function , in the absence of other complicating factors .
To identify which functions of SMN are critical to the pathology of SMA , we aimed to create disease-relevant models that disrupt subsets of SMN interactions . Point mutations are useful in this context , as they can disrupt specific functions of multi-domain proteins , leaving other functions unaffected . To date , twenty-five different SMN1 point mutations have been identified in SMA patients [4] . Many of these mutations are located at residues that are conserved between humans and insects ( Fig . 1A ) . In this report , transgenic Drosophila bearing twelve of these SMA-causing point mutations were generated ( Fig . 1B ) . The salient features of the Smn transgenic cassette have been previously described [28] , including a 3X FLAG tag that was inserted immediately downstream of the ATG start codon . The transgenes were also expressed under control of the native Smn promoter and 3′ flanking sequences [28] and inserted at a landing site located within band 86Fb on chromosome 3R using the PhiC31 integrase system [38] . Note that all of the constructs ( including an SmnWT control ) were injected directly into embryos heterozygous for the SmnX7 microdeletion , a null mutation [39] that was recombined with the appropriate PhiC31 landing site prior to injection ( see Methods for details ) . Note that the SmnX7 deletion removes the promoter region and the entire Smn open reading frame , leaving behind only 44 bp of the 3′ UTR [39] . Thus , this mutant produces no Smn mRNA . As shown in Fig . 1 , one of the mutations lies in the region responsible for Gemin2 binding , five are located in the Tudor domain , and six in the YG box oligomerization domain of SMN , thus mirroring the distribution of point mutations identified in SMA patients . For phenotypic analysis of the point mutants in our model , we crossed the transgenic flies ( SmnX7 , Flag-SmnTg ) with an SmnX7 null allele to obtain flies that are homozygous null for endogenous Smn and hemizygous for the Flag-Smn transgene . We adopted this approach because human SMA patients that express SMN1 missense mutations are also typically hemizygous and because we observed an improvement in the viability of the transgenic mutants when crossed to SmnX7 , as compared to the self-cross . This latter finding indicates the presence of recessive alleles in the transgenic background that contribute to the phenotype of the homozygotes . When expressed in an SmnX7/X7 null background , the hemizygous Flag-SmnWT construct showed robust rescue of adult viability ( ∼67% , see Fig . 2A ) , consistent with our previous findings using an SmnX7/D null background [28] . The degree of rescue achieved by expressing each of the different SMA point mutations varied ( Fig . 2A ) . Three of the transgenes ( SmnM194R , SmnY203C and SmnG206S ) failed to rescue the larval lethality of the null animals ( Fig . 2A ) . In contrast , a second group of point mutants rescued adult viability to a degree roughly similar to that of the SmnWT transgene , including: SmnD20V , SmnF70S and SmnG73R and SmnI93F . A third group of mutants can be characterized as having intermediate phenotypes: SmnV72G , SmnY107C , SmnT205I , SmnG210C and SmnG210V . This last category can be categorized as pupal-lethal , as the majority of these animals die prior to eclosion . The SmnG210C mutation produced a much milder phenotype ( ∼45% rescue to adulthood ) compared to the SmnG210V mutation ( ∼5% rescue ) despite the fact that they are located at the same residue . Interestingly , SmnV72G was the only mutation that rescued the larval lethality of the null animals but then also resulted in complete pupal lethality ( Fig . 2A ) . Thus , when expressed in Drosophila , disease-causing Smn point mutations display a range in the age of symptomatic onset and in life-expectancy . As shown previously [28] , the amount of dSMN protein expressed from transgenes driven by the native Smn promoter and integrated at site 86Fb is several-fold lower than that of the endogenous gene ( Fig . 2B ) . However , we observed robust rescue of the null phenotype in the presence of the SmnWT transgene , as these animals are both viable and fertile . A majority of the SMA alleles expressed equivalent levels of transgenic protein when compared to SmnWT . The exceptions were SmnM194R , SmnY203C and SmnG206S , which showed a significant reduction in dSMN levels ( Fig . 2B ) . Because the constructs were all inserted into the identical genomic location , and this same level of expression was observed in multiple independent transformants ( Fig . S1 ) , the reduction is most likely due to instability of the mutant proteins rather than differences in mRNA transcription . Furthermore , the very low levels of dSMN protein in the SmnM194R , SmnY203C and SmnG206S animals are consistent with their severe , larval-lethal phenotypes . The ability of SMN to interact with itself [40] , [41] is important for its function in RNP assembly [42] , and mutations that disrupt SMN self-oligomerization are unstable in vivo [43] . We have shown previously that dSMN ( Y203C ) and dSMN ( G206S ) proteins are oligomerization defective [28] . We therefore tested dSMN ( M194R ) for its ability to self-interact . The mutant protein was N-terminally tagged with a 3X FLAG tag to distinguish it from the endogenous dSMN , and co-transfected into S2 cells ( a Drosophila embryonic cell line ) along with either a Myc-tagged dSMN ( M194R ) construct or a wild-type control . Immunoprecipitations were performed with anti-FLAG antibodies and the amount of co-precipitating protein was visualized by western blotting . As shown in Fig . 3A , dSMN ( M194R ) protein is also severely defective in its ability to self-oligomerize . Thus , the low levels of dSMN in SmnM194R , SmnY203C and SmnG206S animals can be explained by the inability of these proteins to form higher-order SMN complexes . These observations are consistent with findings that the equivalent mutations in human SMN ( SMNM263R , SMNY272C and SMNG275S ) are defective in self-oligomerization in vitro [40] , [44] , suggesting a high level of conservation between the structure-function relationships of Drosophila and human SMN . Among the proteins that comprise the SMN complex in humans , collectively known as ‘Gemins , ’ orthologues for Gemins2 , 3 , and 5 are the only genes that have been identified and validated in Drosophila . We and others [33] , [45] have thus far failed to detect a biochemically stable interaction between dSMN and dGemin5 ( dGem5 ) , although dGem5 reportedly co-localizes with dSMN in various subcellular organelles [46] . To further establish whether dSMN can recapitulate the biochemical characteristics of SMA patient mutations , and to understand how these characteristics correlate with animal phenotypes , we analyzed the mutant dSMN proteins for their ability to bind dGem2 and dGem3 using a co-transfection assay . The binding of dGem2 was reduced by a single Smn point mutation . As shown in Fig . 3B , dGem2 binding to dSMN ( D20V ) was reduced as compared to wild-type , but not eliminated . None of the other dSMN mutants showed any decrease in binding to dGem2 ( Fig . 3B ) . This was not a surprising finding , as residue D20 lies within the known Gemin2 binding domain of SMN ( Fig . 1A ) , and is consistent with experiments on its human counterpart , SMN ( D44V ) [47] , [48] . Interestingly , SmnD20V flies display a mild phenotype , suggesting that the observed residual interaction with dGem2 is sufficient for its function in vivo . Indeed , SMA patients with the D44V mutation have been diagnosed with the mild form of SMA , type III [49] . We also analyzed the interaction of various Flag-tagged dSMN mutants with Myc-dGem3 by co-immunoprecipitation with anti-Myc antibodies . As shown in Fig . 3C , we found that binding of dGem3 was severely disrupted by two point mutations , SmnY203C ( SMNY272C in humans ) and SmnG206S ( SMNG275S ) . Consistent with our findings , the human SMNY272C mutation was previously shown to disrupt Gemin3 binding in vitro [40] . Notably , we found that the SmnT205I ( SMNT274I ) mutation , which lies between residues Y203 and G206 , did not affect dSMN's ability to interact with dGem3 ( Fig . 3C ) . This finding is consistent with previous observations that SMNT274I is active in snRNP assembly , an activity that also requires Gemin3 [42] . Strikingly , the phenotype of the SmnT205I animals was much less severe than that of either the SmnY203C or SmnG206S mutants ( Fig . 2A ) . Given that dSMN ( T205I ) and dSMN ( G206S ) proteins are both moderately defective in self-oligomerization ( [28] and Fig . 3A ) , we attribute the more severe phenotype of the SmnG206S mutants to the inability of dSMN ( G206S ) to interact with dGem3 . This interpretation is bolstered by the fact that null mutations in Gemin3 were shown to destabilize dSMN in vivo [33] . In summary , these results further demonstrate that important biochemical properties of SMN are conserved between Drosophila and humans [24] , [28] , [45] . To further characterize the Smn point mutants , we crossed each of them to the wild-type ( WT ) rescue line , SmnWT . Since SMA patients display a recessive mode of inheritance , we expected that the SmnWT transgene would rescue organismal viability in the missense mutant backgrounds ( genotype: SmnX7/X7 , Flag-SmnWT/Mut ) to roughly the same extent as it did in the hemizygous rescue line ( SmnX7/X7 , Flag-SmnWT/− ) . However , for many mutations we observed an intermediate level of rescue , between that observed when expressing the wild-type transgene alone versus that of the mutant transgenes alone ( Fig . 4A ) . This dominant phenotype was most evident when the wild-type transgene was expressed in combination with the three most severe YG box alleles ( M194R , Y203C and G206S ) . For example SmnX7/X7 , Flag-SmnWT/Y203C animals displayed a much lower eclosion frequency ( Fig . 4A ) than did either of the two controls , SmnX7/X7 , Flag-SmnWT/− ( hemizygotes ) or SmnX7/X7 , Flag-SmnWT/WT ( homozygotes ) . Importantly , expression of the wild-type Smn transgene in combination with the M194R , Y203C or G206S point mutants rescued the larval lethality associated with expression these alleles on their own ( Fig . 4A ) . Thus , these dominant phenotypes are intermediate in their severity . We note that SmnWT/WT homozygotes ( obtained by an intercross of two different founder lines ) showed a somewhat lower eclosion frequency ( ∼50% ) than did the hemizygotes ( ∼70% ) . These findings suggest that second-site recessive alleles may contribute to the decreased fitness of the homozygotes and that outcrossing to a ‘clean’ SmnX7 chromosome ameliorates these effects . Moreover , the presence of second site recessives may also affect the results of the intragenic complementation analyses ( Fig . 4A ) , as the transgenes were all inserted into the same genetic background . However , when co-expressed with SmnWT , the phenotypic severities observed for the point mutant lines correlated with the degree of rescue when expressed alone . For example , SmnG210V had a less severe phenotype ( pupal lethal ) than SmnG206S ( larval lethal ) , when expressed alone . When co-expressed with the wild-type transgene , SmnG210V/WT flies also showed a milder phenotype than did SmnG206S/WT ( Fig . 4A ) . There was one exception to this rule; SmnV72G displayed an early pupal-lethal phenotype when expressed alone , but had an eclosion frequency comparable to that of the mild mutations when co-expressed with SmnWT . In summary , complementation crosses with the mutant and wild-type Smn transgene display intermediate phenotypes . Using SmnY203C and SmnG210V as test cases , we carried out additional complementation crosses to each of the other point mutant lines . As shown in Fig . 4B , co-expression of SmnY203C with the other transgenes had a negative effect on adult viability . When crossed to Y203C , alleles that showed strong rescue of the null phenotype when expressed alone ( e . g . WT , D20V , G73R , or I93F ) either completely failed to eclose as adults or did so at very low frequencies . However , this dominant effect was not fully penetrant , as a majority of the trans-heterozygous animals developed beyond larval stages and died as pupae ( Fig . 4B ) . The only exceptions were crosses between Y203C and the other two severe mutations ( M194R and G206S ) , which continued to die as larvae . SmnG210V co-expression with the other transgenes had a positive effect on pupation of flies expressing the severe mutations ( Fig . 4C ) . Overall , when crossed to G210V , adult viability decreased . Crosses between G210V and either V72G or G206S were the only exceptions , with a small fraction of flies eclosing . This is in contrast to the pupal lethality observed when V72G was expressed alone and the larval lethality of G206S . Thus the intermediate phenotypes observed in these complementation crosses provide a clear example of the genetic principle of incomplete dominance . The most severe Smn point mutations in our collection ( M194R , Y203C and G206S ) all map to the YG box self-oligomerization domain ( Fig . 1A ) , which is the most well-conserved region of SMN . Consistent with their YG box location , these three alleles are defective in self-interaction assays ( Fig . 3A ) and do not form stable SMN complexes in vivo ( Fig . 2B ) . By such criteria , these three alleles would normally be considered to be protein nulls , yet their phenotype is actually more severe than that of the null allele , SmnX7/X7 . Indeed , these larvae all died by ∼8–9 days after hatching , whereas ∼20% of the null mutants were still alive at this time point ( Fig . 5A ) . The M194R , Y203C and G206S larvae are similar in size to Smn null mutants and are noticeably smaller and less active their SmnWT counterparts ( Figs . 5B and 5C ) . These latter observations are consistent with previous analyses of Smn null mutants [19] , [31] , [33] . The long-lived larval phenotype displayed by the Smn null animals [28] , [33] , but not by the severe point mutants ( Fig . 5A ) , suggests that zygotic expression of the oligomerization-defective point mutations inhibits the function of the endogenous Gemins and/or the maternally contributed ( wild-type ) dSMN . Despite the fact that the severe YG box mutants are unstable and fail to self-interact , we reasoned that the dominant negative phenotype seen upon co-expression of wild-type and mutant dSMN suggests that these two proteins might form heterodimers . We tested this idea by co-transfecting differentially tagged wild-type and mutant dSMN constructs and measuring the amount of co-precipitated protein . We assayed Y203C , G206S , G210V ( Fig . 6A ) along with a number of other mutant Flag-dSMN constructs , and found that they were able to pull down wild-type Myc-dSMN . Consistent with these findings , Martin et al . [44] recently showed that human SMN YG box mutants M263R , Y272C and G275S ( corresponding to M194R , Y203C and G206S in flies ) were completely unable to form dimers or higher-order multimers in vitro , as measured by multi-angle light scattering ( SEC-MALS ) . In contrast , a relatively mild YG box mutation , T274I ( T205I in flies ) , was only partially defective in self-interaction [28] , [40] , [44] , [50] . We conclude that the dominant negative interaction of Smn alleles is not limited to zygotically expressed protein , but extends to the maternal contribution as well . Moreover , these findings show that important structural and functional features of the SMN YG box are conserved between vertebrates and invertebrates . It is important to note that the dominant effect of expressing the M194R , Y203C and G206S transgenes was not observed when these alleles were carried over a balancer or a wild-type third chromosome . Smn transgenes integrated at the 86Fb landing site express relatively low levels of Flag-dSMN , compared to the endogenous gene located at 73A9 ( Fig . 2B ) . One interpretation of these results is that high levels of wild-type dSMN expression from the endogenous gene can squelch the activity of the mutant transgenic protein . Alternatively , the observed decrease in viability of the heteroallelic Smn combinations could be due to second site recessive mutations , as the transgenes were all inserted into the same genetic background . To explicitly address this question , we over-expressed wild-type Flag-dSMN from a transgene located on the second chromosome , using the UAS-Gal4 system [51] . Next , we measured eclosion frequencies of UAS-Flag-Smn , act5C-Gal4; SmnX7/X7 , Flag-SmnWT/Y203C transgenic rescue animals versus those that lack the act5C-Gal4 driver . As shown in Fig . 6B , overexpression of WT Flag-dSMN increased the eclosion frequency from ∼6% to ∼60% . Equivalent results were obtained with a tubulin-Gal4 driver . Thus , we can fully rescue the dominant negative effect of the heteroallelic SmnY203C/WT transgenes by overexpressing WT dSMN , demonstrating that the decrease in viability of these animals is not due to second site recessives . We therefore conclude that severe Smn YG box mutations can act in a dominant fashion , given that our transgenic alleles are expressed at relatively low ( but equivalent ) levels , the observed dominant behavior of the severe YG box mutants could also be viewed as creating a haploinsufficient condition , titrating away wildtype SMN monomers and/or Gemin components . In either event , the ‘squelching’ experiment ( Fig . 6B ) is important because it suggests that overexpression of full-length SMN protein ( e . g . from SMN2 ) may be an effective therapy for SMA patients bearing severe SMN1 point mutations . Within the SMN YG box dimer [44] , conserved tyrosine and glycine residues are thought to form a network of intersubunit interactions , wherein each tyrosine side chain packs against the main chain atoms of the of the i+3 glycine on the opposing helix . As illustrated in Fig . 7 , molecular modeling of Drosophila SMN dimers reveals that the Y203C mutation is predicted to disrupt this intermolecular Y-G contact . Wild-type homodimers can form two such interactions , whereas Y203C:WT heterodimers can form only one of them; Y203C homodimers are incapable of making these contacts ( Fig . 7 ) . Similarly , mutation of G206 to a bulkier serine is predicted to have a disruptive effect . Most important , the Y203-G206 contact occurs precisely at the point at which the two helices cross , so this interaction is expected to be most critical for stability of the SMN dimer . Interestingly , the highly conserved threonine residue present within the 203YxTG206 motif does not participate directly in the dimer interface [44] . Consistent with this observation , we found that the T205I mutant interacts with dGem3 , whereas the Y203C and G206S constructs do not . These results suggest that dGem3 interacts with a multimeric form of SMN . Because the human SMN crystal structure [44] does not contain any information regarding residues that are located upstream of M263 ( M194 in the fruitfly ) , we are currently unable to create an accurate molecular model of the M194R mutation . However , it is interesting to note that constructs containing the proximal N-terminal region of human SMN ( i . e . residues 252–294 instead of 263–294 ) have a strong tendency to form higher-order structures , primarily tetramers and octamers [44] . Taken together with the relative instability of dSMN ( M194R ) protein ( Fig . 2B ) , its inability to self-interact in cultured cells ( Fig . 3A ) , and the severe larval-lethal phenotype of this mutation ( Fig . 4 ) , these results support the view that formation of higher-order multimers of SMN is important for its function in vivo ( Fig . 7 ) . Gaining crucial insight from simpler model organisms is a proven strategy for unraveling complicated biological questions . Using a single point mutation ( SmnT205I ) , we recently showed that the larval locomotion and adult viability defects associated with loss of SMN can be uncoupled from the snRNP assembly function of SMN [28] . More specifically , we found that complete loss of SMN did not result in appreciable changes in the splicing of minor intron genes [28] , [29] , thus arguing for a non-splicing mechanism for SMA etiology . In this report , we have conducted genetic and biochemical characterization of eleven additional Smn mutants . We show that this allelic series captures a range of phenotypic severities , further suggesting a high level of conservation between the human and fruitfly SMN proteins and related pathways . The complex organization and polymorphic nature of the two human SMN genes complicates the analysis of SMA patient phenotypes . Indeed , SMN2 copy number variation is the best known modifier of SMA , potentially masking the phenotype of SMN1 point mutants [4] . However , SMN2 copy number is not always predictive of SMA disease severity [52] . Thus , developing an animal model that is an accurate predictor of SMN protein function is an important goal . We find good correlation between the biochemical properties of the YG box mutants and their phenotypic severities . Patients bearing point mutations in SMN1 are relatively rare , and genotypic information ( SMN2 copy number ) is often not available . Such is the case for the only patient reported to bear a G275S mutation ( G206S in flies ) , who presented with a mild form of SMA but SMN2 copy number was not determined [53] . Biochemically speaking , G275S should be a severe mutation , as this mutant is unable to form oligomers [44] . Consistent with these findings , G206S is a severe mutation in the fly and fails to bind Gemin3 ( this work ) . Similarly , four human patients bearing the T274I mutation presented with intermediate forms of SMA , yet they had only a single copy of SMN2 [53] , [54] . Given that the T274I mutant was shown to be active in Sm core assembly [42] , the milder human phenotype was perhaps to be expected . Concordantly , the corresponding fly mutant , T205I , also displays an intermediate phenotype ( [28]; this work ) . The same is true for D44V ( D20V ) , which is located in the Gemin2 binding domain , and presents with a relatively mild phenotype in both humans and flies . In contrast , the phenotypes of the Tudor domain mutants were harder to compare . Of the five mutations tested , two of them ( F70S and I93F ) were not as severe as one would have predicted from the human data ( W92S , [55]; I116F , [56] ) . Using an S2 cell co-transfection assay similar to that in Fig . 3 , we screened the entire panel of Smn mutants for their ability to bind to SmD1 and found no appreciable differences . Previous results suggest that Drosophila cells are less sensitive to the methylation status of Sm proteins than are human cells [57]–[59] . Given that the Tudor domain is thought to be a methyl-binding module , perhaps Drosophila are more tolerant to mutations in this region of SMN . However , we previously identified a synthetic lethal genetic interaction between Smn and Dart5 ( the fruitfly orthologue of the arginine methyltransferase , PRMT5; see [59] ) , so methylation of Sm proteins ( or other SMN interactors ) may not be completely dispensable in flies . In the absence of more quantitative biochemical assays of SMN function ( particularly for the Tudor domain mutants ) , we are currently unable to make good correlations between genotype and phenotype . Thus , additional efforts in this area will be needed . Additional mutations , those that are patient-derived as well as those that are predicted by ultrastructural studies , should greatly aide future investigations of the SMN YG box oligomerization motif by providing an all-important organismal readout . Moreover , additional phenotypic analyses ( longevity , locomotion , flight , neuromuscular development , etc . ) particularly of adult animals , should provide quantitative measures of differences between the weaker Smn alleles . Correlating this information along with proteomic and RNomic analyses of these alleles will provide important data on SMN function and SMA etiology .
All stocks were cultured on molasses and agar at room temperature ( 24±1°C ) in half-pint bottles . Oregon-R was used as the wild-type allele . The SmnX7 microdeletion allele was a gift from S . Artavanis-Tsakonis ( Harvard University , Cambridge , USA ) . This deficiency removes the promoter and the entire SMN coding region , leaving only the final 44 bp of the 3′ UTR [39] . Control and mutant larvae ( 73–77 hrs post egg-laying ) were imaged at 10× magnification with a stereo dissection microscope ( Leica ) at 10× magnification . Images were captured with a digital camera and larval outlines were traced . Total pixel area was calculated using ImageJ software and measurements converted to square millimeters . For larval locomotion , Smn control and mutant larvae ( 73–77 hours post egg-laying ) were placed on 1 . 5% agarose molasses plates and prodded with a needle to stimulate movement . After 20 s an image was recorded ( 7× magnification ) and the tracks were traced and total distance traveled was measured using ImageJ software . P-values were generated using a 2-tailed student's t-test , assuming unequal variance . As described in Praveen et al . [28] , a ∼3 kb fragment containing the entire Smn coding region was inserted into the pAttB vector [38] . A 3× FLAG tag was inserted immediately downstream of the dSMN start codon . Point mutations were introduced into this construct using Quickchange ( Invitrogen ) site-directed mutagenesis according to manufacturer's instructions . The transgenes were injected directly into embryos heterozygous for the SmnX7 microdeletion [39] that was recombined prior to injection with the 86Fb PhiC31 landing site ( Bloomington Stock Center , IN , USA ) . The injections were performed by BestGene Inc . ( Chino Hills , CA ) . S2 cell lines were obtained from the Drosophila Genome Resource Center ( Bloomington , IL ) . S2 cells were maintained in Express Five SFM ( Gibco ) supplemented with 1% penicillin/streptomycin and 9% L-glutamine . Cells were removed from the flask using a cell scraper and passaged to maintain a density of approximately 106–107 cells/mL . S2 cells were transferred to SF900 SFM ( Gibco ) prior to transfection with Cellfectin II ( Invitrogen ) . Transfections were performed according to Cellfectin II protocol in a final volume of 3 mL in a T-25 flask containing 5×106 cells that were plated one hour before transfection . The total amount of DNA used in transfections was 2 . 5 ug . Larval lysates were prepared by crushing animals in lysis buffer ( 50 mM Tris-HCl [pH 7 . 5] , 150 mM NaCl , 1 mM EDTA , 1% NP-40 ) with 1× protease inhibitor cocktail ( Invitrogen ) and clearing the lysate by centrifugation at 13 , 000 RPM for 10 min at 4°C . S2 cell lysates were prepared by suspending cells in lysis buffer ( 50 mM Tris-HCl [pH 7 . 5] , 150 mM NaCl , 1 mM EDTA , 1% NP-40 ) with 10% glycerol and 1× protease inhibitor cocktail ( Invitrogen ) and disrupting cell membranes by pulling the suspension through a 25 gauge needle ( Becton Dickinson ) . The lysate was then cleared by centrifugation at 13 , 000 RPM for 5 min at 4°C . Western blotting on lysates was performed using standard protocols . Rabbit anti-dSMN serum [28] was affinity purified . For Western blotting , dilutions of 1 in 2 , 500 for the affinity purified anti-dSMN , 1 in 10 , 000 for anti-α tubulin ( Sigma ) , 1 in 10 , 000 for monoclonal anti-FLAG ( Sigma ) , 1 in 10 , 000 for polyclonal anti-Myc and 1 in 5000 for monoclonal anti-Myc ( Santa Cruz ) were used . Anti-FLAG antibody crosslinked to agarose beads ( EZview Red Anti-FLAG M2 affinity gel , Sigma ) or anti-Myc antibody crosslinked to agarose beads ( Sigma ) were used to immunoprecipitate FLAG and Myc tagged proteins from cells . A structural model of a Drosophila SMN YG-box dimer was generated by replacement of side chains in the Martin et al . [44] structure ( pdb code 4GLI ) with those that differ in the Drosophila SMN sequence . For each substitution , the new side chain rotamer was adjusted to be identical to that found in the human structure . No changes to the main chain conformation were made and no steric clashes were observed in the final model .
|
Spinal Muscular Atrophy ( SMA ) is a prevalent childhood neuromuscular disease , which in its most common form causes death by the age of two . One in fifty Americans is a carrier for SMA , making this genetic disease a serious health concern . SMA is caused by loss of function mutations in the survival motor neuron 1 ( SMN1 ) gene . SMN is an essential protein and has a well-characterized function in the assembly of small nuclear ribonucleoproteins ( snRNPs ) , which are core components of the spliceosome . To elucidate the phenotypic consequences of disrupting specific SMN protein interactions , we have generated a series of SMA-causing point mutations , modeled in Drosophila melanogaster . Using this system , we have shown that key aspects of SMN structure and function are conserved between humans and flies . Intragenic complementation analyses reveal the potential for dominant negative interactions between wild-type and mutant SMN subunits , highlighting the essential nature of the YG box in formation of higher-order SMN multimers . These results provide a basis for future studies investigating therapy targeted at restoration of functional SMN oligomers .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biomacromolecule-ligand",
"interactions",
"invertebrates",
"animal",
"genetics",
"molecular",
"complexes",
"neuroscience",
"animals",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"molecular",
"genetics",
"drosophila",
"research",
"and",
"analysis",
"methods",
"molecular",
"biology",
"insects",
"arthropoda",
"biochemistry",
"genetics",
"biology",
"and",
"life",
"sciences",
"human",
"genetics",
"genetics",
"of",
"disease",
"organisms"
] |
2014
|
SMA-Causing Missense Mutations in Survival motor neuron (Smn) Display a Wide Range of Phenotypes When Modeled in Drosophila
|
Equine arteritis virus ( EAV ) is the causative agent of equine viral arteritis ( EVA ) , a respiratory , systemic , and reproductive disease of horses and other equid species . Following natural infection , 10–70% of the infected stallions can become persistently infected and continue to shed EAV in their semen for periods ranging from several months to life . Recently , we reported that some stallions possess a subpopulation ( s ) of CD3+ T lymphocytes that are susceptible to in vitro EAV infection and that this phenotypic trait is associated with long-term carrier status following exposure to the virus . In contrast , stallions not possessing the CD3+ T lymphocyte susceptible phenotype are at less risk of becoming long-term virus carriers . A genome wide association study ( GWAS ) using the Illumina Equine SNP50 chip revealed that the ability of EAV to infect CD3+ T lymphocytes and establish long-term carrier status in stallions correlated with a region within equine chromosome 11 . Here we identified the gene and mutations responsible for these phenotypes . Specifically , the work implicated three allelic variants of the equine orthologue of CXCL16 ( EqCXCL16 ) that differ by four non-synonymous nucleotide substitutions ( XM_00154756; c . 715 A → T , c . 801 G → C , c . 804 T → A/G , c . 810 G → A ) within exon 1 . This resulted in four amino acid changes with EqCXCL16S ( XP_001504806 . 1 ) having Phe , His , Ile and Lys as compared to EqCXL16R having Tyr , Asp , Phe , and Glu at 40 , 49 , 50 , and 52 , respectively . Two alleles ( EqCXCL16Sa , EqCXCL16Sb ) encoded identical protein products that correlated strongly with long-term EAV persistence in stallions ( P<0 . 000001 ) and are required for in vitro CD3+ T lymphocyte susceptibility to EAV infection . The third ( EqCXCL16R ) was associated with in vitro CD3+ T lymphocyte resistance to EAV infection and a significantly lower probability for establishment of the long-term carrier state ( viral persistence ) in the male reproductive tract . EqCXCL16Sa and EqCXCL16Sb exert a dominant mode of inheritance . Most importantly , the protein isoform EqCXCL16S but not EqCXCL16R can function as an EAV cellular receptor . Although both molecules have equal chemoattractant potential , EqCXCL16S has significantly higher scavenger receptor and adhesion properties compared to EqCXCL16R .
Equine arteritis virus ( EAV ) is a single-stranded , positive-sense RNA virus that belongs to the family Arteriviridae in the order Nidovirales [1–3] . It is the causative agent of equine viral arteritis ( EVA ) a respiratory , systemic , and reproductive disease of horses [2 , 4 , 5] . While most naturally acquired EAV infections are clinically inapparent , relatively virulent field strains of EAV periodically emerge around the world giving rise to outbreaks of EVA [6 , 7] . The disease is characterized by fever ( greater than 41°C ) ; depression; leukopenia; rhinitis often accompanied by nasal discharge; urticaria; and edema [8] . Abortion is a frequent outcome in naïve pregnant mares and congenital infection in neonatal foals is characterized by severe , fulminating interstitial pneumonia [9] . In the stallion , EAV is shed in semen during the acute phase of the infection and in some individuals , for a short time during the convalescent period until they clear the virus entirely from all body tissues [10] . However , in contrast , EAV establishes long-term persistent infection in 10–70% of infected stallions and these constantly shed virus in their semen for extended periods ( years or even life long ) [8 , 11 , 12] . The mechanism of long-term persistence of EAV in the reproductive tract of stallions is not well understood . It has been shown that EAV persistence in sexually intact post-pubertal colts or stallions is testosterone dependent [13 , 14] . Persistently infected stallions play an important role in maintenance and perpetuation of the virus in equine populations by transmitting the virus during breeding to naïve susceptible mares and can be responsible for outbreaks of EVA [8 , 13–17] . The use of virus-infective frozen or chilled semen for artificial insemination and embryo transfer can increase the risk of spread of EAV [18] . In previous studies in our laboratory , it has been shown that the experimentally derived virulent Bucyrus strain ( VBS ) of EAV can infect CD3+ T lymphocytes in vitro from some but not all horses [19] . In one study , 310 horses of Thoroughbred , Standardbred , Saddlebred , and Quarter horse breeds were tested and their phenotypes identified with respect to in vitro infection of their CD3+ T lymphocytes by the VBS of EAV [20] . Those whose CD3+ T lymphocytes could be infected in vitro with this virus strain were identified as susceptible , and those whose cells were not infected were identified as resistant [20] . A genome wide association study ( GWAS ) identified a common haplotype associated with the in vitro CD3+ T lymphocyte susceptible phenotype in a region of equine chromosome 11 ( ECA11:49572804–49643932 ) whose distribution was consistent with a dominant mode of inheritance [20] . Subsequently , it has also been demonstrated that stallions with the CD3+ T lymphocyte susceptibility phenotype to in vitro EAV infection are also at a significantly higher risk of becoming persistently infected carriers compared to those that lack this phenotype [21] . The primary objective of this study was to identify the specific gene ( s ) involved in the in vitro CD3+ T lymphocyte susceptibility to EAV infection and also the molecular mechanism ( s ) responsible for this phenomenon . We have combined contemporary genomics , molecular biology , and proteomics techniques to demonstrate that a specific gene in equine chromosome 11 ( ECA11 ) , equine CXCL16 ( EqCXCL16 ) , plays an essential role in the in vitro CD3+ T lymphocyte susceptibility to EAV infection . In this study , we have shown the EqCXCL16 gene has three alleles , two coding for the susceptibility phenotype ( EqCXCL16Sa and EqCXCL16Sb ) and one coding for the resistant phenotype ( EqCXCL16R ) . Furthermore , compelling evidence is provided that allelic variation within EqCXCL16 is a major determining factor for establishment of long-term persistent EAV infection in the stallion reproductive tract . This study also identified key functional differences between the EqCXCL16S and EqCXCL16R proteins , highlighting the biological significance of mutations present in the EqCXCL16 gene .
Three horses were selected for whole genome sequence analysis to identify nucleotide substitutions within the 3 megabase target region of ECA11 ( ECA11: 48M-51M ) associated with in vitro susceptibility or resistance of CD3+ T lymphocytes to infection with EAV . Two horses ( Thoroughbred-10 [TB10] and Standardbred-22 [STB22] ) possessed CD3+ T lymphocytes with the EAV susceptible phenotype while the corresponding cells in Thoroughbred-3 ( TB03 ) were resistant . Sequencing of 500 bp paired end libraries was accomplished using an Illumina HiSeq2000 platform to a depth of approximately 30× coverage . Following alignment to the reference genome ( Ecab 2 . 0 ) , the distribution of SNPs within ECA11: 48M-51M was compared to identify differences between susceptible and resistant horses . SNPs of interest were those which occurred in both susceptible horses but not the resistant horse . The analysis focused on identification of SNPs causing non-synonymous mutations within exons for genes annotated by Ensembl as these were considered to have a high likelihood of influencing the phenotype . In total , 12 non-synonymous SNPs were found in exons for eight annotated genes within the target region ( Table 1 ) . No translational frame-shift insertions or deletions ( indels ) were found within any previously annotated exon . Potential associations between SNPs identified by genomic sequencing and the CD3+ T lymphocyte EAV susceptibility phenotype were determined by sequencing PCR amplified fragments spanning each of the 12 SNP locations from a subset of the 240 horses ( resistant [n = 2] and susceptible [n = 8] ) ( Table 2 ) . Only the SNPs associated with the equine orthologue of the gene encoding the chemokine CXCL16 ( EqCXCL16 ) showed a complete association; the two resistant horses were homozygous for the A , G , T , and G SNPs at ( XM_00154756 ) c . 715 , c . 801 , c . 804 , and c . 810 , while all eight susceptible horses were either homozygous for the SNPs T , C , A , and A or were heterozygous A/T , G/C , T/A , and G/A at those positions , respectively , This observation is consistent with genetic dominance for the EAV CD3+ T lymphocyte susceptibility phenotype as reported previously [20] . To investigate the distribution of the four SNPs described above in a larger , more representative sample of the horse population , exon 1 of EqCXCL16 was PCR amplified and sequenced from 240 horses ( Thoroughbred [n = 67] , Standardbred [n = 60] , Quarter Horse [n = 53] , and Saddlebred horses [n = 60]; see S1 Table for primer sequences ) . The 240 horses had previously been characterized as EAV CD3+ T resistant ( n = 85 ) or susceptible ( n = 155 ) . Based on the annotation for the equine reference CXCL16 ( XM_001504756 ) sequence , all resistant horses ( n = 85 ) were homozygous for the SNPs A , G , T , and G at positions 715 , 801 , 804 , and 810 . However , all susceptible horses were homozygous or heterozygous at those positions and possessed the SNPs T , C , A , A or T , C , G , A ( n = 154 ) at the respective positions . A single ( n = 1 ) homozygote was observed for a putative haplotype possessing the combination T , C , G , A; that individual was phenotyped as susceptible . These two susceptibility alleles ( T , C , A , A and T , C , G , A ) were designated EqCXCL16Sa and EqCXCL16Sb , respectively . The predicted amino acid sequences were identical for EqCXCL16Sa and EqCXCL16Sb as described below . The allele for resistance was recessive and was characterized by A , G , T , and G at positions 715 , 801 , 804 , and 810 . This allele was designated EqCXCL16R and was identical to the reference genome sequence ( Ecab 2 . 0 ) . No additional alleles were found in exon 1 of EqCXCL16 among 47 horses of 20 other breeds tested during the study . The gene frequencies of the three CXCL16 alleles are shown in Table 3 for Thoroughbred , Quarter Horse , Saddlebred , and Standardbred horse breeds . The originally identified SNPs were non-synonymous mutations , predicting a change of amino acids in the protein ( Table 1 ) . The predicted amino acid sequences for this region of exon 1 specify phenylalanine ( Phe ) , histidine ( His ) , isoleucine ( Ile ) and lysine ( Lys ) at amino acid positions 40 , 49 , 50 , and 52 respectively , for both EqCXCL16Sa and EqCXCL16Sb based on the horse reference sequence , XM_001504756 . The predicted amino acids at those positions for EqCXCL16R were tyrosine ( Tyr ) , aspartic acid ( Asp ) , phenylalanine ( Phe ) , and glutamic acid ( Glu ) . These are non-conservative substitutions . The distribution of the two predicted EqCXCL16 proteins among the 240 Thoroughbred , Quarter Horses , Saddlebred , and Standardbred was compared to the distribution of their CD3+ T lymphocyte phenotypes for resistance or susceptibility to equine arteritis virus infection . As can be seen in Table 4 , the association was complete . This observation is particularly noteworthy since it occurred across a group including 4 different horse breeds . The 5 carriers with the resistance genotype came from 5 different breeds ( Saddlebred , Friesian , Lusitano , Thoroughbred , and Quarter Horse ) while the 11 non-carriers with susceptibility genotypes came from 8 different breeds ( Belgian Warmblood , Belgian draft , Saddlebred , Standardbred ( 4 ) , Paint , Friesian , Hanoverian , and Quarter Horse . The predicted amino acid sequences within exon 1 of CXCL16 was compared for seven species using the reference genome sequences from white rhinoceros ( Ceratotherium simum ) , horse ( Equus caballus ) , dog ( Canis lupus familiaris or Canis familiaris ) , human ( Homo sapiens ) , domestic rat ( Rattus norvegicus ) , cattle ( Bos taurus ) , and African elephant ( Loxodonta africana ) ( Fig 1 ) . As noted above , the horse genome reference sequence ( Ecab 2 . 0 ) is identical to EqCXCL16R ( Fig 1 ) . Although the sequences depicted in Fig 1 have some shared elements such as a GN-GS motif ( positions 2–6 Fig 1 ) or a C residue at position 11 , this region is not highly conserved between species and different amino acid side chains are permitted at sites equivalent to those involving polymorphism within CXCL16 in the horse . However , position 40 of EqCXCL16R differs in that Tyr is hydrophilic; while in EqCXCL16S and all other species examined , this site is occupied by a non-polar residue . In addition , EqCXCL16S differs at position 49 in that His has a basic side chain; whereas in the case of EqCXCL16R , along with other species , this position is generally occupied by amino acids with acidic or polar side chains ( Fig 1 ) . In contrast , the identities of amino acids or amino acid side chain properties at positions 50/52 in EqCXCL16S and EqCXCL16R are observed in other species , with Phe in EqCXCL16R seen in humans along with another perissodactyl , the white rhinoceros , and Ile is present in cattle . Similarly , at position 52 , Glu in EqCXCL16R is also present in the white rhinoceros; although Lys in EqCXCL16S is not found in any of the other species shown in Fig 1 , positively charged Arg is observed in cattle and African elephants . These results suggest that exon 1 of CXCL16 has been exposed to considerable selective pressure over time and imply adaptive evolution for this region unique to each mammalian species . We previously reported an association between susceptibility of CD3+ T lymphocytes to infection and the ability of EAV virus to establish the long-term carrier state in stallions ( 7 ) . Following natural infection , some stallions continue to shed virus in the semen for long periods ( more than one year to lifelong ) after EAV is no longer present in blood and nasal secretions [2 , 22] and as such become reservoirs for the maintenance of this virus in equid populations . Therefore , we analyzed semen from 77 stallions comprising 24 different breeds that had been infected with EAV and their status as long-term carrier animals determined ( shedders; Table 5 ) . Since the CXCL16 alleles showed complete association with the CD3+ T lymphocyte phenotype across several breeds , data for the carrier-status phenotype were pooled for 77 stallions from different breeds . Of these , 37 were identified as long-term carriers ( stallions EAV seropositive and shedding detectable levels of virus in semen more than one year after initial infection ) while 40 were identified as non-carriers ( stallions seropositive for EAV , but had apparently cleared the virus from the reproductive tract [non-shedders] ) based on absence of detectable virus in semen post-infection . Exon 1 of EqCXCL16 was PCR amplified and sequenced from all 77 stallions to determine each stallion’s genotype . The results demonstrated a strong , although incomplete , association between long-term EAV carrier status in stallions and the presence of at least one copy of the EqCXCL16Sa or EqCXCL16Sb allele among 86% of the stallions with that phenotype ( Table 6 ) . In contrast , 73% of stallions that had been infected with EAV but were negative for virus in their semen following infection were homozygous for EqCXCL16R ( Table 6 ) . Recently , we demonstrated that the membrane-associated form of the chemokine encoded by the EqCXCL16Sa allele can function as a host-cell receptor for EAV binding and entry [23] . Since the four predicted amino acid substitutions discovered in exon 1 of EqCXCL16R compared to EqCXCL16Sa/b appear to be non-conservative , we hypothesized they might modify the functional EAV receptor properties of the resultant EqCXCL16R molecule . Consequently , as described previously for EqCXCL16S [23] , we established an HEK-293T cell line for stable expression of the transmembrane form of the EqCXCL16R protein . Stable expression of the respective proteins from HEK-EqCXCL16S and HEK-EqCXCL16R cells were compared using confocal microscopy ( Fig 2A ) and Western blot ( WB ) analysis ( Fig 2B ) using guinea pig ( Gp ) α-EqCXCL16 antisera . The results confirmed that the variant forms of EqCXCL16 were produced in apparently similar amounts and that they were both clearly associated with the plasma membrane ( Fig 2A ) . Infection of HEK-EqCXCL16S with the EAV sVBSmCherry encoding the mCherry fluorescent protein confirmed our previous reported findings that by 12 hpi the fluorophore was expressed in almost every cell ( Fig 2C , panel b ) . In contrast , expression of mCherry in HEK-EqCXCL16R cells infected at the same MOI with EAV sVBSmCherry was almost undetectable at the same time point and as such was equivalent to that observed in mock transfected HEK-293T cells ( Fig 2C , panels a and c ) . This result was confirmed in experiments where EAV gene expression following infection was detected by an indirect immunofluorescence assay ( IFA ) using a monoclonal antibody against viral nonstructural protein 1 ( nsp-1 ) . Although nsp-1 expression was detectable in a few EAV infected HEK-EqCXCL16R and mock transfected HEK-293T cells ( Fig 2D , panels b and c ) , the numbers were markedly lower than observed in similarly infected HEK-EqCXCL16S cells ( Fig 2D , panel a ) . These results demonstrate that the plasma membrane-associated EqCXCL16S isoform can function as an efficient cellular receptor for EAV while EqCXCL16R does not . By employing a combination of the virus overlay protein binding assay ( VOPBA ) and Far-Western blot ( Far-WB ) techniques , we previously demonstrated that EAV VBS could directly bind with EqCXCL16S [23] . In view of the results outlined above , the experiment was repeated with the stable cell lines expressing both isoforms of EqCXCL16 ( HEK-EqCXCL16S and HEK-EqCXCL16R cell lines ) . Equal amounts of total lysate from HEK-EqCXCL16S , HEK-EqCXCL16R , and mock transfected HEK-293T cells were subjected to polyacrylamide gel electrophoresis and transferred onto a polyvinylidene difluoride ( PVDF ) membrane prior to sequential denaturation and renaturation by treatment with guanidine-HCl ( Gn-HCl ) AC buffer . Refolded membrane-bound prey proteins were incubated with purified EAV VBS ( bait protein ) which was detected using a mouse monoclonal antibody directed against the GP5 envelope protein of EAV ( mouse α-GP5 Ab ) . Two strong signals were detected with apparent molecular weights ( MW ) of approximately 52 and 30 kDa in HEK-EqCXCL16S , but not HEK-EqCXCL16R or mock transfected HEK-293T ( naïve HEK ) cell lysates ( Fig 3A , panel a ) . No bands were visible in a control experiment where the EAV VBS binding step was omitted ( Fig 3A , panel b ) , demonstrating the absence of non-specific interactions between mouse α-GP5 Ab and cellular lysate proteins . The monomeric form of CXCL16S has an apparent molecular weight of 30 kDa , and this was confirmed by stripping the membrane depicted in panel a of Fig 3 and re-probing with polyclonal Gp α-EqCXCL16 antisera ( pAb ) . In this experiment , co-migrating bands with an apparent MW of 30 kDa were visible in both HEK-EqCXCL16S and HEK-CXCL16R lysates ( Fig 3 , panel c ) . The detection of a strong signal at 52 kDa is unlikely to be the result of interactions between EAV VBS and another cellular protein because this was only visible in HEK-EqCXCL16S lysates . Therefore , while initial polyacrylamide gel separation of the cell lysates was conducted under denaturating conditions , it is conceivable that EqCXCL16S may associate strongly with other proteins . Some support for this hypothesis is provided by the fact Gp α-EqCXCL16 antisera reacts with material with apparent MWs between 52 and 30 kDa in HEK-EqCXCL16S and HEK-EqCXCL16R , but not in mock transfected HEK-293T cell lysates ( Fig 3 , panel c ) . To further confirm these observations , biotinylated EAV VBS was incubated with HEK-EqCXCL16S , HEK-EqCXCL16R , and naïve control HEK-293T cells at 4°C to prevent internalization . Subsequently , the bound virus particles were visualized using streptavidin FITC . Fluorescence microscopy revealed that significantly higher number of HEK-EqCXCL16S cells bound to biotinylated EAV VBS compared to the HEK-EqCXCL16R or control HEK-293T cells ( Fig 3B ) . Taken together , these results ( Fig 3A and Fig 3B ) confirm that EAV VBS is capable of binding to EqCXCL16S , but that this association is almost completely abrogated by the four amino acid substitutions encoded by exon 1 of EqCXCL16R . The CXCL16 protein was originally identified as a scavenger receptor for oxidized low-density lipoprotein ( OxLDL ) in humans [25 , 26] . Therefore , it was decided to determine if scavenger receptor activity differed between EqCXCL16 isoforms . The HEK-EqCXCL16S , HEK-EqCXCL16R , and naïve HEK-293T control cells were incubated with Dil labelled OxLDL ( Dil-OxLDL ) and analyzed by fluorescence microscopy . Interestingly , the highest levels of Dil-OxLDL binding and potential internalization were observed in the HEK-EqCXCL16S while amounts bound to HEK-EqCXCL16R were very low , similar to those observed in mock transfected naïve HEK-293T cells ( Fig 4 , panel b vs panels a and c ) . Furthermore , Dil-OxLDL binding to HEK-EqCXCL16S cells was significantly reduced by prior treatment with a polyclonal antibody against EqCXCL16 ( Gp α-EqCXCL16 pAb; Fig 4 , panel d ) , suggesting a high degree of specificity in reactivity between Dil-OxLDL and the “S” form of the EqCXCL16 chemokine . These data strongly suggest that the EqCXCL16S isoform retains the function of an efficient scavenger receptor for Dil-OxLDL , whereas this functional property appears to be almost completely absent in the EqCXCL16R isoform . This confirms that four amino acid substitutions in the ectodomain of the EqCXCL16S isoform play a critical role in scavenger properties which may also favor the infection of cells with EAV as compared to the EqCXCL16R isoform . Although the experiments described above were not quantitative , they did reveal obvious differences between the two EqCXCL16 isoforms in terms of OxLDL scavenger receptor activity . Therefore , Far-WB analysis was employed to determine if similar differences existed between EqCXCL16S and EqCXCL16R in binding to the CXCR6 receptor molecule . Purified recombinant HA-tagged EqCXCR6 protein was separated in SDS-PAGE and incubated with purified recombinant EqCXCL16S or EqCXCL16R protein and binding detected using EqCXCL16 specific antibody produced in rabbits ( Rb α-EqCXCL16 Ab ) . Very similar signal levels were observed with both EqCXCL16 isoforms ( Fig 5 , panels a and b ) at a position approximating a MW of 25 kDa which is the predicted size for cloned EqCXCR6 . This was confirmed by stripping the membrane depicted in Fig 5 ( panel a ) before re-probing with anti-His antibody , thus demonstrating that the location of the His-tagged EqCXCR6 protein was identical to that observed for the interaction between EqCXCL16 and EqCXCR6 ( Fig 5 , panel c ) . There was no non-specific reactivity between the EqCXCL16 isoforms and the anti-His antibody with BSA . These data demonstrated there are no major qualitative differences between EqCXCL16S and EqCXCL16R in binding to the EqCXCR6 receptor protein . The fact that human CXCL16 can act as a chemoattractant and recruit lymphoid cell types to sites of inflammation within the body suggests that this property may also be present in EqCXCL16 [25 , 27] . Therefore , experiments to determine if EqCXCL16 possessed chemoattractant potential and if this differed between the two isoforms were conducted using purified recombinant soluble forms of EqCXCL16S and EqCXCL16R proteins produced in E . coli . Equine CD3+ T lymphocytes derived from peripheral blood mononuclear cells ( PBMCs ) were enriched using anti-CD3 conjugated magnetic beads to > 95% purity based on the proportion of CD4+ and CD8+ T lymphocytes ( S1A Fig top row ) present in the population with the remainder comprised mainly of CD21+ B cells and CD14+ monocytes ( S1A Fig bottom row ) . Approximately 6% of this enriched CD3+ T lymphocyte population had detectable cell-surface expression of EqCXCR6 as determined by staining with rabbit anti-EqCXCR6 antibody ( PA7511 ) followed by analysis using flow cytometry ( S1B Fig ) . Incubation of enriched equine CD3+ T lymphocyte preparations at 37°C with the recombinant chemokines separated using a Boyden chamber with a 3 μm pore size provided clearest evidence that EqCXCL16 possessed chemoattractant potential compared to the medium only control ( P<0 . 001 ) ( Fig 6 ) . Furthermore , there were no statistically significant differences between the S and R isoforms in numbers of EqCXCR6 expressing CD3+ T lymphocytes entering the chamber ( P = 0 . 210 ) ( Fig 6 ) . Evidence of the chemoattractant properties can be attributed directly to EqCXCL16 and not to potential contaminants in the recombinant protein preparations was substantiated by the fact CD3+ T lymphocyte migration into the chamber was not statistically different from the medium only control where the equine chemokine isoforms were pre-incubated with Gp α-EqCXCL16 pAb ( Fig 6 ) . Membrane-associated forms of human CXCL16 have been reported to function as cellular adhesion molecules [25] . Preliminary observations suggested differences between HEK cells expressing the two EqCXCL16 isoforms and adhesion to cell culture plates . Therefore , HEK-EqCXCL16S , HEK-EqCXCL16R , and naïve control HEK-293T cells were tested using a previously described protocol [53] to evaluate resistance to detachment from the cell culture plate following incubation for 10 min at 37°C with 0 . 5M EDTA . Surprisingly , HEK cells expressing EqCXCL16S were found to be significantly ( P<0 . 0001 ) more resistant to treatment with EDTA than those expressing EqCXCL16R or naïve control HEK-293T cells ( Fig 7A and Fig 7B ) . Furthermore , the superior adhesion properties of HEK-EqCXCL16S were abrogated by pre-treatment prior to plating with Gp anti-EqCXCL16 polyclonal antibody providing strong evidence of resistance to EDTA-induced detachment was mediated only by the “S” isoform of the chemokine ( Fig 7A and Fig 7B ) .
Previously , we have shown that horses can be categorized into two distinct phenotypic groups based on the susceptibility of their CD3+ T lymphocytes to in vitro infection with EAV [19 , 20] . It was also discovered that stallions with EAV-susceptible CD3+ T lymphocytes were at higher risk of becoming long-term carriers following exposure to this virus than those with a resistant CD3+ T-cell phenotype [21] . Furthermore , a genome wide association study ( GWAS ) based on single nucleotide polymorphism ( SNP ) detection using the Equine SNP50 BeadChip suggested an association between these phenotypic traits and a dominant genetic marker ( s ) located on equine chromosome 11 ( ECA11 ) between nucleotide positions 49 , 572 , 804 and 49 , 643 , 932 [20] . In this study , we present genetic and functional data confirming that susceptibility or resistance of CD3+ T lymphocytes to in vitro EAV infection is indeed under genetic control and can be explained based on mutations found in the chromosome region predicted by GWAS . Whole genome sequencing of one resistant and two susceptible horses revealed 12 non-synonymous nucleotide substitutions distributed among eight previously annotated genes in the targeted region on ECA11 . CXCL16 was implicated as the putative cause of the phenotype based on tests of the SNPs among 10 horses ( 8 susceptible and 2 resistant ) that excluded the seven other genes . Furthermore , genetic association studies based on typing 240 horses of different breeds demonstrated three alleles of CXCL16 encoding for two proteins , EqCXCL16R and EqCXCL16S . The gene for EqCXCL16S was dominant and correlated completely with the EAV-susceptible CD3+ T lymphocyte phenotype . Specifically , EAV susceptibility of CD3+ T lymphocyte population members was conferred by two distinct alleles ( EqCXCL16Sa and EqCXCL16Sb ) , although the resultant proteins were predicted to possess identical amino acid sequences . These susceptibility alleles exhibited a dominant mode of inheritance , while that associated with CD3+ T lymphocyte EAV resistance ( EqCXCL16R ) was recessive , in that the phenotype was only found in equids homozygous for EqCXCL16R . At least three of the SNPs located within EqCXCL16Sa , EqCXCL16Sb , and EqCXCL16R are predicted to induce non-conservative or even radical amino acid substitutions within exon 1 of the resultant protein . If analogous to the human orthologue of CXCL16 , exon 1 should encode the N-terminal ectodomain of the chemokine and , therefore , play a major role in its biological properties . Although there is to our knowledge no published information concerning the functional properties of EqCXCL16 , the human equivalent of this protein ( CXCL16 ) is expressed as a type-I membrane protein and is comprised of several distinct domains ( chemokine , extracellular [mucin stalk] , hydrophobic transmembrane , and intracellular cytoplasmic ) [28] . The protein can also exist in an unbound soluble form resulting either from cleavage with cellular metalloproteinases such as ADAM10 [28] or alternative mRNAsplicing [29] . Membrane-associated forms of human CXCL16 ( HuCXCL16 ) are expressed on dendritic cells , CD14+ monocytes/macrophages , CD21+ B lymphocytes , endothelial cells , smooth muscle cells , and keratinocytes [30–37] . Moreover , expression on these cell types is upregulated by inflammatory mediators and bacterial lipopolysaccharides [27 , 29 , 36 , 38–40] . Both soluble and membrane-bound forms of HuCXCL16 specifically interact with its receptor CXCR6 ( also known as STRL33/BONZO/TYMSTR ) expressed on the surface of CD4+ and CD8+ T lymphocytes , NKT cells , and NK cells [27 , 41–43] . Binding to the CXCR6 receptor is facilitated by the mucin-stalk located within the extracellular domain of the molecule [44 , 45] . Soluble HuCXCL16 has strong chemotactic potential in that it effectively recruits CXCR6+ T lymphocytes to sites of inflammation . The extracellular domain of HuCXCL16 also recognizes oxidized low-density lipoprotein ( OxLDL ) along with phosphatidylserine; therefore , the protein is multifunctional , acting as a scavenger receptor in addition to possessing chemokine activity [34] . Furthermore , aberrant expression of HuCXCL16 is implicated in the pathogenesis of certain viral infections , arthritis , atherosclerosis , and the metastasis of some cancers [46–50] . Preliminary in silico studies suggested that EqCXCL16 possesses a domain structure very similar to its human counterpart including the presence of six cysteine residues within the chemokine domain . However , a Tyr-X-Pro-Val motif in the C-terminal intracellular domain believed to act as a potential substrate for tyrosine kinase phosphorylation in human and mouse variants of the protein is replaced by Tyr-X-Pro-Val in the horse . Experiments described here demonstrate EqCXCL16 binds to the equine orthologue of CXCR6 ( EqCXCR6 ) . In addition , the soluble form of EqCXCL16 has chemotactic activity for CXCR6 expressing T lymphocytes . Although there are no detectable qualitative changes in chemotactic property between the two equine isoforms described here , the predicted amino acid substitutions between EqCXCL16R and EqCXCL16S consisting of Tyr to Phe at position ( p ) 40 , Asp to His at p49 , Phe to Ile at p50 , and Glu to Lys at p52 ( p . Tyr40Phe , p . Asp49His , p . Phe50Ile , and p . Glu52Lys ) respectively , had considerable effects on the ability to bind OxLDL . Moreover , there were dramatic differences between HEK-293T cells expressing each of these proteins to adhere to plastic culture vessels in the presence of EDTA . Collectively , these results indicate that EqCXCL16S , in common with its human counterpart , has both chemotactic and scavenger receptor activity , whereas the latter property is likely to be substantially reduced in the EqCXCL16R isoform [51] . However , in terms of EAV , the most significant finding is that while the membrane-bound variant of EqCXCL16S can bind this virus and function as a cellular receptor [28] , these properties are completely abrogated in the EqCXCL16R isoform . This suggests amino acid residues located at positions 40 to 52 within the chemokine domain of EqCXCL16 have a direct role in viral attachment via interactions with EAV membrane surface glycoproteins . These studies confirm and extend our previous findings [28] that although some viruses such as HIV and severe acute respiratory syndrome ( SARS ) virus may specifically interact with CXCL16 to influence the course of viral pathogenesis [32 , 33] , EAV is at present unique in its ability to utilize this chemokine as a receptor protein . However , the choice of a scavenger receptor as a portal for viral entry does have a precedent among the arteriviruses in that porcine reproductive and respiratory syndrome virus ( PRRSV ) uses CD163 [48] . It has been reported that CXCL16 in humans is expressed by a subpopulation ( s ) of T lymphocytes [49 , 50] . Consequently , the simplest explanation for the two phenotypically distinct horse groups identified in our earlier studies [16] is that EAV “susceptible” animals ( EqCXCL16Sa , b/EqCXCL16Sa , b or EqCXCL16Sa , b/EqCXCL16R ) possess a subpopulation ( s ) of T lymphocytes with cell-surface expression of EqCXCL16S and are thus permissive for viral infection , while the equivalent CD3+ T lymphocyte population in those that are “resistant” express only EqCXCL16R ( EqCXCL16R/EqCXCL16R ) . Further studies are needed to investigate whether EqCXCL16 is constitutively expressed on CD3+ T lymphocytes or if expression can be increased by mitogen , TNFα , and IFNγ stimulation and/or by EAV infection of equine PBMCs . Another caveat to this hypothesis is that EAV exhibits a broad host-cell tropism and can infect several common laboratory cell lines from different species that certainly do not express EqCXCL16 . Previous studies from our laboratory and others have shown that EAV can utilize a number of different non-related , host-specified molecules as cellular entry receptors or accessory molecules [52] . Consequently , demonstrating that susceptibility of a CD3+ T lymphocyte subpopulation ( s ) to EAV infection results from the presence of EqCXCL16S acting directly as an entry receptor will necessitate not only showing the presence of this protein , but also that alternative receptors are either not expressed in these cells or , for some reason , not utilized . Furthermore , the exact mechanism ( s ) of CXCL16 in infection of equine CD3+ T lymphocytes is yet to be determined . One possibility is that CXCL16S ( or CXCL16R ) is constitutively expressed or can be induced to express on subpopulation of CD3+ T lymphocytes , but only a CD3+ T lymphocyte subpopulation expressing CXCL16S becomes infected with EAV ( Fig 8A and 8B; Model 1 ) . Alternatively , viral entry is mediated by the interaction between soluble isoform of CXCL16S and its cellular receptor CXCR6 expressed on CD3+ T lymphocytes ( Fig 8C and 8D; Model 2 ) . However , these two models are not mutually exclusive and , as such , both need to be tested to unequivocally confirm the mechanism of CD3+ T lymphocyte infection dynamics . Although differences between horses in EAV CD3+ T lymphocyte subpopulation susceptibility is an interesting observation and may correlate with the severity of acute clinical signs [53] , it is the association between the EqCXCL16 genotype and long-term carrier status in stallions that is of major importance because these animals are key to the survival of this virus in equid populations [10] . In contrast to the CD3+ T lymphocyte EAV-susceptible/resistant phenotype , the correlation between carrier status in stallions and EqCXCL16 allelic content was strong ( P<0 . 000001 Fisher's Exact Test without regard to breed ) but not absolute . This P-statistic has limited value in that the stallions in Table 6 came from diverse breeds ( Table 5 ) , however the association of the genotype with the phenotype without regard to breed was remarkable . In an analysis of 77 EAV-infected stallions , only 14% of the shedders and 86% of non-shedders were homozygous for the resistance genotype . Conversely , 74% of the shedders and 26% of non-shedders had the allele for the EqCXCL16S protein . Although results presented here are consistent with existence of the long-term carrier status in the majority of stallions being dependent on the membrane-bound form of EqCXCL16S acting as a cellular receptor for EAV , the fact that the EqCXCL16 allelic association is not complete suggests additional genetic , immunological , and viral factors or even environmental factors also play a role in this determination . Furthermore , the EAV carrier state in stallions has a number of features that distinguish it from many other persistent viral infections of the male reproductive tract . These add additional layers of complexity and must be considered in any proposed mechanism . For example , recent virus isolation and immunohistochemistry studies have confirmed that EAV persists primarily in the ampulla along with other accessory sex glands rather than in immunologically privileged sites such as the Sertoli cells within the testis ( Carossino et al . manuscript submitted ) [10] . This is despite the fact that carrier stallions possess active immune responses against EAV , and the virus is not detectable in any organ or tissue except the reproductive tract in carrier animals . Indeed inflammatory infiltrates in close proximity to viral antigen-expressing cells are frequently observed in the stallion reproductive tract indicating that EAV persistence is a continuous , dynamic process that occurs in the presence of active local immune responses . Clearly these responses are not completely effective in clearance of the virus for reasons that are unknown at present . Potential explanations range from some form of localized immunosuppression based on the fact that androgens such as testosterone can down-regulate immune responses , to more specific activity such as that mediated by T regulatory lymphocytes or even to differences in functional properties between EqCXCL16S and EqCXCL16R . These are not mutually exclusive , and so the failure to eliminate EAV from the stallion’s reproductive tract could result from a combination of mechanistic factors . In addition , the virus probably contributes to its own survival via antigenic drift as evidenced by the continual emergence of novel variants during the course of persistent infections [16 , 54 , 55] . However , it is also possible that variation exists between individual stallions in the efficacy of EAV-specific immune responses within the reproductive tract that can operate independently of the EqCXCL16 genotype . If so , this could explain why 25% of stallions that possess at least one EqCXCL16Sa/b allele cease shedding shortly after infection with EAV , while approximately one in six that are homozygous for EqCXCL16R become long-term ( more than one year ) carriers . Based on the EqCXCL16 gene polymorphisms and its association with long-term carrier status , it would be possible to develop an allelic discrimination real-time PCR assay to distinguish horses that are prone to become long-term versus short-term shedders . It is interesting that while polymorphisms of CXCL16 have been reported in , for example , the human CXCL16 gene [44 , 51 , 56–58] , these are located in regions other than exon 1 . Furthermore , these polymorphisms have not been shown to completely abrogate the chemokine , scavenger receptor properties [43] . The horse may therefore be highly unusual in possessing allelic variants of this gene with SNPs situated in exon 1 that in the case of EqCXCL16R appear to disrupt the ability of the resultant protein to act as a scavenger receptor for OxLDL . In conclusion , these genomic studies unequivocally demonstrate that horse genomic sequences encoding the EqCXCL16 chemokine are associated with in vitro susceptibility of equine CD3+ T lymphocytes to EAV infection , as well as the establishment of long-term carrier state in stallions . Although the molecular mechanisms associated with these phenotypic traits have not been fully elucidated , there is compelling evidence the plasma membrane-associated variant of EqCXCL16S can function as a cellular entry receptor for EAV , whereas this property is absent in the EqCXCL16R isoform of the protein . It is interesting that the sequence associated with abrogation of the virus binding site costs the horse the scavenging capability of CXCL16 . If we assume the EqCXCL16Sa/b form is ancestral , the virus will have made use of a functional part of EqCXCL16 , potentially to deter adaptation; selective pressure may have been sufficient for evolution of a variant that does not allow virus binding at the cost of losing the scavenging capability for OxLDL . The origins and selection pressures for these variants of EqCXCL16 warrant further study .
This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The animal protocol involving horses was approved by the University of Kentucky Institutional Animal Care and Use Committee ( IACUC; protocol number 2013–1098 ) . The animal protocol involving rabbits and guinea pigs was approved by the Thermo Scientific , Rockford , IL IACUC ( NIH OLAW assurance number: A3669-01 , USDA research license registration number: 23-R-0089 , and PHS assurance number: A3669-01 ) . This study was performed according to these IACUC-approved protocols . Equine pulmonary artery endothelial cells ( EECs ) were maintained in Dulbecco’s modified Eagle’s medium ( DMEM; Mediatech , Herndon , VA , USA ) with sodium pyruvate , 10% fetal bovine serum ( Hyclone Laboratories , Inc . , Logan , UT , USA ) , 100 U/ml of penicillin/100 μg/ml of streptomycin ( Gibco , Carlsbad , CA , USA ) , and 200 mM L-glutamine [59–61] . The high passaged rabbit kidney cells ( HP-RK-13 [KY] P399-409 , originally derived from ATCC CCL-37; American Type Culture Collection , Manassas , VA , USA ) were propagated in Eagle’s minimal essential medium with 10% ferritin-supplemented bovine calf serum ( Hyclone Laboratories , Inc . ) and 100 U/ml of penicillin/100 μg/ml of streptomycin ( Gibco ) . Human embryonic kidney ( HEK-293T ) cells ( ATCC CRL-3216 ) were propagated in DMEM with 10% ferritin-supplemented bovine calf serum ( Hyclone Laboratories , Inc . ) , 100 U/ml of penicillin , and 100μg/ml streptomycin ( Gibco ) . HEK-293T cells stably expressing EqCXCL16S and EqCXCL16R were maintained in DMEM with 10% ferritin-supplemented bovine calf serum ( Hyclone Laboratories , Inc . ) and puromycin ( Clontech Laboratories Inc . , Mountain View , CA , USA ) at 3 μg/ml of medium . Isolation of PBMCs from peripheral blood of horses ( n = 9 ) was performed as described previously [19 , 45] with some modifications . Briefly , blood ( 20 ml ) was collected aseptically using Vacutainer tubes containing 0 . 1 ml of 15% EDTA solution ( Covidien , Dublin , Ireland ) . PBMCs were isolated from the buffy-coat fraction by centrifugation through Ficoll-Paque Plus ( Amersham Biosciences , Piscataway , NJ , USA ) at 500 × g for 30 min at 25°C . The PBMC layer was collected and washed twice with Hanks balanced salt solution ( pH 7 . 4 ) ( Life Technologies , Grand Island , NY , USA ) by centrifugation at 300 × g for 10 min to eliminate the platelets . The cells were resuspended in RPMI-1640 medium with 2 mM GlutaMAX and 25 mM HEPES ( Life Technologies ) without FBS and counted using a Countess Automated Cell Counter ( Life Technologies ) . CD3+ T lymphocytes were enriched indirectly using magnetic microbeads according to the manufacturer’s protocol ( Miltenyi Biotec , Bergisch Gladbach , Germany ) . Briefly , PBMCs were blocked with 5% normal mouse serum ( Innovative Research , Novi , MI , USA ) and then incubated with anti-equine CD3 ( Clone UC F6G ) conjugated with Alexa Fluor 647 for 30 min . After washing with MACS buffer , the cells were incubated with anti-Alexa Fluor 647 microbeads for 20 min . The cells were then washed with MACS buffer and applied to an LS column , washed three times , and the bound CD3+ T lymphocytes were eluted with MACS buffer in the absence of the magnet , washed , and counted using a Countess Automated Cell Counter . The purity of the CD3+ T lymphocytes was confirmed by flow cytometric analysis . Staining was performed following the protocol as described previously [20 , 45] . These cells were used for studying the chemokine function of EqCXCL16S and EqCXCL16R . Two strains of EAV , the experimentally derived virulent Bucyrus strain ( VBS ) ( ATCC VR-796 ) [4 , 62] and the recombinant EAV VBS virus expressing mCherry ( EAV sVBSmCherry ) [24] were used . Both viruses were propagated in EECs to generate high titer working stocks as previously described [63 , 64] . Briefly , EECs infected with each virus were frozen at -80°C when 90–100% cytopathic effect ( CPE ) was observed . Cell lysates were clarified by centrifugation ( 500 × g ) at 4°C for 15 min , followed by ultracentrifugation ( Beckman Coulter , Miami , FL , USA ) at 121 , 600 × g through a 20% sucrose cushion in NET buffer ( 150 mM NaCl , 5 mM EDTA , and 50 mM Tris-HCl , pH 7 . 5 ) at 4°C for 4 h to pellet the virus . Purified preparations of each strain of EAV were resuspended in phosphate buffered saline ( PBS , pH 7 . 4 ) and frozen at -80°C . Virus stocks were titrated by standard plaque assay in RK-13 cells and titers were expressed as PFU/ml [65] . A total of 240 horses from four different breeds , Thoroughbred ( TB; n = 67 ) , American Saddlebred ( ASB; n = 60 ) , Standardbred ( STB; n = 60 ) , and Quarter Horses ( QH; n = 53 ) were randomly selected for sequencing . These horses were a random subset of horses randomly selected from farms in Central Kentucky and previously tested for CD3+ T lymphocyte phenotyping [15] . A panel of 77 archived semen samples from EAV carrier stallions ( n = 37 ) and non-carrier ( n = 40 ) that had been stored at -80°C was tested . These semen samples were previously submitted to the EVA OIE Reference Laboratory at the Maxwell H . Gluck Equine Research Center , Lexington , KY for testing . To confirm carrier status , isolation of EAV from equine semen samples was attempted in a high passage ( P399-409 ) rabbit kidney-13 ( RK-13 ) ( HP-RK-13 [KY] ) cell line according to the OIE described protocol [66] . Serum neutralizing antibodies to EAV were demonstrated by microneutralization assay in both carrier and non-carrier stallions to confirm that they were seropositive for EAV [67] . Based on clinical histories , none of the stallions had been vaccinated against EAV , and seroconversion was the result of natural infection . EAV long-term shedders were defined as those horses that had detectable EAV in their semen for more than one year following infection where date of exposure was known . Non-shedders of EAV were those which likely shed EAV in their semen during the acute phase of infection but that had ceased shedding virus at the time of initial testing for presence of the carrier state . These horses came from a wide range of breeds including Warmbloods , Standardbreds , Thoroughbreds , Quarter Horses , Belgian draft , Andalusian , Friesian , Rocky Mountain , Selle Francais , Tennessee Walking Horse , Arabian , Lusitano and American Saddlebred horses ( Table 5 ) . Comparisons of shedders to non-shedders for EqCXCL16 genotypes were made using Fisher's Exact Test . The susceptible or resistant phenotype of each animal was defined by dual-color flow cytometric analysis of in vitro EAV-infected CD3+ T lymphocytes as described previously [28] . Horses were classified as susceptible or resistant to in vitro EAV infection based on the in vitro susceptibility or resistance of their CD3+ T lymphocytes . Genomic DNA ( gDNA ) was obtained from PBMCs or semen of each animal by using the Puregene whole-blood extraction kit ( Qiagen , Valencia , CA , USA ) in accordance with the manufacturer’s instructions as previously described [20] . DNA quality and concentration were assessed using Nanodrop ( Thermo Scientific , Wilmington , DE , USA ) at an absorbance ratio of optical density at 260 nm/280 nm ( OD260/280 ) . Three horses were selected from those SNP genotyped and phenotyped for the CD3+ T lymphocyte susceptibility ( S ) or resistance ( R ) phenotype based on dual-color flow cytometric analysis in the previous study [19]; specifically , they were TB03 ( R ) , TB10 ( S ) , and ST22 ( S ) . Genomic DNA was submitted to BGI Americas ( Davis , CA , USA ) for sequencing from each of these three horses . Approximately 20 μg of DNA was submitted for construction of short insert ( 500 bp ) libraries for sequencing using the Illumina HiSeq2000 . To achieve 20–35 GB of raw data per lane , 100 bp paired end sequencing across 7 lanes was conducted . Approximately 30× coverage was obtained per sample; there were 874 , 258 , 138 clean reads of 95 . 86 Q20 with ST22; 838 , 009 , 546 clean reads of 95 . 61 Q20 with TB03; and 867 , 086 , 552 clean reads of 95 . 96 Q20 with TB10 . Reads were mapped to the horse genome reference sequence ( Ecab 2 . 0 ) [68] using CLC workbench 8 . 0 . 1 ( CLC Bio , Boston , MA , USA ) . The whole genome sequence data from the three horses have been submitted to the Sequence Read Archive and can be found under BioSample/experiment accession numbers SAMN03838869/SRX1097022 , SAMN03838867/ SRX1097495 and SAMN03838868/SRX1097492 for CXCL16 of TB10 , TB3 , and ST22 , horses respectively . The variant discovery and genotyping were done with the Genome Analysis Toolkit UnifiedGenotyper using arguments -nt 4 -gt_mode DISCOVERY—validation_strictness LENIENT . [69] . Genome annotation was from Ensembl version Equus caballus . EquCab2 . 75 . The primer sequences for the eight candidate genes found using the target region ( ECA11: 48M-51M ) are shown in S2 Table . A PCR was performed in a final volume of 18 μL and the reaction for amplification consisted of: 100 ng of genomic DNA , 0 . 15 mM of each primer , and 10 μL of AmpliTaq GoldFast PCR Master Mix ( Applied Biosystems , Foster City , CA , USA ) . Amplifications consisted of the following steps: initial denaturation at 95°C for 10 min; 30 cycles of denaturation at 95°C for 30 sec , annealing of primers at 56°C for 30 sec , extension at 72°C for 30 sec; and a final extension at 72°C for 10 min . Amplicons were shipped overnight for sequencing to Eurofins Genomics ( Louisville , KY , USA ) . DNA sequences were analyzed using Chromas Lite ( Technelysium Pty Ltd , South Brisbane , Australia ) . For annotation of bases for equine CXCL16 , the sequence XM_001504756 was used . The NCBI database was used to obtain and compare reference genome predicted-protein sequences of CXCL16 for white rhinoceros ( Ceratotherium simum; XM_004433427 ) , horse ( Equus caballus; XM_001504756 ) , dog ( Canis lupus familiaris; XM_844211 ) , human ( Homo sapiens; AY358909 ) , domestic rat ( Rattus norvegicus; DQ025528 ) , cattle ( Bos taurus; NM_001046095 ) and African elephant ( Loxodonta Africana; XM_003416737 ) . Recombinant plasmids for cloning EqCXCL16 and EqCXCR6 were designed by CLC Main Workbench 7 programs using horse genome sequences obtained as part of this study . Synthetic sequences encoding the predicted exposed part of EqCXCL16 ( aa 25–199 ) [45] and the entire EqCXCR6 were produced by IDT ( Coralville , IA , USA ) . These sequences were designed to have flanking BamHI and XhoI sites . In order to produce recombinant plasmids encoding the full size of S and R versions of CXCL16 , synthetic sequences flanked by PstI and XhoI were also produced . These sequences encoded the N terminus half ( aa 1–143 ) of both versions of EqCXCL16 ( S and R ) . Following digestion with BamHI and XhoI or PstI and XhoI ( Thermo Scientific , Rockford , IL , USA ) , fragments of amplicon and synthetic DNA were separated in E-Gel EX 1% agarose ( Life Technologies ) and extracted from gel using Zymoclean Gel Recovery Kit ( Zymo Research , Irvine , CA , USA ) . Purified fragments encoding aa 17–247 and aa 25–199 of EqCXCL16 and entire EqCXCR6 were ligated into pET15b ( Novagen , Temecula , CA , USA ) followed by transformation into E . coli NovaBlue ( Novagen ) . Ligation and transformation were performed using Rapid DNA Ligation and TransformAid kits ( Thermo Scientific ) , respectively . Recombinant plasmids p15-16A ( aa 17–247 ) , p15-16B ( aa 25–199 ) , and p15-R6 were isolated from ampicillin-resistant clones using ZR Plasmid Miniprep™ Kit ( Zymo Research ) . Plasmids p15-16R and p15-16S encoding R and S versions of full EqCXCL16 were produced by substitution of the smaller PstI/XhoI fragment of p15-16A to synthetic sequences encoding an N terminal half of R and S versions of this protein , respectively . In order to express recombinant polypeptides , plasmids were transformed into E . coli BL21 ( DE3 ) expression vector ( Novagen ) . Several ampicillin-resistant clones were grown overnight in 1 ml of MagicMedia ( Life Technologies ) supplemented with 50 μg/ml of ampicillin ( Sigma-Aldrich ) . Production of polypeptides was confirmed by SDS-PAGE electrophoresis following mini-scale isolation of recombinant proteins using Talon Metal Affinity Resin ( Clontech Laboratories Inc . ) . For large scale production of recombinant EqCXCL16 ( S and R ) and EqCXCR6 , 500 ml cultures of BL21 ( DE3 ) strain of E . coli with respective plasmids were grown overnight at 37°C in MagicMedia supplemented with 50 μg/ml of ampicillin . Following centrifugation at 6000 × g for 15 min , the cell pellet was resuspended in Buffer A ( 50 mM sodium phosphate , 6 M guanidine-HCl , and 300 mM NaCl; pH 7 . 0 ) and subjected to several short cycles of sonication to reduce viscosity . The lysate was centrifuged at 16 , 000 × g for 30 min at 4°C to remove debris . The His-tagged recombinant EqCXCL16 and EqCXCR6 proteins were purified from the supernatant by affinity chromatography using Talon Superflow Metal Affinity Resin ( Clontech Laboratories Inc . ) in combination with an FPLC apparatus ( Amersham Pharmacia Biotech Inc . ) . Columns were equilibrated and washed with Buffer A , while proteins were eluted using Buffer B ( 45 mM sodium phosphate , 5 . 4 mM Gn-HCl , 270 mM NaCl , and 150 mM imidazole; pH 7 . 0 ) . Eluted recombinant protein was dialyzed against PBS using Slide-A-Lyzer Dialysis Cassettes ( Thermo Scientific ) . The purity and integrity of recombinant EqCXCL16S , EqCXCL16R , and EqCXCR6 proteins were evaluated by subjecting them to electrophoresis on a 4–20% gradient gel SDS followed by staining with PageBlue Protein Staining Solution ( Thermo Scientific ) and WB analysis . Protein concentration was determined with BCA Protein Assay ( Thermo Scientific ) using BSA as the standard . Protein-specific rabbit antipeptide sera ( Rb α-EqCXCL16 [rabbit Ab , PA7509] ) , and guinea pig polyclonal antibody ( Gp α-EqCXCL16 pAb ) to detect EqCXCL16 proteins ( EqCXCL16S and EqCXCL16R ) were generated by immunizing rabbits with two synthetic peptides and guinea pigs with recombinant EqCXCL16 expressed in E . coli as previously described [45] . For this study anti-EqCXCR6 peptide antibody was generated by immunizing two rabbits with the 14 amino acid peptide ( amino acid residues 17–30: DSSQEHERFLQFKK ) . These antibodies were extensively characterized by ELISA , confocal microscopy , and WB analysis . The monoclonal antibody ( MAb ) to equine CD3 surface molecule , UC F6G , was kindly provided by Dr . Jeff Stott , University of California , Davis . The R-PE conjugated F ( ab′ ) 2 fragment of goat anti-mouse IgG1 ( Southern Biotech , Birmingham , AL , USA ) was used as the secondary antibody . Mouse EAV α-GP5 and mouse EAV α-nsp-1 monoclonal Abs ( MAb 6D10 and MAb 12A4 , respectively ) have been described previously [70 , 71] . Detection of EAV antigen in infected cells was conducted using Alexa Fluor 488-labeled MAb against nonstructural protein 1 ( nsp1; MAb 12A4 ) [19 , 70] . Goat α-rabbit IgG ( H+L ) -HRP and goat α-mouse IgG ( H+L ) -HRP were purchased from Cell Signaling Technology , Inc . ( Danvers , MA , USA ) . Goat α-guinea pig IgG ( H+L ) -HRP , goat α-rabbit IgG ( H+L ) conjugated to Alexa Fluor 488 , and goat α-guinea pig conjugated to Alexa Fluor 488 were purchased from Life Technologies . Streptavidin conjugated to FITC was purchased from Southern Biotech . Stable HEK-293T transfectants expressing EqCXCL16S ( HEK-EqCXCL16S cells ) and EqCXCL16R ( HEK-EqCXCL16R cells ) were generated as described earlier [45] . Briefly , for expression of the EqCXCL16R protein in eukaryotic cells , a codon-optimized full-length EqCXCL16R sequence ( obtained from whole genome sequencing in this study ) was commercially synthesized and cloned into the pJ609 plasmid , into which the puromycin resistance gene was incorporated , by DNA2 . 0 ( Menlo Park , CA , USA ) . This molecular construct was identified as pJ609-EqCXCL16R and used to transform E . coli DH10B cells ( Life Technologies ) . For the establishment of stable cells , the HEK-293T cells were seeded in 6-well plates ( 2 x 106 cells/well ) and transfected with 3 μg of codon-optimized pJ609-EqCXCL16R plasmid DNA ( DNA2 . 0 ) using lipofectamine 3000 ( Life Technologies ) following the manufacturer’s instructions . At 24 h post transfection , the medium was replaced with fresh medium containing 4 μg/ml of puromycin ( Clontech Laboratories Inc . ) and cells were incubated at 37°C in a 5% CO2 incubator . This process was repeated every other day until only puromycin-resistant colonies remained . These puromycin resistant cells were cloned by limiting dilution in 96-well plates and screened by IFA , after which clones showing the highest level of EqCXCL16S and EqCXCL16R protein expression were frozen in commercial cell-freezing medium ( Recovery Cell Culture Freezing Medium; Life Technologies ) and stored in liquid nitrogen until needed . At every 5th passage and up to the 50th serial passage , cells were analyzed by IFA using Gp α-EqCXCL16 pAb to confirm the expression of EqCXCL16R . All the experiments in HEK-EqCXCL16S and HEK-EqCXCL16R cells were performed within passage levels 5 to 10 . Naïve HEK-293T and stable HEK-EqCXCL16S and HEK-EqCXCL16R cells in 8-well Thermo Scientific Lab-TeK chamber slides were washed in cold phosphate buffered saline ( PBS , pH 7 . 4 ) and fixed in 4% paraformaldehyde ( PFA; Sigma-Aldrich , St Louis , MO , USA ) for 30 min at room temperature ( RT ) . Cells were then stained as described previously [72] . Following fixation , cells were washed 5 times in ice-cold 10 mM glycine ( Sigma-Aldrich ) in PBS , pH 7 . 4 ( PBS-Glycine ) and were then permeabilized with 0 . 2% saponin ( Sigma-Aldrich ) in PBS or left untreated with detergent where examination of surface staining was required . All cells were washed again in 10 mM PBS-glycine and blocked with 5% normal goat serum ( MP Biomedicals , Santa Ana , CA , USA ) for 30 min at RT prior to incubation with specific primary antibodies ( 1:100 dilution ) for 1 h at 37°C in a humidified chamber . After washing in 10mM PBS-glycine , the cells were incubated with anti-mouse or anti-guinea pig IgG ( H+L ) secondary antibodies conjugated with Alexa Fluor 488 ( AF488 , 1:200 dilution ) for 1 h at 37°C in a humidified chamber maintained in total darkness . After washing , slides were mounted in Vectashield mounting medium containing 4′ , 6-diamidino-2-phenylindole ( DAPI; Vector Laboratories , Burlingame , CA , USA ) . The slides were observed either under a Leica TSP SP5 confocal microscope in an environmental chamber at the University of Kentucky imaging core facility or with an inverted fluorescence microscope ( ECLIPSE Ti; Nikon , Melville , NY , USA ) . Cells were lysed in RIPA lysis buffer ( Santa Cruz Biotechnology , Dallas , TX , USA ) in Halt protease and phosphatase inhibitor cocktails ( Thermo Scientific ) . The solubilized proteins were mixed with Pierce lane marker reducing 5X sample buffer containing 100 mM dithiothreitol ( DTT; Thermo Scientific ) and heated for 5 min at 95°C . Samples were resolved in SDS-polyacrylamide gel ( 5% stacking and 12% resolving; Bio-Rad ) at 200 V for 45 min and then transferred onto a PVDF membrane ( Bio-Rad , Hercules , CA , USA ) at 100 V for 1 h using the trans-blot transfer system ( Bio-Rad ) [73 , 74] . The membrane was blocked with 5% non-fat milk powder ( Bio-Rad ) in TBS-T ( 10 mM Tris-HCl [pH 7 . 6] , 150 mM NaCl , and 0 . 1% Tween 20 ) for 1 h at RT and incubated with primary antibodies ( Abs ) : rabbit α-EqCXCL16 PA7509 ( 1:500 ) , guinea pig α-EqCXCL16 ( 1:1000 ) , mouse monoclonal α-EAV GP5 MAb 6D10 ( 1:2000 ) , and mouse monoclonal α-EAV nsp-1Mab ( 12A4 ) . The Abs were diluted in TBS-T with 5% bovine serum albumin ( Sigma-Aldrich ) overnight at 4°C . The following day , the membranes were washed with TBS-T and then incubated with anti-rabbit , anti-mouse , or anti-guinea pig IgG , as appropriate , and conjugated with horseradish peroxidase ( HRP , 1:3000; Cell Signaling Technology , Inc . ) for 1 h at RT . The membranes were washed again and antibody binding was visualized with an ECL-detection system using SuperSignal West Pico chemiluminescent substrate ( Thermo Scientific ) . The cell adhesion assay was performed in accordance with a published protocol [75] . Approximately 1 x 105 HEK-EqCXCL16S , HEK-EqCXCL16R , and naïve HEK-293T cells were plated in a 96-well plate . After 24 h incubation , the cells were washed in 37°C PBS . Cells were then incubated with 0 . 5M EDTA for 10 min at 37°C following which they were washed again with PBS and fixed with 4% PFA for 15 min at RT . Cells were then washed in distilled water and stained with 0 . 1% crystal violet solution for 20 min at RT , visualized with an inverted light microscope and photographed . After washing with distilled water , the cells were air-dried , then incubated with 10% acetic acid for 20 min with shaking prior to the transfer of 50 μl of the lysate to a new 96-well ELISA plate for determination of OD595nm values using a Synergy H1MD microplate reader ( BioTek Instruments Inc . , Winooski , VT , USA ) . Cell migration in response to EqCXCL16S and EqCXCL16R proteins was determined using a Chemotaxis assay kit ( Cell Biolabs , Inc . , San Diego , CA , USA ) . Briefly , FPLC purified soluble recombinant EqCXCL16S and EqCXCL16R proteins were diluted in serum-free RPMI medium with 0 . 5% BSA ( cell culture grade ) at a concentration of 2 μg/ml in 500 μl of RPMI and were added to the lower well of a 24-well chemotaxis chamber; the lower and the upper wells were separated by a polyvinylpyrrolidone-free polycarbonate filter insert with the pore size of 3 μm . To the upper wells of the chamber , 100 μl of purified equine CD3+ T lymphocytes ( 5 x 105 cells/ml ) labelled with Calcein-AM ( Life Technologies ) was added . After 6 h of incubation at 37°C in 5% CO2 , media from the inside of the insert was removed and insert was placed in a clean well with 400 μl of Cell Detachment solution and incubated at 37°C for 30 min . After complete dislodging of the cells from the underside of the insert , 400 μl of the cells containing Cell Detachment solution was mixed with 400 μl of the migratory cells from the original well . Then 50 μl of the mixed cell solution was added onto a glass slide , and cells that passed through the filter were counted under a fluorescent microscope ( 40× objective ) and represented in a bar diagram ( averages of six different fields were included ) . Stable HEK-293T cells expressing EqCXCL16S and EqCXCL16R and naïve HEK-293T cells were seeded in 12-well plates . At about 85% confluency , cells were washed with warm ( 37°C ) PBS ( pH 7 . 4 ) and replenished with warm ( 37°C ) growth medium containing Dil-OxLDL at 10 μg/ml ( Kalen Biomedical LLC , Montgomery Village , MD , USA ) and incubated at 37°C for 3 h . Cells were washed with ice-cold PBS and examined under an inverted epifluorescence microscope for the uptake of Dil-OxLDL by cells . Approximately 100 μg of total protein lysate from naïve HEK-293T cells or stable HEK-EqCXCL16 cells were separated in 12% SDS-PAGE and transferred onto a PVDF membrane for Far-WB analysis following a modification of the published protocol [76] . The bound proteins were then denatured and gradually renatured on the membrane by sequential incubation with 6 M , 3 M , 1 M , and 0 . 1 M Gn-HCl in freshly prepared AC buffer ( 100 mM NaCl , 20 mM TRIS [pH 7 . 5] , 10% glycerol , 0 . 5 mM EDTA , 0 . 1% Tween-20 , 2% non-fat dry milk , and 5 mM DTT ) for 30 min at RT or with only AC buffer in the absence of Gn-HCl overnight at 4°C . The membrane was blocked with 5% non-fat dry milk in TBS-T ( 0 . 1% Tween-20 ) and overlaid with purified EAV VBS ( 15 μg/ml ) and incubated overnight at 4°C . The next day , the membrane was washed vigorously ( 3 washes each of 10 min ) and incubated with mouse monoclonal Ab ( α-GP5; MAb 6D10 ) directed against EAV GP5 envelope glycoprotein . Monoclonal antibody binding was detected by the ECL-detection system using SuperSignal West Pico chemiluminescent substrate ( Thermo Scientific ) . To study EqCXCL16-EqCXCR6 interaction in a separate Far-Western blot experiment , 20 μg of purified recombinant hemagglutinin ( HA ) -tagged EqCXCR6 was separated on 12% SDS-PAGE and transferred onto a PVDF membrane which was incubated with recombinant purified EqCXCL16S or EqCXCL16R proteins ( 5 μg/ml ) . The membrane was then developed using Rb-anti EqCXCL16 antibody . EAV VBS was purified by ultracentrifugation ( 121 , 600 × g for 4 h ) through a 20% sucrose cushion and protein concentration determined using the BCA protein assay kit . About 2 mg of purified EAV was biotinylated using EZ-Link Sulfo-NHS-Biotin ( Thermo Scientific ) following the manufacturer’s protocol . Excess unbound biotin was removed by filtering through a Zeba desalt spin column ( MWCO 7000; Thermo Scientific ) equilibrated in PBS ( pH 7 . 4 ) . The naïve HEK-293T and HEK-EqCXCL16 cells were washed in cold PBS ( pH 7 . 4 ) and removed from the culture dish using a non-enzymatic cell dissociation solution ( Cellstripper; Mediatech Inc . ) . Cells were resuspended in cold PBS ( pH 7 . 4 ) containing 2% FBS ( PBS-F ) , centrifuged at 1000 × g for 5 min at 4°C and incubated with biotinylated EAV at an MOI of 100 on ice for 2 h in total darkness . Excess EAV was removed by washing three times in cold PBS-F . Subsequently , the cells were stained with Streptavidin-FITC ( 1:100 ) and incubated at 4°C for 30 min in total darkness . Cells were washed in PBS-F at 1000 × g for 5 min at 4°C , transferred onto glass microscope slides using a Shandon CytoSpin III Cytocentrifuge with Shandon single cytofunnel with white filter cards ( Thermo Scientific ) and incubated with DAPI solution to permit visualization of cell nuclei . Cells were then analyzed using a Nikon inverted fluorescence microscope and the percentage of cells bound to EAV was calculated . Statistical tests for association of CXCL16 genotypes with phenotypes , including susceptibility and resistance as well as carrier status , were conducted using the Fisher’s Exact Test . Differences among multiple treatment groups were analyzed by statistical analysis software Sigmaplot 12 . 3 ( SystatSoftware Inc . , San Jose , CA , USA ) , by ANOVA with pairwise multiple comparison procedures by the Holm-Sidak method . P-values less than 0 . 05 were considered to be statistically significant .
|
A variable proportion of EAV infected stallions ( 10–70% ) may become persistently infected and continuously shed the virus exclusively in their semen after recovery from acute infection . Previous studies in our laboratory have shown that stallions with the CD3+ T lymphocyte susceptibility phenotype to in vitro EAV infection are at higher risk of becoming persistently infected carriers compared to those that lack this phenotype . Here genetic and experimental studies were used to demonstrate that CXCL16 in the horse codes for two proteins , one associated with resistance and the other associated with susceptibility of CD3+ T lymphocytes to EAV infection . The two proteins are the result of four nucleotide substitutions in exon 1 of the equine CXCL16 gene . These alleles determine the outcome of in vitro infection of CD3+ T lymphocytes with EAV and are strongly associated with the establishment and maintenance of long-term carrier state in stallions . In vitro studies demonstrated that one form of CXCL16 protein ( CXCL16S ) is one of the cellular receptors for EAV and has higher scavenger activity and adhesion ability as compared to the form of the protein associated with resistance ( CXCL16R ) .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"sequencing",
"techniques",
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"body",
"fluids",
"immunology",
"vertebrates",
"animals",
"mammals",
"membrane",
"proteins",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"white",
"blood",
"cells",
"semen",
"animal",
"cells",
"proteins",
"t",
"cells",
"recombinant",
"proteins",
"molecular",
"biology",
"cell",
"membranes",
"chemotaxis",
"biochemistry",
"cell",
"biology",
"anatomy",
"horses",
"physiology",
"chemokines",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"equines",
"amniotes",
"organisms"
] |
2016
|
Allelic Variation in CXCL16 Determines CD3+ T Lymphocyte Susceptibility to Equine Arteritis Virus Infection and Establishment of Long-Term Carrier State in the Stallion
|
The X chromosome constitutes a unique genomic environment because it is present in one copy in males , but two copies in females . This simple fact has motivated several theoretical predictions with respect to how standing genetic variation on the X chromosome should differ from the autosomes . Unmasked expression of deleterious mutations in males and a lower census size are expected to reduce variation , while allelic variants with sexually antagonistic effects , and potentially those with a sex-specific effect , could accumulate on the X chromosome and contribute to increased genetic variation . In addition , incomplete dosage compensation of the X chromosome could potentially dampen the male-specific effects of random mutations , and promote the accumulation of X-linked alleles with sexually dimorphic phenotypic effects . Here we test both the amount and the type of genetic variation on the X chromosome within a population of Drosophila melanogaster , by comparing the proportion of X linked and autosomal trans-regulatory SNPs with a sexually concordant and discordant effect on gene expression . We find that the X chromosome is depleted for SNPs with a sexually concordant effect , but hosts comparatively more SNPs with a sexually discordant effect . Interestingly , the contrasting results for SNPs with sexually concordant and discordant effects are driven by SNPs with a larger influence on expression in females than expression in males . Furthermore , the distribution of these SNPs is shifted towards regions where dosage compensation is predicted to be less complete . These results suggest that intrinsic properties of dosage compensation influence either the accumulation of different types of trans-factors and/or their propensity to accumulate mutations . Our findings document a potential mechanistic basis for sex-specific genetic variation , and identify the X as a reservoir for sexually dimorphic phenotypic variation . These results have general implications for X chromosome evolution , as well as the genetic basis of sex-specific evolutionary change .
When the X chromosome stops recombining with the Y chromosome , a unique genomic environment is formed . Several important population genetic parameters are presumably affected , and collectively this sets the scene for different patterns of evolution on the X chromosome [1–3] . In particular , the level of standing genetic variation on the X chromosome is expected to differ from the autosomes , however , depending upon the precise conditions , either an increase or a decrease is expected . Classic population genetic theory predicts that the X chromosome will host lower levels of standing genetic variation than the autosomes ( e . g . [4–7] ) . The lower census size of the X chromosome ( 3/4 of the autosomes ) should result in a lower effective population size and thus a higher rate of genetic drift , while hemizygosity in males results in unconditional expression of deleterious mutations . Together , these two features are expected to remove both neutral and deleterious allelic variants at a higher rate on the X chromosome and result in reduced standing genetic variation on the X chromosome relative to the autosomes [6] . Despite these predictions there are nevertheless a range of theoretical arguments why the difference in standing genetic variation between the X and the autosomes could be reduced , or even reversed . In species where neither the X chromosome nor the autosomes recombine in males ( e . g . Drosophila ) , the rate of recombination could be lower on the autosomes . Lower autosomal rates of recombination would , because of Hill-Robertson interference , result in a larger reduction in the effective population size of the autosomes , and consequently a larger reduction in genetic variation [8–11] . Another factor which reduces the effective population size disproportionately on the autosomes is sexual selection on males [12] . Because the X chromosome spends 2/3 of its time in females it is less affected by higher variance in male fitness . In extreme cases sexual selection can theoretically cause the effective population size of the X chromosome to surpass that of the autosomes [12] . Sexually antagonistic selection has also been suggested to increase standing genetic variation on the X chromosome . Theory predicts that hemizygosity of the X chromosome in males , and its presence in females 2/3 of its time , allow for a wider parameter space where allelic variants with sexually opposing effects on fitness can be maintained in polymorphisms [13–15] ( but see [16] ) . The X chromosome has also been suggested to host a disproportionately large fraction of the genome which contributes to sexual dimorphism [14 , 17–19] . Sexual dimorphism requires elements with sex-specific or sex-limited effects and , because deleterious mutations in such elements are primarily selected in one sex [20] , they are predicted to host higher levels of standing genetic variation at mutation-selection-drift balance [21] . However , more recent theory predicts that sexual dimorphism may develop more easily on the autosomes [22] . A further possibility that may affect levels of standing genetic variation on the X chromosome is the way in which dosage compensation ( DC ) influences the effect size of mutations . Most studies assume an equal effect of mutations on the X and the autosomes [6 , 7 , 23 , 24] , but this may be violated when DC is incomplete [6] . Incomplete DC may be more widespread than previously assumed [25] . For example , in mammals , substantial parts of the X chromosome escape DC [26 , 27] and in several species with ZW sex chromosomes , DC may not occur at all [25 , 28] . Under incomplete DC , mutations may have smaller phenotypic effects on the X or Z chromosome in the hemizygous sex [6 , 29] . Deleterious mutations in regions of incomplete DC could therefore experience weaker net purifying selection than mutations in dosage compensated regions . As a consequence , regions of incomplete DC could host more genetic variation . Genome-wide standing genetic variation has been most extensively studied empirically in Drosophila melanogaster . Evidence from this species points to reduced sequence ( e . g . [2 , 30 , 31] ) and transcriptional variation [32] on the X chromosome . In contrast to this , the X chromosome shows no reduction in genetic variation for a range of phenotypic traits [33 , 34] and seems enriched for sexually antagonistic genetic variation for fitness [35 , 36] . With respect to sex-bias , genes with female-biased expression are enriched on the X chromosome , while those with male-bias are depleted [3 , 37 , 38] ( but see [39] ) , indicating that at least some types of sexual dimorphism develop more easily on the X [1 , 40] . A number of studies have also compared the distribution of genetic variation in sexual dimorphism over the X and the autosomes for a range of phenotypic traits . The results from these studies are mixed ( reviewed in [3 , 41] ) . Sex-specific transcriptional variation [42] and general sex-specific genetic variation [43] do however appear enriched on the X chromosome , while the number of eQTLs with a male-specific effect is reduced [44] . In summary there are theoretical as well as empirical reasons to expect depletion as well as enrichment of standing genetic variation on the X chromosome . These predictions depend on the type of mutations and potentially also their location along the X chromosome . To test these predictions we contrast the genomic distribution of trans-acting SNPs which associate with sexually concordant and sexually discordant standing genetic variation in gene expression , in a population of D . melanogaster . We focus on trans-regulation because this covers a SNP’s impact on variation within , as well as between , chromosomes . Our results show that SNPs with a sexually concordant effect are significantly depleted on the X chromosome . However , SNPs with a sexually discordant effect are enriched on the X chromosome compared to SNPs with a sexually concordant effect . Furthermore , we show that this relative enrichment of sexually discordant SNPs is driven by SNPs with a female-biased effect and these SNPs tend to accumulate away from regions where DC is initiated and predicted to be strongest . Our results suggest that contrasting patterns of standing genetic variation on the X and autosomes , while influenced by the factors discussed above , might also depend on patterns of DC along the X chromosome .
The X chromosome contains ~18 . 8% of the D . melanogaster genome [30] , but only ~13 . 7% of the SNPs . We use these benchmarks to test if the X chromosome is enriched or depleted of trans-regulatory SNPs . We find that SCV SNPs are depleted on the X chromosome with respect to the proportion of SNPs on the X ( and thus also with respect to the size of the X ) ( Table 1 , Fig . 1 ) . The proportion of SDV SNPs does not differ from 0 . 137 ( Table 1 , Fig . 1 ) . This also applies to the two categories of SDV SNPs ( SDV . M and SDV . F ) , apart from intergenic SDV . M SNPs which are depleted ( Table 1 , Fig . 1b ) . While the 95% confidence interval of most classes of SDV . F SNPs clearly overlaps with a proportion of 0 . 188 , this is not the case for any SDV . M SNPs classes ( Table 1 ) . When comparing the proportions of SNPs associated with SCV and SDV genes , we find that SDV SNPs are more common on the X chromosome than SCV SNPs ( Fig . 1a , Table 1 ) , a result which holds when we relax and strengthen the gene selection criteria ( S2 Table ) . This pattern is primarily driven by genic SNPs ( Fig . 1b , c , Table 1 ) located in introns ( Table 1 ) . To test for an enrichment of SDV SNPs with trans-acting effects , mediated through cis-regulation of X-linked genes , we restricted the intergenic class of SNPs to only include SNPs located 500bp before , or after , a transcription start , or end , site ( since most cis-regulatory SNPs are found within this region in D . melanogaster [44] ) . We found no evidence for such an enrichment ( P = 0 . 76 ) . When we divide the SDV SNPs into SDV . F and SDV . M SNPs , and compare these proportions against the proportion of SCV SNPs , we find that the relative enrichment of SDV SNPs compared to SCV SNPs on the X is mainly driven by SDV . F SNPs ( Fig . 1 , Table 1 ) . SDV . M SNPs were not significantly more common on the X , but the trend was in the same direction as for female-biased SNPs . The relative effect size of X-linked to autosomal SNPs in general did not differ between SDV and SCV SNPs ( S3 Table ) . Dosage compensation in D . melanogaster is initiated at a large number of high affinity sites ( HAS ) on the X-chromosome , to which the dosage compensation complex ( DCC ) binds and from where it spreads along the X-chromosome [45] . Dosage compensation of the male X chromosome is therefore expected to be more complete closer to HAS , than further away . Accordingly , we predicted that SDV . F SNPs should be less likely to be found in close proximity to HAS . Using information on the position of HAS from Straub et al [46] we indeed found that SDV . F SNPs lie further from HAS than SDV . M SNPs ( Fig . 2 , S4 Table ) . SDV . F SNPs were also located further away from HAS than SCV SNPs ( S5 Table ) . The distribution of SDV . M SNPs , with respect to location of HAS , did not significantly depart from that of SCV SNPs ( S5 Table ) . SDV . F SNPs seem to accumulate at a moderate distance to HAS ( Fig . 2 ) . HAS only function as docking platforms for the DCC , while the means by which the DCC is believed to increase transcription rate at the male X-chromosome is through acetylation of histone H4 at lysine 16 ( H4K16ac ) [45] . We find complementary evidence that SDV . F SNPs are preferentially located away from regions where DC is predicted to occur when we analyse locations of SNPs with respect to enrichment scores of H4K16ac ( S6 Table , S7 Table ) . Because DC may not operate in the testis [47] we also removed all gonad specific SDV and SCV genes from the analyses , identified as genes with the highest expression in the testis or the ovaries [48] . Using this subset we still see an over-representation of X-linked SDV SNPs ( P = 0 . 018 ) . When breaking the SDV SNPs down into SDV . F and SDV . M SNPs , we again see an enrichment of SDV . F SNPs on the X ( P = 0 . 0076 ) , but not for SDV . M SNPs ( P = 0 . 86 ) , in line with our previous analyses . In addition , using this subset of genes , SDV . F SNPs are , again , concentrated away from HAS ( S3 Fig . ) . We next looked at the relationship between the sex-bias of X-linked genes and their distance from HAS , to see if the patterns we find for SDV SNPs are consistent with the sex-bias of X-linked genes . When using whole fly expression data from the DGRP lines we find results consistent with a previous study [49] , which showed that X-linked genes become more male-biased with increasing distance to HAS ( slope = 1 . 12 × 10–6 , P < 0 . 0001 ) . We also find no significant difference in the sex-bias of genes containing intronic SCV SNPs , SDV SNPs with female-biased ( SDV . F ) or male-biased ( SDV . M ) effect size ( median sex-biased expression of X-linked gene log2 ( M/F ) : SDV . M = -0 . 016 , SDV . F = 0 . 012 , SCV = -0 . 016; Wilcoxon test: SCV-SDV . M P = 0 . 757 , SCV-SDV . F P = 0 . 083 ) . These results suggest that SDV SNPs with a female-biased effect are not necessarily located in regions with female-biased expression in whole flies . However , Vesenko and Stone [50] recently pointed out that it is problematic to test for an association between sex-biased expression and distance to HAS , with reference to DC , using whole fly expression data . They showed that the positive association disappears when genes with highest expression in the testis , a tissue where DC may not operate [47] , are removed . We also confirm this lack of association when testis-biased genes are removed from the whole fly expression data we use ( P = 0 . 069 ) . This result highlights the fact that sex-biased expression is dependent on tissue [51 , 52] . In our data we have no information on which tissue ( s ) each of the SDV genes show variation for sex-biased expression in . To limit our analyses to one tissue we repeated the analyses using gene expression in the brain , a shared tissue for which expression data from both males and females is available [53] . Previous analyses of brain [53] and head [54] gene expression data have , in contrary to whole fly gene expression analyses , shown that male-biased genes cluster closer to HAS . In accordance , when we look at the sex-bias of X-linked genes containing SDV and SCV SNPs , we find that genes with intronic SDV . F SNPs are significantly more female-biased than genes with intronic SCV SNPs ( median log2 ( M/F ) : SDV . F = -0 . 153 , SCV = -0 . 100; Wilcoxon test: P = 0 . 0117 ) . We find no difference in sex-bias between genes with intronic SDV . M SNPs and genes with intronic SCV SNPs ( median log2 ( M/F ) : SDV . M = -0 . 113 , SCV = -0 . 100; Wilcoxon test: P = 0 . 241 ) . Furthermore , we find that X-linked genes containing intronic SDV SNPs become more female-biased with increasing distance to HAS ( slope = -1 . 13 × 10–6 , P = 0 . 009 ) . If incomplete DC explains the enrichment of trans-acting SNPs with a female-biased effect size on the X ( compared to SCV SNPs ) , we would expect to see a relative decline in SNP effect size in males compared to females with distance to HAS . To test this we separately calculated the regression coefficient between effect size and distance to HAS for each SDV gene , using only significant SDV . M or SDV . F SNPs . In contrast to predictions , we do not find this pattern ( median slope: SDV . M SNPs = 1 . 52 × 10–7; SDV . F SNPs = -5 . 14 × 10–8; Wilcoxon test: P = 0 . 262 ) . Under mutation-selection balance , and the assumption that the deleterious effect of a mutation scales with its effect size , a negative association between effect size and minor allele frequency ( MAF ) is expected [30 , 55] . We found the correlation between effect size and MAF to be negative for SCV , SDV . M , as well as SDV . F SNPs ( Table 2 ) . The correlation was similar in all three cases , but significantly less negative for SDV . M SNPs . The predominant view is that all sex-specific gene expression in D . melanogaster is regulated by sex-specific transcription factors at the terminal end of the sex determination pathway , which interact with cis-regulatory elements [56 , 57] . We therefore investigated whether expression variation in SDV genes is associated with sequence variation in genes within the sex determination pathway . Of the genes in , and associated with , the sex determination pathway ( Sxl , tra , dsx , fru , tra-2 , ix and her ) only Sxl is located on the X chromosome . Two SDV and two SCV genes had one SNP each in this gene . One SNP is associated with both of the SCV genes and one of the SDV genes . Of the other genes in the sex determination pathway , only tra , dsx and fru had SNPs associated with them . The number of SDV and SCV genes with associated SNPs located in genes in the sex determination pathway was very similar ( χ2 = 0 . 6757 , d . f . = 1; P = 0 . 4111 ) . These findings suggest that the sex-specific gene expression variation we observe in our SDV genes is not a result from sequence variation within the sex determination pathway . To test if the higher proportion of X-linked SDV SNPs with female-biased effect size is driven by a small set of SNPs , each controlling the expression of many SDV genes , or if X-linkage is a more general phenomenon , we investigated whether the number of genes associated with female-biased SNPs was greater than expected by chance . We permuted the number of times a SNP was associated with SDV genes , and calculated the 95% confidence limits of the maximum . We find that three SNPs are associated with more genes than expected by chance ( X: 1617959; X: 18634687; X: 18634824 associates with 10 , 12 , 12 genes respectively; 95% CI of expected maximum distribution: 6–8 genes ) . These three SNPs are all intergenic and lie between genes CG3795 and Scgdelta ( X: 1617959 ) , or CG6873 and CG12609 ( X: 18634687 , X: 18634824 ) . Removing these three SNPs from our analyses does not change the general conclusion that there is a relative excess of SDV SNPs compared to SCV SNPs on the X chromosome ( all SDV SNPs vs SCV SNPs; Wilcoxon test P = 0 . 0010 ) .
Classic population genetic theory predicts that the X chromosome should be depleted of genetic variation compared to the autosomes , due to a combination of enhanced selection against recessive deleterious mutations and reduced effective population size [4–7] . In accordance with this prediction we find that the X chromosome is depleted of trans-acting SNPs with a sexually concordant effect , and that our genes with high transcriptional genetic variation are underrepresented on the X chromosome [32] . However , when we compare SNPs with sexually concordant and sexually discordant trans-effects , we observe that the latter are substantially more frequent on the X chromosome and not necessarily depleted . This in particular concerns SNPs with a female-biased effect size , which , depending on site class , are 50–98% more common than SNPs associated with sexually concordant variation . We explore several possible explanations for this pattern . The relative enrichment of trans-acting SNPs with a female-biased effect on the X chromosome compared to SCV SNPs could be interpreted to result from resolved intralocus sexual conflict . Sexual antagonism over expression level is expected to occur for genes on the X and the autosomes alike , but theory shows that the invasion criteria for alleles with a beneficial effect in one sex , and a detrimental effect in the other , are more relaxed on the X chromosome [14] ( but see [16] ) . Accordingly it has been predicted that the X chromosome should be enriched with sexually antagonistic variation [14] , which also has been confirmed in D . melanogaster [35 , 36 , 58] . It has also been predicted that enrichment of sexually antagonistic variation would be followed by enrichment of sex-biased genes , as modifiers of gene expression should evolve to reduce expression in the disfavoured sex , which would allow for the sexually antagonistic allele to fix [14] . In line with these predictions , studies have shown that the X chromosome is enriched with female-biased genes in Drosophila [3 , 37 , 38] . The same studies do , however , show that enrichment does not apply to male-biased genes . Since theory predicts that sexually antagonistic dominant female beneficial and recessive male beneficial alleles can invade the X chromosome [14] , it has been suggested that enrichment of exclusively female-biased genes supports theory , provided that beneficial mutations are in general dominant [49] . Our results can be interpreted in light of this scenario . Dominant female beneficial male detrimental trans-acting alleles accumulate on the X chromosome . These trans-factors are subsequently followed by modifiers which reduce their impact in males . Mutations in such trans-factors will now have a female-biased effect , and because the female-biased trans-factors are enriched on the X chromosome , so will SNPs with female-biased effect . Our finding , that trans-SNPs with a female-biased effect size locate away from HAS and away from regions with high intensity of acetylation at H4K16 sites ( through which the DCC is believed to achieve increased transcription in males ) , does however point to a different explanation as to why SNPs with a female-biased effect accumulate on the X chromosome . These results suggest that there are regions of the X chromosome where DC is incomplete , causing female-biased expression , and mutations to have a female-biased effect size . Although our data clearly show that the SNPs with a female-biased effect avoid regions where DC is predicted to be strongest , there are nevertheless a couple of potential caveats with this explanation . In whole fly samples , which we study here , gene expression in general becomes more male-biased , and not female-biased , further away from HAS [49] . This positive association does however disappear when genes with the highest expression in the testis , where DC probably does not occur [47] , are removed [50] . This highlights the fact that sex-biased expression , and thus also genetic variation in sex-biased expression , depends on tissue [51 , 52] . Since our selected genes were identified from whole fly expression data , we have no information on which tissue ( s ) they show variation in . Looking at the sex-bias of genes in whole fly samples may therefore not accurately reflect what is relevant for our SNPs . When we look at gene expression in a single tissue ( brain ) , which is shared between the sexes and for which there is available data , we do find that intronic female-biased SNPs are more likely associated with female-biased genes than SNPs with a sexually concordant effect . Furthermore the sex-bias of these genes becomes more female-biased with increasing distance to HAS , corroborating a lack of complete DC away from HAS . A second potential problem for the incomplete DC hypothesis is that it relies on the male effect size of SNPs declining with distance to HAS relative to the female effect size . This is a pattern we do not see in our data , but there are at least two explanations which could obscure a decline with distance to HAS . First the effect size of a SNP is measured across all tissues in our whole body samples , and it may therefore not accurately reflect the true effect size in the tissue ( s ) in which it has its effect , and second , GWAS studies are biased towards finding SNPs with large effect size , preventing SNPs with smaller effect size to influence the association with distance to HAS . With these potential caveats in mind , we next explore what incomplete DC may entail for the evolution of the X chromosome from the perspective of our data . The most feasible explanation for why we see an enrichment of trans-SNPs with female-biased effect size on the X chromosome is that incomplete DC provides genomic regions where expression is naturally female-biased . This results in relatively more X-linked trans-factors , and associated mutations , with female-biased effect . Under this scenario the female-bias of trans-factors is not adaptive , but simply a consequence of incomplete DC . In addition , incomplete DC is also expected to increase the proportion of X-linked SNPs with female-biased effect size , through reduced intensity of net purifying selection , mediated through a reduced effect size of mutations in males . It has previously been shown that expression of most genes evolves under stabilizing selection [59] , and it therefore seems reasonable to assume that this is also the case for trans-factors . The negative correlation we observe for all classes of trans-SNPs , between effect size and MAF , also supports this view . Assuming that the strength of selection ( s ) against a deleterious mutation is halved in males , when it is located in a region without DC , the equilibrium frequency of a mutation at mutation-selection balance is given by 3u/ ( s/2+2hs ) in a non-dosage compensated region , and 3u/ ( s+2hs ) in a dosage compensated region [10] ( where u is the mutation rate and h is the dominance factor ) . Since the equilibrium frequency is always higher in regions of incomplete DC , this implies that relatively more SNPs with a female-biased effect size , than those with an equal effect size in both sexes , will segregate at the X chromosome , all else being equal . On the autosomes , the frequency at mutation-selection balance is given by u/sh , which is always higher than the equilibrium frequency for mutations at dosage compensated regions , and when h < 0 . 5 for mutations in regions without DC . Reduced net purifying selection in regions with incomplete DC thus acts to increase genetic variation on the X chromosome . Given that h < 0 . 5 , which is a reasonable assumption since most deleterious mutations are recessive [60 , 61] , the genetic variation in non-dosage compensated regions is not expected to surpass the genetic variation at the autosomes , which is also what we observe . This simplified scenario assumes complete lack of DC , which is probably not the case , but it provides a framework for how to understand how incomplete DC may influence the relative amounts of genetic variation on the autosomes and at different regions of the X chromosome . Theoretical models concerning standing genetic variation and the rate of evolution of the X chromosome relative to the autosomes primarily focus on dominance [4–7 , 62] and effective population size [1 , 63] . However , the possibility that the strength of selection may differ for some classes of mutations between the X chromosome and the autosomes has , to our knowledge , not been empirically considered [64] . Reduced efficacy of selection will , apart from allowing for a higher frequency of deleterious mutations , also result in more frequent fixation of deleterious mutations . Our findings may therefore have relevance for the relative evolutionary rate of the X and the autosomes . Some studies addressing the faster X hypothesis have made efforts to control for gene content ( e . g . [65 , 66] ) , but also taking incomplete DC into account may be important for a full understanding of the intrinsic differences between mutational effects on the X and the autosomes [1] . An intriguing but speculative possibility is that trans-factors have accumulated in regions with incomplete DC , because the natural female-bias in these regions helps them resolve intralocus sexual conflict over gene expression elsewhere in the genome . While the established view suggests that sex , and all somatic sexual differentiation , is exclusively regulated by sex-specific transcription factors or hormones at the terminal end of the sex-determining pathway initiated by one sole master sex-switch gene [56 , 57] , an emerging view suggests that chromosome karyotype also controls sex-differences [54 , 67–73] . The exact mechanism by which this occurs is not yet understood , but trans-factors acting independently from the sex determining pathway , placed in regions with incomplete DC as our findings support , offers an exciting possibility [74] . Interestingly , trans-factors in regions of incomplete dosage compensation could use the exact same simple mechanism as the master sex-switch gene does in mammals , flies and nematodes , to initiate different signalling cascades in the sexes . These genes are located on one of the sex chromosomes [56 , 75–77] , where they trigger sex differences by a dose effect ( presence in one vs versus zero copies if it is Y linked and one vs versus two copies if it is X linked ) . This same mechanism would allow trans-acting elements to produce female- as well as male-biased gene expression of target genes , as higher expression of an X-linked activator ( repressor ) of gene expression will cause female- ( male- ) biased expression ( Fig . 3 , see also figure 10 in [29] ) . Genome-wide gene expression studies comparing expression in males and females of different Drosophila species , show that gene expression changes more rapidly in males compared to females [38 , 78 , 79] . This phenomenon is probably also true for many other species , because males are often exposed to sexual selection both when interacting with females and when competing with males , while females are exposed to sexual selection primarily when interacting with males . It thus seems as if males , in general , could benefit more than females from an additional mechanism to control gene expression ( males for instance have more cis-regulatory elements than females [44 , 80] ) , particularly in species locked into a perpetual arms race between the sexes and male phenotypes . The potential mechanism we describe here instead gives females more opportunity to change than males . However , strong sexual selection on males should often displace the female phenotype from its optimum . The mechanism described here may enable females to maintain their optimum with only minor effects on the male phenotype . In species where males are the homogametic sex ( as opposed to the species studied here ) there is the opportunity for males to use this mechanism directly , and we would therefore predict a larger Z-linked trans-regulatory effect on sex-biased gene expression compared to that observed here . Regulation of sex differences through trans-factors located in regions without DC could potentially be part of the explanation as to why we see less DC in ZW sex chromosome systems [81] . If the Z chromosome accumulates trans-acting factors which influence gene expression in the direction favoured by males , at the time the W chromosome degrades and loses its gene content , this may reduce selection on females to develop DC in these systems . The X-linked SNPs we find associated with female-biased transcriptional variation are located in intergenic , and specifically intronic , regions . This suggests that the effect we observe is caused by non-coding trans-regulatory factors . The view which has emerged over the last few years suggests that these regions , previously assumed to be inert , are host to a range of non-coding RNAs ( ncRNA ) with gene regulatory functions [82] . Studies have also shown that introns are particularly enriched with ncRNAs . These studies have foremost been conducted on humans and mice [83–86] , but evidence from species of other taxa , such as Xenopus tropicalis [87] Caenorhabditis elegans [88 , 89] and D . melanogaster [88–90] , is also accumulating . However , function has only been verified for a small fraction of all non-coding transcripts . Interestingly , we note that two recent studies on mouse have found ncRNAs with female biased expression , which localize to genomic regions which escape X chromosome inactivation [91 , 92] . Much of what we discuss here relies on the interpretation that trans-SNPs with a female-biased effect size are enriched in regions where the DC machinery appears less active , because this should cause trans-factors , and thus SNPs within these factors , to have a female-biased effect size . If a trans-factor is not female-biased in itself we find it difficult to understand how a mutation can have a female-biased effect . We would nevertheless like to raise the possibility that some idiosyncrasy of DC , that we do not yet understand , is the underlying cause to the intriguing patterns we observe . In summary we find that trans-SNPs with a sexually concordant effect on gene expression are depleted on the X chromosome . We interpret this to primarily result from an exposure of recessive deleterious mutations in males and a reduced effective population size of the X chromosome . The pattern we observe for SCV trans-SNPs is in large mirrored by SDV . M SNPs . With respect to SDV . F SNPs we do however find a striking difference , as their genomic distribution is significantly shifted towards that expected on the X chromosome , compared to SCV trans-SNPs . This implies that mutations with a female-biased effect size are governed by a population genetics parameter setting different from that of the other types of trans-SNPs and/or the presence of other idiosyncrasies of the X chromosome which facilitates their accumulation . Our finding that SDV . F SNPs are enriched in regions where the DC machinery appears less involved opens the possibility for differences between the X and the autosomes that has not been discussed before . Most feasibly the enrichment in regions with predicted incomplete DC is a simple consequence of such regions providing an environment where female-biased expression naturally occurs . In addition , the lower effect size of deleterious mutations in males may reduce the net strength of purifying selection and allow a higher equilibrium frequency of mutations at mutation-selection-drift balance . On a more speculative note , regions with incomplete DC can constitute a platform where trans-factors accumulate to directly modify gene expression in a sex-specific manner . This possibility hints that a lack of complete DC should not always be viewed as a problem , but also as an opportunity for the genome to resolve intralocus sexual conflict over gene expression . In closing , we note that the X chromosome in general is depleted of trans-acting SNPs with a sexually concordant effect , but that mutations with a trans-acting female-biased effect either occur more frequently on the X chromosome or have the capacity to counterbalance the factors which mediate erosion of genetic variation on the X chromosome .
We used whole body microarray data from 40 inbred lines of D . melanogaster ( the DGRP lines from the Raleigh population ) from the study by Ayroles et al [32] . The raw data were downloaded from http://www . ebi . ac . uk/arrayexpress/experiments/E-MEXP-1594 and normalized using RMA [93] . SNP data for these lines was taken from Mackay et al [30] . We identified two classes of genes: those with genetic variation for sexually discordant ( SDV genes ) and those with sexually concordant genetic variation ( SCV genes ) . To accomplish this we first fit a linear mixed model using Restricted Maximum Likelihood ( REML ) of gene expression levels independently for all genes , specifying Sex ( fixed factor ) , Line ( random factor ) and Sex × Line ( random factor ) as predictors . From these models we extracted the Line and the Line × Sex variance components to create a variation index ( I ) , that measures the percentage contribution of the Sex × Line variance to the total genetic variance [I = VSex × Line / ( VLine + VSex × Line ) ] . We then classified genes as SDV or SCV according to the following characteristics: Genes with an I > 0 . 95 and a large ( > 0 . 2 ) and significant ( P < 0 . 0001 ) Sex × Line variance component were classified as SDV , and genes with an I < 0 . 05 and a large ( > 0 . 2 ) and significant ( P < 0 . 0001 ) Line variance component were classified as SCV . Using this procedure , 121 genes were classified as SDV and 152 genes as SCV ( S1 Fig . ; S1 Table ) . To verify that our procedure captured genes with sexually discordant and concordant variation we calculated the intersexual genetic correlation rMF for SDV and SCV genes . SDV genes , on average ( ± standard deviation ) , had a low rMF ( -0 . 03 ± 0 . 15 ) , whereas SCV genes had a high average rMF ( 0 . 93 ± 0 . 07 ) . 98% ( mean abs[log2{male/female}] expression ± SD is 0 . 63 ± 0 . 39 ) of the SDV genes and 49% ( 0 . 07 ± 0 . 06 ) of the SCV genes had sex-biased expression at a P-value < 0 . 05 . For each SDV and SCV gene , mean gene expression for each Line was calculated independently for each sex . Gene expression values ( G . E ) for each gene were uploaded onto the DGRP website ( http://dgrp . gnets . ncsu . edu/ ) to identify SNPs . Analyses include only those SNPs with two alleles and where the minor allele appears in at least 10 percent of the lines . A linear mixed model for each SNP was run using the model G . E = mean + S + M × S + L ( M ) + E where M is the Marker ( SNP ) , S is Sex , L ( M ) is the random Line effect nested within Marker and E is Error . To contrast SNPs with a sex-specific effect on expression to SNPs with a concordant effect we only retained SNPs with concordant association with gene expression across the sexes for SCV genes ( i . e . significant pooled P-value across the sexes ) and only SNPs with a sex-specific association with gene expression for SDV genes ( i . e . significant SNP × Sex P-value ) . Since we were interested in the chromosomal distribution of SNPs rather than identifying particular SNPs with association beyond doubt , there is sequence and gene expression data for only 40 of the DRGP lines , and since trans-SNPs have been shown to have relatively small effect size and escape detection when using too stringent criteria [94] , we chose a P-value cut off at 1×10–5 . This P-value cut off corresponds to a median FDR of 0 . 19 per SCV gene and 0 . 27 per SDV gene . This slightly higher FDR for SNPs associated with FDR genes rendered our results conservative , if anything , since the estimated true proportions of SDV and SCV SNPs ( 0 . 085 and 0 . 141 ) that associate with the X chromosome differ more than the observed proportions ( 0 . 095 and 0 . 140 ) . We calculated the expected true proportions using the formula [observed proportion X-linked SNPs] = FDR x [proportion X-linked SNPs of all tested SNPs] + [1-FDR] x [true proportion of X-linked SNPs] , where [proportion X-linked SNPs of all tested SNPs] = 0 . 137 . To control for differences in linkage between the autosomes and the X chromosome we grouped together , and counted as one SNP , any SNPs that associated with a particular gene that were within 10 base pairs of each other on the autosomes and 30 base pairs of each other on the X chromosome . The different distances standardised the linkage between the autosomes and X chromosome to an r2 of 0 . 2 [30] . For each gene we calculated the proportion of SNPs on the X chromosome and the relative effect size of SNPs on the X chromosome ( average effect size of SNPs on X / average effect size across all chromosomes ) . SDV SNPs were classified as either female- or male-biased based on the sex the SNP had the larger effect size in . The proportion of male- and female-biased SNPs associating with SDV genes was compared with all SNPs associating with SCV genes ( their sex-bias was approximately 0 , given that these genes were chosen to have a very low sex-specific genetic variation and that only SNPs with a concordant effect on gene expression in the two sexes was chosen ) . All analyses were conducted in R v2 . 13 . 0 [95] .
|
Theory provides contrasting predictions with respect to the relative magnitude of standing genetic variation on the X chromosome and the autosomes . While most classic population genetics theory suggests a dearth of variation on the X chromosome , theory which concerns selection operating differently in males and females sometimes suggests the opposite . In support of classic theory we find that genetic variants influencing gene expression far from where they are located in the genome ( trans-acting ) are depleted on the X chromosome . Trans-acting genetic variants with a larger effect in females than in males do , however , depart from this pattern . Allelic variants with a trans-acting female-biased effect are primarily located in regions where the dosage compensation machinery is less active . This suggests that the intrinsic properties of dosage compensation influence the accumulation of different types of trans-factors and/or their propensity to accumulate mutations . These findings have general implications for the evolution of the X chromosome , but may also have relevance for the evolution of sexual dimorphism .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Sex-specific Trans-regulatory Variation on the Drosophila melanogaster X Chromosome
|
Inhibitory pathways are an essential component in the function of the neocortical microcircuitry . Despite the relatively small fraction of inhibitory neurons in the neocortex , these neurons are strongly activated due to their high connectivity rate and the intricate manner in which they interconnect with pyramidal cells ( PCs ) . One prominent pathway is the frequency-dependent disynaptic inhibition ( FDDI ) formed between layer 5 PCs and mediated by Martinotti cells ( MCs ) . Here , we show that simultaneous short bursts in four PCs are sufficient to exert FDDI in all neighboring PCs within the dimensions of a cortical column . This powerful inhibition is mediated by few interneurons , leading to strongly correlated membrane fluctuations and synchronous spiking between PCs simultaneously receiving FDDI . Somatic integration of such inhibition is independent and electrically isolated from monosynaptic excitation formed between the same PCs . FDDI is strongly shaped by I ( h ) in PC dendrites , which determines the effective integration time window for inhibitory and excitatory inputs . We propose a key disynaptic mechanism by which brief bursts generated by a few PCs can synchronize the activity in the pyramidal network .
The mammalian neocortex consists of neurons that form an intricate network of recurrent circuits [1]–[3] . The synaptic wiring between cells follows a number of stereotypic rules including targeting specific domains of neurons , specific connection probabilities , target neuron preferences , and specific short-term synaptic dynamics [1]–[5] . Revealing these rules is essential to understand the mechanisms that generate the response of a cortical column ( or functional unit ) to any external input . In particular , it is crucial to identify the synaptic pathways that enable the neocortex to appropriately respond to all possible environmental stimuli . Neocortical neurons receive excitatory and inhibitory inputs over a variety of different network activity states [6] that seem to be proportionally regulated [7] . This balanced excitatory and inhibitory activity is remarkable since the large majority of cells in the neocortex are ( excitatory ) pyramidal cells ( PCs ) , only around 25% are inhibitory GABAergic interneurons [8] , [9] , and almost 90% of the neocortical synapses are presumably excitatory [10] . This relatively small population of interneurons is responsible for generating a precisely matched inhibition for a variety of cortical network states . One synaptic principle for dynamically adjusting the level of excitation within a neocortical column is the use of dynamically depressing excitatory synapses [11]–[13] , but how inhibitory synaptic pathways ensure dynamic application of balanced inhibition as a function of the moment-to-moment excitation of the neocortical column is not clear . A disynaptic pathway and dynamic circuit mechanism allowing an activity-dependent recruitment of inhibition was recently reported: frequency-dependent disynaptic inhibition ( FDDI ) between PCs is indeed a common pathway in multiple cortical areas that is dynamically regulated by the firing rate and the number of presynaptic PCs [14]–[17] . In contrast to many other cortical connections , the PC–Martinotti cell ( MC ) synapse is strongly facilitating . In response to high frequency stimulation of a PC , spiking activity of MCs can be recruited , thus providing a level of inhibition that depends on the previous excitation level in the network . MCs display a characteristic ascending axonal arborization up to layer 1 [18] , and they are the only interneurons that target the combination of oblique , apical , and tuft dendrites of their neighboring PCs [3] , [14] . FDDI has so far been explored mainly as a pairwise interaction between PCs and MCs , but little is known about how this synaptic pathway could operate to dynamically apply inhibition to the microcircuit as a function of multi-cellular activity . Here , we used multi-neuron whole cell recordings to characterize summation properties of FDDI between layer 5 thick tufted PCs within the dimensions of a neocortical column . FDDI tends to summate linearly with coincident excitatory postsynaptic potentials ( EPSPs ) from neighboring PCs but may also shunt some input arriving at the apical dendrite . Three to four PCs firing simultaneously are sufficient to generate FDDI in all PCs within the dimensions of a cortical column , and eight to nine PCs can saturate the amount of hyperpolarization recorded from their somata . A brief , high frequency burst in only a few PCs can therefore constitute a gating mechanism for further excitatory input to the apical dendrites of the entire column . This inhibition promotes subthreshold correlations and synchronous spiking in PCs .
Figure 1 illustrates the basic components ( A , B ) that mediate FDDI . A presynaptic PC ( red ) projecting onto an MC ( blue ) excites the MC using a strongly facilitating synapse , which in turn gives rise to a delayed inhibition in another postsynaptic PC ( FDDI , black ) . Monosynaptic excitatory connections between PCs occurred in 14% of all tested cases ( probability of occurrence was 0 . 14; 463 out of 3 , 342 tested connections ) , while PC-MC connections occurred far more frequently ( 0 . 43; 26/61 ) and MC-PC connections had a probability of occurrence of 0 . 31 ( 18/58 ) . The entire FDDI loop occurs with a probability of 0 . 283 ( 859/3 , 041 ) , which is more than double the monosynaptic connectivity between two PCs . Silberberg and Markram ( 2007 ) previously showed a strong modulation of FDDI by Ih currents [14] . Blocking Ih currents with extracellular application of zd7288 leads to larger amplitudes ( average 75% increase , n = 23 , μctrl = 0 . 99±0 . 5 mV , μzd7288 = 1 . 73±0 . 99 mV , p = 0 . 0002 , paired t test ) and longer decay time constants ( 250% increase , μctrl = 0 . 051±0 . 01 s , μzd7288 = 0 . 182±0 . 071 s , p = 7 . 64e-9 ) of FDDI ( Figure 1C ) . In some cases ( 3 out of 26 ) FDDI disappeared after Ih block . Since zd7288 blocks Ih irreversibly [19] , we do not know whether the disappearance is due to a drug action or a general rundown . On the other hand , Ih block never leads to FDDI appearance de novo ( n = 19 ) . In order to understand whether the effects can be attributed to Ih on the intermediate interneuron or on the postsynaptic PC , we recorded from the entire disynaptic pathway while Ih was blocked . Facilitating EPSPs from PCs to MCs were only slightly changed in the presence of zd7288 ( average 8% decrease of maximal depolarization; n = 5 , μctrl = 2 . 275±1 . 961 mV , μzd7288 = 2 . 431±1 . 825 mV , p = 0 . 384 , paired t test ) , whereas MC input onto PCs displayed increased synaptic summation ( Figure 1D ) . Thus , the strong effect of zd7288 on FDDI is likely to be mediated by Ih in PCs . PCs receiving both disynaptic inhibition and monosynaptic excitation from their neighboring PCs displayed the tendency of a frequency-dependent transition from a net depolarization to hyperpolarization ( Figure 1E , n = 4 ) . Blockage of Ih resulted in increased frequency dependence , enhancing both low-frequency depolarization and high-frequency hyperpolarization . Together with the observed shortening of synaptic events , this suggests that Ih in PCs acts to localize synaptic inputs , both spatially and temporally . Monosynaptic excitation between PCs mainly targets their basal dendrites [20] while FDDI mainly targets their apical and tuft dendrites [14] . It is not clear to what extent these two inputs interact . We therefore activated both pathways simultaneously and quantified the linearity of summation . Clusters of three PCs , with a PC receiving FDDI from a neighboring PC and a direct excitatory connection from another PC , were stimulated in a way that FDDI and a direct EPSP coincided ( Figure 2A ) . We observed supra- , sub- , and linear amplitude summation in the soma ( Figure 2B ) in different experiments , and on average there was no significant difference in EPSP amplitude between control and coinciding FDDI ( Figure 2C , n = 21 , μctrl = 1 . 885±1 . 334 mV , μFDDI = 1 . 808±1 . 154 mV , p = 0 . 295 , paired t test ) . Inhibition in the distal dendrites may not shunt the peak amplitude of fast AMPA-mediated EPSPs from the basal dendrites but could reduce the total charge . We did not , however , observe any significant change in the integral of the EPSPs ( μctrl = 0 . 08±0 . 057 mV*ms , μFDDI = 0 . 074±0 . 048 mV*ms , p = 0 . 15 , paired t test ) . Next , we used the same protocol to investigate the summation of FDDI with excitatory input to the apical dendrite ( Figure 2A ) . Instead of stimulating a neighboring PC , we synchronously injected a brief current ( aEPSC ) into the trunk of the apical dendrite ( 50–350 µm away from the soma ) that mimicked EPSP kinetics ( τrise = 0 . 5 ms , τdecay = 2 ms ) and peak amplitude ( 200–500 pA , tuned to match a somatic voltage depolarization of 1–4 mV ) . The somatic amplitude ( Figure 2D ) and integral of dendritic aEPSPs was slightly reduced by FDDI input in a distance dependent manner and as a function of the number of presynaptic PCs . We used fast AMPA kinetics for the aEPSPs , which might underestimate the shunting effect by FDDI on events with slower kinetics , namely NMDA components and EPSPs filtered by dendritic attenuation . A further technical limitation of artificial EPSPs via dendritic recording besides the focalization is the fact that excitatory synapses rather target spines , not the trunk like the patch electrode . Nevertheless , these data suggest that FDDI is more effective in shunting synaptic input from the apical and tuft dendrites than input from the basal dendrites , revealing a dual and separable action between layer 5 PCs: direct excitation mostly onto basal dendrites , and indirect inhibition mostly onto the apical and tuft dendrites . This finding is supported by the anatomical separation of the inputs ( Figure 1A , see also [3] , [14] , [20] ) . We performed a set of experiments to estimate the number of MCs that meditate FDDI between two PCs . We stimulated a presynaptic PC that synapses onto an MC , which in turn projects to another PC ( Figure 3A ) . Every other iteration , the MC was prevented from discharge by a hyperpolarizing step current , thereby isolating the effect of this one MC on the FDDI recorded in the postsynaptic PC . FDDI amplitude was reduced to 47 . 5%±38 . 1% ( integral to 45 . 3%±35% ) when the single MC was prevented from participating ( Figure 3B , n = 7 , amplitude: μwMC = 0 . 692±0 . 417 mV , μw/oMC = 0 . 460±0 . 446 mV , p = 0 . 0011 , integral: μwMC = 0 . 08±0 . 082 mV*ms , μw/oMC = 0 . 053±0 . 049 mV*ms , p = 0 . 1148 , paired t tests ) . These results show that although on average multiple MCs participate in FDDI , a single MC can make a significant contribution to the overall FDDI produced in a target PC . The exact number of intermediate MCs is not straightforward to extrapolate . Assuming linear amplitude summation of the MCs' inhibitory postsynaptic potentials ( IPSPs ) , three MCs ( μwMC/ ( μwMC−μw/oMC ) ) participate on average in FDDI upon stimulation of one layer 5 PC ( range 1–28 MCs ) . We might have indirectly prevented further neighboring MCs from spiking through electrical coupling by hyperpolarizing the recorded MC , which might have resulted in an underestimate of participating MCs . Figure 3C shows an example of two MCs coupled via electrical synapses . Their coupling coefficient was 0 . 11 for hyperpolarizing step currents , which is within the range that has been found in previous studies [21] , [22] . Due to low pass filtering , miniature EPSPs in one MC do not pass to the other MC ( arrows in Figure 3C ) . For the same reason , the coupling coefficient was only 0 . 02 for action potentials . Thus , electrical synapses can only play a role in the communication in the FDDI network if synaptic inputs summate with a sufficiently slow time constant so that the signal is not eliminated by low-pass filtering . The same two MCs were targeted by two PCs that were recorded at the same time ( Figure 3D ) providing direct evidence for PC-MC divergent and PC-MC convergent connectivity . We also found multiple cases of MC-PC divergent connectivity ( data not shown ) , indicating that neighboring PCs might share a common pool of MCs for feed-forward and feed-back inhibition . The high degree of interconnectivity between PCs and MCs results in subthreshold correlations between PCs ( Figure 4A , B ) , showing a high correlation coefficient for simultaneous FDDI in different PCs ( n = 28 , μFDDI-FDDI = 0 . 892±0 . 125 ) and significantly lower ones for control conditions ( n = 28 , μCTRL-CTRL = 0 . 085±0 . 364 , n = 26 , μFDDI-CTRL = −0 . 070±0 . 331 , p<0 . 00001 , ANOVA with Scheffe correction ) . This correlation was calculated with average traces and is therefore based on mean responses . In order to estimate the similarity of FDDI in different PCs arising from stimulating a single PC , we performed a trial-to-trial analysis of divergent FDDI responses . In principle , divergent FDDI connectivity may be mediated by a high degree of divergence from PCs onto many different MCs and/or a high degree of divergence from MC to PCs ( see Figure 4C for illustration ) . To quantify the amount of common FDDI input , we defined a “Dissimilarity Index” ( DI ) , which is the root mean squared of mean subtracted traces ( see Methods ) . DI was calculated pairwise between single trial traces , either between simultaneous traces of different cells , or , as a control , between traces of the same ( or different , data not shown ) cells but from different trials . If each postsynaptic PC received FDDI from a different set of interneurons ( as illustrated in the left part of Figure 4C ) , the inhibitory response in the different postsynaptic PCs would not co-vary from trial to trial , resulting in a strong dissimilarity ( high DI , as control ) . In contrast , if each postsynaptic PC received common input from the same set of interneurons ( right part of Figure 4C ) , single-trial FDDI responses between different PCs should be more similar ( low DI , smaller than control ) . If single-trial responses in PCs were identical , DI would be zero . In all tested cases except one , we found a lower DI of simultaneously acquired traces than that of non-simultaneously acquired traces , indicating a high degree of common MC input to neighboring PCs . The data of the illustrated example as well as 43 more cases suggest high MC to PC divergence ( Figure 4D , n = 44 , μac = 0 . 0148±0 . 0033 mV , μar = 0 . 0178±0 . 0028 mV , p = 2e-12 , paired t test ) . Direct connections diverging from a PC to two or more postsynaptic PCs did not have a significantly different DI ( n = 11 , μac = 0 . 0162±0 . 0054 mV , μar = 0 . 0168±0 . 0058 mV , p = 0 . 1063 , paired t test ) . These results show that divergent FDDI from a single PC onto multiple neighboring PCs is not because of a large set of MCs but can be accounted for by a highly divergent MC-PC connectivity . Combined with these findings on the contribution of a single MC on FDDI ( Figure 3A–C ) , we conclude that the high prevalence of FDDI is supported by both PC-MC divergence as well as a high degree of MC-PC divergent connectivity . This MC-PC divergence causes the inhibitory inputs onto neighboring PC to be precisely timed and , together with the mean-based correlations ( Figure 4B ) , enables FDDI to facilitate synchronization of PC activity . Figure 5 shows this synchronization of multiple PCs in the suprathreshold regime . A single presynaptic PC ( Figure 5A , red ) was stimulated with high frequency ( 15 spikes at 70 Hz ) and elicited FDDI in multiple postsynaptic PCs ( black , left column ) . Postsynaptic PCs were stimulated with a suprathreshold step current ( resulting in low frequency spiking of 2–8 Hz ) in the presence of FDDI input ( right column ) , and as a control , without FDDI input ( middle column ) . Without FDDI input , firing of PCs already displayed some variability from trial to trial , probably due to spontaneous membrane potential fluctuations and drifts over the long duration of the stimulus paradigm . As can be seen in the peristimulus time histogram ( Figure 5B ) , however , the probability of spiking is reduced during the beginning of FDDI ( blue color ) , followed by a period of “rebound spiking” at the end and briefly after FDDI ( red color ) . We quantified this effect by counting spikes during this first ( left part of Figure 5C , n = 11 , μCTRL = 10 . 7±3 . 92 , μFDDI = 6 . 3±5 . 1 , p = 0 . 0048 , paired t test ) and second 100 ms time window ( right part of Figure 5C , μCTRL = 12 . 6±7 . 7 , μFDDI = 18±5 . 1 , p = 0 . 0015 , paired t test ) of 22 repetitions in control and FDDI condition . This effect is also quantified by a correlation-based spike timing reliability measure ( Figure 5D; standard deviation of the Gaussian used for convolution with the spike trains was 10 ms; for details on the method , see [23] ) . Spike timing reliability between single repetitions of pairs of postsynaptic PCs increases during FDDI ( left part of Figure 5D , n = 18 , μCTRL = 0 . 197±0 . 109 , μFDDI = 0 . 241±0 . 1034 , p = 0 . 018 , paired t test ) , which also holds true if a time window before FDDI onset is chosen as a control ( right part of Figure 5D , see methods , μBEFORE = 0 . 162±0 . 064 , μDURING = μFDDI , p = 0 . 003 , paired t test ) . Thus , FDDI can lead to synchronous pauses followed by subsequent synchronous spiking in neighboring PCs . Next , we investigated the spatial and temporal integration properties of FDDI in a single PC when multiple presynaptic PCs are stimulated at the same time . Figure 6A shows that an increased number of stimulated PCs leads to a reduced delay of MC firing . The MC was recorded in cell-attached mode so that the intracellular medium remained undisturbed . Not only does the discharge onset take place earlier by tens of milliseconds ( μ3pre = 0 . 114±0 . 021 s , μ2pre = 0 . 202±0 . 036 s , p = 0 . 000021 , two-sample t test ) , but also the number of APs fired by the MC increases ( Figure 6B ) . Figure 6C shows the same type of experiment with the MC recorded in whole-cell mode . Stimulation of two PCs simultaneously can lead to earlier and more numerous spikes ( Rep . 1 , blue traces ) , a PC-MC convergence configuration , which would lead to earlier and larger FDDI in a PC postsynaptic to the MC ( supralinear summation ) . On the other hand , simultaneous stimulation may also lead to earlier MC spiking only ( Rep . 2 , light blue traces ) , which should lead to reduced FDDI amplitude in postsynaptic PCs ( sublinear summation ) . In view of the latency shortening and increased discharge in MCs , we analyzed both amplitudes and onset latencies of FDDI mediated by several presynaptic PCs onto a single postsynaptic one ( Figure 6D–F , Figure 7 ) . Similarly to the previous summation experiments we compared FDDI in response to synchronous stimulation of two PCs ( Figure 6D , black traces in the right column ) to the off-line calculated sum of the separate stimulations ( left and middle column , gray dashed traces , and gray traces in right column ) . As expected , a variety of different responses were found ( Figure 6E ) , ranging from linear summation ( left ) , reduced amplitude with reduced onset delay ( left middle ) , increased amplitude with reduced onset delay ( right middle ) , and increased amplitude with same delay ( right ) . Possible underlying connectivity schemes are depicted in Figure 6F . In cases where the onset delay was shortened , it is very likely that the FDDI is mediated by MCs receiving convergent common excitation from both PCs . The common input decreases the discharge onset of the MC ( s ) and results in earlier onset of inhibition ( see Figure 6A–C ) . Networks that did not exhibit a latency decrease following co-stimulation may also involve MCs receiving common input but not exclusively ( Figure 6F , right ) . The origin of amplitude summation is more complicated , since both supra- and sublinear summation can be explained by convergent PC-MC inputs: if an intermediate MC can be reliably activated only by convergent input , this will result in an average supralinear increase in amplitude . However , if an MC discharges reliably following inputs from both PCs individually , such that co-activation does not significantly increase the number of APs , the result is sublinear amplitude summation . Our results show that on average the latency was shortened by 33 . 7±35 . 8 ms ( n = 103 , p<0 . 0001 , two-tailed t test ) , and the amplitude increase was supralinear ( μsync = 1 . 232±0 . 723 ms , μsummed = 1 . 111±0 . 877 mV , n = 103 , p = 0 . 00096 , paired t test ) . These results indicate that co-stimulation of presynaptic PC pairs increases FDDI in a supralinear manner due to the high degree of PC-MC convergence . How does FDDI summate when more than two neighboring PCs are active ? We stimulated an increasingly larger number of PCs and recorded FDDI in another PC ( Figure 7A , gray shades of the traces according to the number of stimulated cells ) . FDDI monotonically increased in amplitude and voltage integral , which saturated when eight to nine PCs were simultaneously stimulated ( Figure 7B ) . In order to compare the pooled data of many recorded clusters , we used a nonlinearity index for amplitude and integral summation [15] . FDDI summated on average supralinearly for the amplitude as well as for the integral following stimulation of two presynaptic PCs ( Figure S1 , n = 103 , pamp = 1 . 977e-6 , pint = 2 . 710e-13 , one-sample t test ) . Stimulation of three or more ( Figure 7C and S1 ) presynaptic PCs increased the supralinearity of integral and amplitude of FDDI , and also decreased the onset delay . Saturation levels of amplitude and integral difference were reached at around 60% and 70% when six to seven PCs were stimulated simultaneously ( Figure 7C ) . A remarkable feature of FDDI was its abundance in the layer 5 network . Upon stimulation of four PCs simultaneously , all recorded neighboring PCs were inhibited ( Figure 7D ) . Our typical stimulation protocol used to elicit and reliably identify FDDI contained multiple APs ( 15 ) and high frequencies ( 70 Hz ) , a condition that is presumably unlikely to be experienced by PCs in the intact brain . However , the onset of FDDI after this long-train stimulation is variable between cells ( Figure 8A ) , and in several cases , less APs would have been sufficient to trigger FDDI by a single PC , since the hyperpolarization can start off briefly after stimulus onset ( Figure 8B , n = 439 , mean = 0 . 110 s , eight presynaptic APs ) . We also know that synchronous activation of multiple PCs can significantly decrease the FDDI onset ( Figures 6 , 7 and S1 ) . In order to examine whether FDDI can be triggered with few APs only , we stimulated three presynaptic PCs with only three APs at 70 Hz , mimicking the spiking output evoked by dendritic calcium spikes [24] . As can be seen in Figure 8C , even this condition is sufficient to elicit FDDI reliably , with a probability of occurrence of 0 . 23 ( 23 out of 99 tested different quadruplet combinations of cells ) , a mean onset delay of 0 . 068±0 . 012 s , and amplitudes of up to several millivolts ( μ = 1 . 22±0 . 77 mV; range 0 . 25–3 . 7 mV ) . This illustrates that brief synchronous bursts ( ∼three APs at 70 Hz ) of only three PCs are able to trigger FDDI in neighboring PCs , a condition that is likely to be relevant in the in vivo situation .
Previously , the summation properties for convergent FDDI have been investigated for two [15] or three [14] stimulated presynaptic PCs , and the activity-dependent recruitment of MCs was extrapolated for the case of multiple active PCs [15] . Two of our most important findings are that ( a ) every neighboring ( <150 µm intersomatic distance ) thick tufted layer 5 PC is affected by FDDI when four or more PCs burst simultaneously and that ( b ) the FDDI amplitude saturates at the somatic recording site at resting condition when eight to nine PCs are stimulated simultaneously . We find the low number of PCs necessary to trigger FDDI in all neighboring PCs especially remarkable—it shows that in the high frequency , high correlation range the major signaling between PCs is ( after an initial brief excitatory response ) inhibitory . The observed FDDI saturation may be caused by several reasons: limited recruitment of MCs ( due to limited connectivity or limited number of MCs ) , reduction in the driving-force of the inhibitory signal in the apical dendrite when it reaches the GABAA reversal potential , saturating firing rates in MCs , and frequency-dependent synaptic depression of the MC-PC connection . It is likely that all these factors contribute to this early saturation . Summation properties as shown in Figures 3 , 4 , and 6 indicate that only a few MCs are actually recruited by a cluster of PCs . We cannot , however , state that MC recruitment is saturated by stimulation of eight to nine PCs since multiple MCs could mutually shunt their inhibitory signals in a postsynaptic PC and therefore mask the contribution of additional MCs to FDDI . Further , it should be considered that the saturation might not hold in different cortical activity states . For example , if reduced driving force was a major reason for saturation at rest , FDDI might saturate at a later stage at more active cortical states . A high level of excitatory synaptic input to the apical dendrite would require larger activation of the FDDI pathway in order to reach saturation . Kapfer and colleagues [15] extrapolated PC-PC and PC-MC connectivity data to predict a saturation curve for MC recruitment ( see Figure 6 therein ) . Although the studies are not directly comparable and were performed in different cortical layers , our data suggest a smaller dynamic range for recruitment of inhibition and earlier saturation than previously reported . What is not exactly known and not addressed in the current study is the degree of synchrony that brief bursts of neighboring need to have in order to trigger FDDI . We only tested simultaneous stimulations of PCs; a jitter in the PC firing is likely to alter the efficiency of FDDI recruitment and amplitude . Neighboring neocortical cells can show highly correlated activity patterns both in vitro [25] and in vivo [26] , [27] . Recently , it has been shown that the synchrony of subthreshold membrane potential fluctuations depends on the behavioral state of the animal [28] . FDDI acts as a synchronizer of subthreshold membrane potential between PCs in two ways . Multiple PCs , targeted by the same MC , receive FDDI simultaneously , resulting in a high correlation coefficient ( Figure 5B ) . Moreover , due to the reliability of the MC-PC synapse , and its high divergence , the inter-trial variability is mainly due to the summation of the facilitating PC-MC synaptic response . A previous study has also shown that the synaptic dynamics from interneurons are virtually identical across postsynaptic neurons of the same class , which may also underlie the high subthreshold correlations mediated by MCs [29] . Simultaneous responses in different postsynaptic PCs are therefore more similar to each other than the responses of the same PC for different iterations . The high correlation in FDDI across PCs suggests that inhibitory inputs from MCs to PCs may contribute to subthreshold correlations observed between neighboring PCs under in vivo conditions [26] , [27] . Photostimulation studies have suggested that interneurons with adapting firing pattern ( like MCs ) are less specific or selective concerning the targeting of their synaptic input and output [30] , a finding which is in agreement with the high degree of FDDI divergence we report ( Figure 5A ) . Two dynamically different disynaptic inhibitory pathways have been identified in the neocortex [14] and their equivalents in the hippocampus [31] . The pathways differ in their dynamical as well as morphological properties , with the delayed , frequency-dependent pathway activated by MCs ( belonging to the low threshold spiking ( LTS ) class of interneurons ) , triggered by facilitating connections from PCs and target PC dendrites . The other inhibitory pathway conversely is “immediate” and time-locked to PC single APs . It is mediated by depressing connections onto fast-spiking cells , typically PV-expressing basket cells , which in turn target PC perisomatic regions . These interneurons also mediate strong feed-forward inhibition activated by the thalamocortical pathway that has received attention in recent studies , showing that FS interneurons respond to thalamic input by discharge that precedes that of their excitatory neighbors [32]–[34] . LTS cells , on the other hand , receive only weak thalamic input [34] , [35] ( but see [36] ) , suggesting that their activation is primarily intracortical , optimally driven by high-frequency burst discharge of PCs . One implication of the dendritic locus of MC-PC connections , reaching up to the distal dendritic tuft [14] , suggests that FDDI has a role in regulating dendritic excitation , including intrinsic excitability in the form of calcium [24] , [37] and NMDA spikes [38] . Indeed , in a recent study , Murayama and colleagues [39] demonstrated direct blocking of dendritic calcium spikes by FDDI in older animals ( 24–40 d old ) , showing that FDDI is preserved in development and can regulate dendritic excitability in layer 5 PCs . The authors also showed that GABAergic inhibition to PC dendrites originated from layer 5 interneurons and was crucial for enabling a wide dynamic range of calcium responses in vivo , correlated to the intensity of sensory stimulus . Therefore , FDDI might be a precisely matching antagonist of active excitatory conductances like calcium spikes , both being triggered by high frequency bursts . Our study was performed in younger animals , suggesting that development of FDDI onto PC dendrites precedes the maturation of their excitability , which occurs after the third postnatal week [40] . Ih is a prominent current with increasing channel density along the dendrites of layer 5 PCs [41] , [42] . It renders the apical ( and presumably also the basal ) dendrites disconnected from the soma [43] by counteracting any polarization deviating from the resting potential . The decay times of de- and hyperpolarizing inputs are substantially shortened , allowing for a higher temporal precision in the processing of information . Due to its increasing density along the dendrites it also renders EPSP shape and time course site independent [44] . Here we showed that Ih can change the gain between excitation and inhibition for train stimulations , thus increasing the dynamic frequency range . A modulation of this channel conductance might be an approach to profoundly alter this inhibitory pathway [42] , [43] . Studies showed Ih presence in MCs as well , but there seem to be exceptions to this finding , with not all MCs expressing the Ih mediated sag in response to hyperpolarizing step currents [18] . We also did not find a prominent sag in MCs that participated in FDDI ( n = 3 , see also Figure 3C ) . The relative contribution of the various MC populations to FDDI remains to be elucidated [45] . MCs can be modulated by various means . Acetylcholine receptor agonists lead to increased firing in MCs [46] , which might influence the plasticity rules at the apical dendrite of PCs [47] . LTS cells , have been shown to synchronize and oscillate in response to a G-protein coupled glutamate receptor antagonist [48] . This synchronization , mediated by electrical synapses , should enhance and broaden the effect of FDDI in the PC population . Compared to other cell types , MCs seem to be particularly susceptible to changes in the general cortical activity state [49] . Spiking activity ( in certain frequency ranges ) in MCs can trigger intracellular endocannabinoid signaling that eventually leads to hyperpolarization , and thus reduced excitability [50] , [51] . It remains to be elucidated which modulations play strong roles under physiological conditions and to what extent FDDI properties reported in the present study are altered . Several aspects of FDDI still remain to be elucidated . So far , layer 5 thick tufted PCs and layer 3 PCs have been shown to display FDDI [14] , [15] . Cortical-callosal layer 5 PCs with a slender apical dendrite lacking tuft dendrites do not seem to feature this type of inhibition [52] . Also , PCs in layer 6 do not show any measurable FDDI ( Berger and Markram , unpublished data ) . It is , however , not clear whether these potential pathways require a larger number of active neurons to become observable . It remains to be shown whether other PC classes are inhibited in a similar manner and whether this inhibition is mediated via the same MCs . Apart from the FDDI mediated within the same layer , it is possible that presynaptic activity in one layer will inhibit PCs in a different layer . Kapfer and colleagues showed that MCs in layer 5 mediated FDDI between layer 3 PCs [15] , which is in agreement with the neurons' axonal terminal distribution [18] , [45] . It is not known , however , if these MCs are the ones that also mediate FDDI onto layer 5 PCs and whether they are also recruited by layer 5 PCs . It also remains to be elucidated whether supragranular MCs also participate in FDDI between layer 5 PCs . Layer 5 PCs do innervate layer 2/3 , and these MCs target preferentially supragranular layers and possibly also the apical trunk or tuft of layer 5 PCs . Subpopulations of SOM expressing interneurons are now GFP labeled in various mouse strains [45] , facilitating future investigations of FDDI in different layers . A recent study described differences in monosynaptic excitatory connectivity between different types of layer 5 PCs [53] , according to their long-rage projections . It would be of great importance to determine the properties of disynaptic inhibition between these populations as well .
Fourteen- to 18-d-old Wistar rats ( mean age 15 . 0 d , range 14–18 d ) were quickly decapitated according to the Swiss national and institutional guidelines . The brain was carefully removed and placed in iced artificial cerebrospinal fluid ( ACSF ) . Three hundred µm thick parasagittal slices of the primary somatosensory cortex ( hindlimb area ) were cut on a HR2 vibratome ( Sigmann Elektronik , Heidelberg , Germany ) . Slices were incubated at 37°C for 30–60 min and then left at room temperature until recording . Cells were visualized by infrared differential interference contrast videomicroscopy utilizing either a C2400-03 camera ( Hamamatsu , Hamamatsu City , Japan ) mounted on an upright Axioscope FS microscope ( Zeiss , Oberkochen , Germany ) or a VX55 camera ( Till Photonics , Gräfeling , Germany ) mounted on an upright BX51WI microscope ( Olympus , Tokyo , Japan ) . Thick tufted layer 5 PCs were selected according to their large soma size ( 15–25 µm ) and their apparent large trunk of the apical dendrite . Care was taken to use only “parallel” slices , i . e . slices that had a cutting plane parallel to the course of the apical dendrites and the primary axonal trunk . This ensured sufficient preservation of both the PCs' and MCs' axonal and dendritic arborizations . Some experiments included recording of MCs . They were targeted by their soma , which is oval and bitufted , and often oriented sideways . Slices were continuously superfused with ACSF containing ( in mM ) 125 NaCl , 25 NaHCO3 , 2 . 5 KCl , 1 . 25 NaH2PO4 , 2 CaCl2 , 1 MgCl2 , and 25 D-glucose , bubbled with 95% O2–5% CO2 . The intracellular pipette solution ( ICS ) contained ( in mM ) 110 K-gluconate , 10 KCl , 4 ATP-Mg , 10 phosphocreatine , 0 . 3 GTP , 10 N-2-hydroxyethylpiperazine-N9-2-ethanesulfonic acid ( HEPES ) , and 13 biocytin , adjusted to a pH 7 . 3–7 . 4 with 5 M KOH . Osmolarity was adjusted to 290–300 mosm with D-mannitol ( 25–35 mM ) . The membrane potential values given were not corrected for the liquid junction potential , which was approximately −14 mV . 4- ( N-ethyl-N-phenylamino ) -1 , 2-dimethyl-6- ( methylamino ) pyridinium chloride ( zd7288 ) was bought from Biotrend ( Zurich , Switzerland ) , and all other drugs and chemicals were from Sigma-Aldrich ( St . Louis , MO ) or Merck ( Darmstadt , Germany ) . Multiple somatic whole cell recordings ( 2–12 cells simultaneously ) were performed with Axopatch 200B or Multiclamp 700B amplifiers ( Molecular Devices , Union City , CA ) in the current clamp mode . We selected PCs that were located close to each other , preferentially in clusters of near to adjacent cells . When 12 cells were recorded at the same time , the pairwise intersomatic distance increased due to limited accessibility with multiple patch electrodes in the tissue but did not exceed 150 µm . In some experiments , MCs were first recorded in voltage clamp in the cell-attached configuration , leaving the intracellular medium unperturbed , and then in whole-cell mode , thus perfused with the ICS and Biocytin contained in the pipette , allowing a subsequent staining and cell type identification . In experiments including dendritic recordings , dendrites were patched before the somata . Alexafluor594 ( Invitrogen , Eugene , OR ) was sometimes included in the dendritic patch electrode , revealing the corresponding soma unambiguously . The temperature was 34°C±1°C during recording . Data acquisition was performed via an ITC-18 or ITC-1600 board ( Instrutech Co , Port Washington , NY ) , connected to a PC or Macintosh running a custom written routine under IgorPro ( Wavemetrics , Portland , OR ) . Sampling rates were 5–10 kHz , and the voltage signal was filtered with a 2 kHz Bessel filter . Patch pipettes were pulled with a Flamming/Brown micropipette puller P-97 ( Sutter Instruments Co . , Novato , CA ) and had an initial resistance of 3–8 MΩ ( 10–15 MΩ for dendritic patches ) . 3D morphological reconstruction of biocytin-labeled cells was done under an Olympus BX 51 W microscope fitted with a water-immersion 60× ( numerical aperture ( NA ) 0 . 9 ) or an oil-immersion 100× ( NA 1 . 35 ) objective using Neurolucida software ( MicroBrightField , Magdeburg , Germany ) . Monosynaptic , direct connections were usually identified by stimulation of a presynaptic cell with a 20 Hz train of eight strong and brief current pulses ( 1–3 . 5 nA , 2–4 ms ) , followed by a so-called recovery test response ( RTR ) 0 . 5 s after the end of the train , all precisely and reliably eliciting APs . Disynaptic connections were characterized by the same protocol but at a higher frequency ( usually 70 Hz ) and with longer trains ( usually 15 APs ) . Postsynaptic PCs were slightly depolarized from a potential of ∼−62 mV to −57 to −60 mV to increase the driving force for inhibitory connections . This was usually not necessary to detect FDDI but gave larger amplitudes occasionally . Due to the dendritic location and the resulting space clamp effect in layer 5 PCs , especially MC-PC synapses have a very hyperpolarized apparent somatic reversal potential that deviates strongly from the calculated one [14] . We did not find any depolarizing FDDI responses , possibly because we used rats older than 13 d [54] . Connectivity ratios were calculated as the ratio between observed versus tested connections between a pair of cells . A pair of cells could therefore maximally have two connections ( both directions ) , a triplet could have six connections , and a cluster of n neurons could potentially have n * ( n−1 ) connections . “Autaptic” connections—that is , FDDI elicited and received by the same PC—were not taken into consideration . The balance between de- and hyperpolarization due to FDDI and direct EPSPs as a function of stimulation frequency ( Figure 1E ) was calculated as the net polarization deviating from baseline in the time window starting from stimulation onset and ending just before the RTR , i . e . 0 . 5 s after the stimulation train ended . Bath application of zd7288 resulted in a strong hyperpolarization of PCs ( ∼10–12 mV; [14] ) , which was counteracted by a positive holding current to reestablish resting membrane potential of around −60 mV . The waiting time between stimulations was 10–20 s . Especially for FDDI summation experiments ( Figures 6–8 , S1 ) long waiting times were crucial as FDDI amplitudes would decrease otherwise ( much more dramatic than , e . g . , EPSP amplitudes ) . For these figures , we only included “pure” FDDI responses ( without monosynaptic EPSP contamination ) in the analysis . Stimulations were given in an alternating manner ( ABAB… instead of AABB… ) . For summation experiments as shown in Figure 7 and S1 , linearity of amplitude ( and likewise integral ) was calculated as a normalized difference according to L = ( Ainput ( 1 , 2 , … , n ) − ( Ainput1+Ainput2+…+Ainputn ) ) /Ainput ( 1 , 2 , … , n ) , where Ainput ( 1 , 2 , … , n ) is the amplitude of the simultaneous stimulation and Ainput1+Ainput2+…+Ainputn is the offline calculated sum of the separately stimulated presynaptic cells . For Figures 7C , D and S1 , data were included if the FDDI evoked by synchronous stimulation exceeded 0 . 5 mV , as the signal-to-noise ratio was too high for the difference measures otherwise . All statistical analysis ( paired and unpaired student's t test , ANOVA ) was done with MATLAB ( The Mathworks , Natick , MA , USA ) . The DI was defined aswhere xi and yi are single repetitions of baseline-subtracted ( mean of the first 100 ms before stimulation was taken as a reference ) traces of different or identical cells and different or identical repetitions , and x′ and y′ are the baseline-subtracted mean responses . DI is the point-wise squared difference between mean- and baseline-subtracted traces , calculated for every possible pair of traces , i . e . “across cells , same repetition , ” “same cell , across repetitions , ” and “across cells , across repetitions . ” It quantifies the deviation from the average response and shows whether noise coming along the FDDI signal co-varies between two cells or not . Given one stimulated presynaptic PC , two postsynaptic PCs receiving FDDI , and n repetitions of stimulation , one obtains n “across cells , same repetition” conditions , n * ( n+1 ) /2 “same cell , across repetitions” conditions , and n * ( n−1 ) “across cells , across repetitions” conditions . For the latter two conditions the DI measure was nearly identical , therefore the “across cells , across repetitions” condition is not displayed in Figure 4E . DI was taken for the interval from 0 to 0 . 5 s after stimulation onset . Note that DI is intended to compare traces that have been stimulated in the same way . It is therefore not meaningful to compare DI values of FDDI ( stimulated with 15 APs at 70 Hz , disynaptic ) with EPSPs ( 8 APs at 20 Hz , monosynaptic ) . Cross-correlation and Pearson's correlation coefficient of two mean FDDI responses ( Figure 4A and 4B ) were calculated with Igor Pro . The effect of FDDI on spiking postsynaptic PCs ( Figure 5 ) was quantified by counting spikes at specific 100 ms time windows of the peristimulus time histogram , namely during the second half of ( first window ) and immediately after ( second window ) presynaptic train stimulation . Spiking responses of postsynaptic PCs without coincident FDDI input served as control condition . Peristimulus time histograms contained spike counts of around 22 repetitions . Correlation-based spike timing precision was calculated according to [23] and on 0 . 5 s long time windows .
|
The neocortex of the mammalian brain contains many more excitatory neurons than inhibitory neurons , yet inhibitory neurons are essential components of neocortical circuitry . Inhibitory neurons form dense and intricate connections with excitatory neurons , which are mainly pyramidal cells . One prominent pathway formed between pyramidal cells and inhibitory Martinotti cells is frequency-dependent disynaptic inhibition ( FDDI ) , which mediates a strong inhibitory signal in the microcircuitry of the neocortex . Here , we reveal deeper insight into how FDDI is mediated and recruited within the circuit , showing that short simultaneous bursts in four pyramidal cells are sufficient to exert FDDI in all neighboring pyramidal cells within the dimensions of a cortical column . As few as three synchronous action potentials in three pyramidal cells can trigger FDDI . This powerful inhibition is mediated by only a few inhibitory neurons yet correlates membrane potential fluctuations , leading to synchronous spiking between pyramidal cells that simultaneously receive FDDI . The inhibitory signals are independent and electrically isolated from excitation mediated by neighboring PCs via basal dendrites . We propose FDDI as an important pathway that is readily activated by brief bursts of action potentials and correlates neocortical network activity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"neuroscience",
"neuroscience/sensory",
"systems"
] |
2010
|
Brief Bursts Self-Inhibit and Correlate the Pyramidal Network
|
Diverse mechanisms for DNA-protein recognition have been elucidated in numerous atomic complex structures from various protein families . These structural data provide an invaluable knowledge base not only for understanding DNA-protein interactions , but also for developing specialized methods that predict the DNA-binding function from protein structure . While such methods are useful , a major limitation is that they require an experimental structure of the target as input . To overcome this obstacle , we develop a threading-based method , DNA-Binding-Domain-Threader ( DBD-Threader ) , for the prediction of DNA-binding domains and associated DNA-binding protein residues . Our method , which uses a template library composed of DNA-protein complex structures , requires only the target protein's sequence . In our approach , fold similarity and DNA-binding propensity are employed as two functional discriminating properties . In benchmark tests on 179 DNA-binding and 3 , 797 non-DNA-binding proteins , using templates whose sequence identity is less than 30% to the target , DBD-Threader achieves a sensitivity/precision of 56%/86% . This performance is considerably better than the standard sequence comparison method PSI-BLAST and is comparable to DBD-Hunter , which requires an experimental structure as input . Moreover , for over 70% of predicted DNA-binding domains , the backbone Root Mean Square Deviations ( RMSDs ) of the top-ranked structural models are within 6 . 5 Å of their experimental structures , with their associated DNA-binding sites identified at satisfactory accuracy . Additionally , DBD-Threader correctly assigned the SCOP superfamily for most predicted domains . To demonstrate that DBD-Threader is useful for automatic function annotation on a large-scale , DBD-Threader was applied to 18 , 631 protein sequences from the human genome; 1 , 654 proteins are predicted to have DNA-binding function . Comparison with existing Gene Ontology ( GO ) annotations suggests that ∼30% of our predictions are new . Finally , we present some interesting predictions in detail . In particular , it is estimated that ∼20% of classic zinc finger domains play a functional role not related to direct DNA-binding .
The past decade has witnessed tremendous progress in genome sequencing [1]–[5] . According to the Genomes On Line Database , the complete sequenced genomes of almost 1 , 000 cellular organisms have been released , and about 5 , 000 active genome sequencing projects are on the way [6] . The unprecedented amount of genetic information has provided hundreds of thousands of protein sequences [7] . This poses a challenging problem to elucidate their functions , as experimental characterization of all newly sequenced proteins is obviously impractical . Fortunately , many of them are homologous to proteins that have been experimentally studied . Consequently , it would be highly desirable to develop computational approaches that automatically annotate a new protein sequence through its functionally characterized homologs [8]–[10] . The key component of such approaches is the ability to detect homologous relationships between un-characterized and characterized proteins . For this purpose , many sequence and structural similarity comparison methods have been developed [11]–[15] . While sequence-based methods are powerful and widely adopted for function inference [16]–[18] , structure-based methods are more sensitive in detecting homologs with low or no sequence similarity [19]–[21] . However , significant sequence or structural similarity does not necessarily lead to identical function , since the functional roles of related proteins can diverge during the course of evolution [22] , [23] . To address this problem , it is often necessary to examine the conservation of functionally discriminating residues when predicting enzymatic functions [17] , or to evaluate the interaction energy when predicting protein-protein [24] or protein-DNA interactions [19] . DNA-binding function is a key characteristic of many proteins involved in various essential biological activities; these include DNA transcription , replication , packaging , repair and rearrangement . These DNA-binding proteins have a diversified classification according to their structures and the way they interact with DNA [25] , [26] . Due to the importance of DNA-binding proteins , a few dedicated computational approaches have recently been proposed for the prediction of DNA-binding function from protein structure [19] , [27]–[31] . These methods can be classified into two groups: structure template-based and template-free , depending on how ( or if ) they use the information from the known structures of DNA-binding proteins . Template-based methods utilize a structural comparison protocol to detect significant structural similarity between the query and a template known to bind DNA at either the domain or the structural motif level and use a statistical or electrostatic potential to assess the DNA-binding preference of the target sequence [19] , [29] . The latter assessment reduces the number of false positives , which is important for the success of these methods . Template-free methods do not perform direct structural comparison , but typically follow a machine-learning framework and use features such as sequence composition and biophysical properties of surface patches [27] , [28] , [30] , [31] . Although they can potentially detect a novel DNA-binding protein fold , template-free methods generally have lower accuracy than template-based methods , which perform well on large-scale datasets and have been applied to structural genomics targets [19] . In addition to DNA-binding function , it is also of interest to predict the amino acids that directly participate in DNA-binding . This is often straightforward for a template-based approach , as one can infer the binding residues directly from the identified template [19] . By comparison , in a template-free approach , one needs to design a new prediction protocol [32]–[36] . Recently , a new approach has been developed to predict DNA-binding residues through DNA-protein docking [37] . This approach , which takes the advantage of the non-specific DNA-binding ability of DNA-binding proteins , provides a coarse model of the DNA-protein complex in addition to the prediction of DNA-binding sites . Although structural information is helpful for predicting DNA-binding function , it can also limit the scope of application because less than 1% of all proteins have an experimentally determined structure [38] . To overcome this limitation , we introduce a threading-based method , DBD-Threader , for the prediction of DNA-binding domains and associated functional sites . Threading-based approaches , which require only sequence as query input , have been successfully applied to the prediction of protein-protein interactions [24] , [39] and protein-ligand interactions [40] . Below , we first describe the framework of our approach , and then compare its performance with three established methods , including the standard sequence alignment tool PSI-BLAST [11] , the threading method PROSPECTOR [41] , and the experimental structure-based DNA-binding prediction method DBD-Hunter [19] . Finally , we present the application of DBD-Threader to the human genome , for which DBD-Threader detected ∼7 , 000 DNA-binding domains in 59 SCOP superfamilies . We also predict that ∼20% of classic zinc finger domains play a functional role not related to direct DNA-binding .
DBD-Threader uses fold similarity evaluated by the threading Z-score , and DNA-binding propensity , evaluated by DNA-protein interaction energy , as two properties to discriminate DNA-binding proteins from non-DNA-binding proteins . The effectiveness of these two properties are demonstrated through an analysis of 179 DNA-binding proteins ( DB179 ) and 3797 non-DNA-binding proteins ( NB3797 ) ; two non-redundant datasets collected previously [19] . The sequences of these ∼4 , 000 proteins were used as input . For each target , we excluded from the library any template whose sequence is more than 30% identical to the target , since we are mostly interested in detecting homologs at low sequence identity . A significant threading Z-score for a pair of target/template proteins typically suggests a high level of structural similarity . Since two proteins with similar structures more likely share the same function than those in different structures , the threading Z-score can serve as a good indicator not only for structure similarity , but also for function similarity . As shown in Figure 1A , 70% ( 126 ) of the proteins in DB179 hit at least one template from the DNA-binding domain library with a significant Z-score>6; 25% ( 44 ) hit with a high Z-score>20 . By contrast , only 3 . 9% ( 149 ) proteins of NB3797 hit at least one template with a Z-score>6 , and only two targets from NB3797 hit a template with a high Z-score>20 . These results suggest that one can utilize threading to filter out the vast majority of non-DNA-binding proteins , while keeping many homologs with DNA-binding function . However , since the numbers of proteins with a significant hit are about the same in the DNA-binding and the non-DNA-binding protein sets , about half of the predictions would be incorrect if one chooses a Z-score of 6 as the threshold to determine the DNA-binding function . One can raise the threshold to a high Z-score of 20 , which would greatly improve the precision of the predictions to 96% . But , it would also reduce dramatically the sensitivity ( coverage ) of the predictions to only 25% . Thus , use of threading alone has a limited accuracy when applied for functional inference . To further improve the precision without seriously compromising the sensitivity of the predictions , we introduce a DNA-protein statistical pair potential to assess DNA-binding propensity [19] . It has been shown that this term can be used to differentiate DNA-binding protein residues from non-DNA-binding residues , independent on the specific DNA substrates involved [19] , [37] . If a pair of target/template proteins has similar structure , then the target protein might favorably interact with the template DNA in a similar way as the template protein . This assumption is generally valid , as shown in the distributions of the DNA-protein interaction energy of targets with at least one significant ( Z-score>6 ) template ( Figure 1B ) . For each target , the lowest energy is shown if more than one significant hit is identified . 94 of 126 targets from DB179 have attractive DNA-protein interaction energy values <−5 , whereas only 28 of 149 targets from NB3797 have an energy value <−5 . The analysis suggests that a functional relationship between remote homologs can be established at quite high precision through a combination of threading and interaction energy calculations , which is the strategy adopted by DBD-Threader . To benchmark the performance of our approach , DBD-Threader is compared with three methods: PSI-BLAST [11] , PROSPECTOR [41] , and DBD-Hunter [19] . Two sequence libraries from NCBI and from UniProt were used to derive the position specific sequence profile for PSI-BLAST , respectively . Details of the assessment procedures are given in Methods . Figure 2A shows the precision-recall ( PR ) and Figure 2B shows the Receiver Operator Characteristic ( ROC ) curves for benchmark tests on DB179 and NB3797 . DBD-Threader generally performs better than PROSPECTOR and PSI-BLAST , especially at a precision higher than 0 . 75 and at False Positive Rate ( FPR ) lower than 0 . 01 , the regime relevant to practical applications . Correspondingly , the sensitivity obtained by DBD-Threader can be higher than 0 . 55 within this regime . If one considers only fold similarity suggested by the threading Z-score or sequence similarity measured by the PSI-BLAST E-value , one obtains an inferior precision/FPR at the same level of sensitivity . For example , at a sensitivity value of 0 . 55 , the precision/FPR for DBD-Threader , PROSPECTOR , and PSI-BLAST ( NCBI ) , and PSI-BLAST ( UniProt ) is 0 . 85/0 . 004 , 0 . 69/0 . 012 , 0 . 24/0 . 085 , and 0 . 24/0 . 081 , respectively . Therefore , the results suggest that the quality of predictions by DBD-Threader is significantly improved when both threading Z-score and protein-DNA interaction propensity are taken into account . We also note that threading itself ( PROSPECTOR ) typically performs better than PSI-BLAST . The comprehensive performance of these methods can be assessed by the Matthews Correlation Coefficient ( MCC ) [42] . A perfect prediction at 100% accuracy yields a MCC of one , whereas a random prediction gives a MCC of zero . The best MCCs of these four methods are provided in Table 1 . The highest MCC of DBD-Threader is 0 . 680 , corresponding to a sensitivity of 0 . 56 and a precision of 0 . 86 , whereas the best MCCs of PROSPECTOR , PSI-BLAST ( NCBI ) , and PSI-BLAST ( UniProt ) are 0 . 609 , 0 . 540 , and 0 . 553 , both as shown in Table 1 at lower sensitivity and precision than DBD-Threader . Moreover , the best performance of DBD-Threader is only slightly lower than that ( MCC 0 . 681 ) of DBD-Hunter , which requires the structure of the target as input . Note that the previous results of DBD-Hunter were obtained on a smaller template library [19] . The results reported here are based on the updated template library employed by all methods . Direct structural comparison allows DBD-Hunter to detect homology between a pair of template/target proteins with no sequence similarity , resulting in the highest sensitivity of 0 . 61 among all four methods at the same precision of 0 . 79 . Nevertheless , the performance of DBD-Threader is comparable to that of DBD-Hunter in terms of its MCC . The optimal thresholds corresponding to the best performance of DBD-Threader were adopted in the application to the human genome below . The contributions by threading and by energy to the optimal performance of DBD-Threader are further dissected through an analysis of DNA-binding and non-DNA-binding proteins that share common structural folds . Here , we use the Structural Classification of Proteins ( SCOP ) to classify structural folds [43] . Table 2 shows the numbers of proteins ( and their relevant domains ) that belong to the same SCOP folds across two benchmark sets DB179/NB3797 . In total , there are 109/599 proteins that contain 127/646 domains from 24 common SCOP folds . The vast majority of non-DNA-binding proteins were filtered out after the threading procedure , resulting in a 90% reduction in non-DNA-binding proteins but only a 31% reduction in DNA-binding proteins to 75/58 DB/NB proteins . After applying the optimal energy criteria , the number of DNA-binding proteins is reduced by 23% to 58 , whereas the number of non-DNA-binding dramatically decreases again by 86% to 8 . We note that in some sparsely populated ( number of DB targets ≤4 ) folds , successive filtering by threading and energy left no true positive from the DB set . This is mainly due to the absence of a suitable template under the specified sequence identity cutoff of 30% . By ignoring these folds , one still obtains about 75% and 80% reduction rates on non-DNA-binding proteins through threading and energy filtering , respectively , while the majority of DNA-binding proteins are kept . Overall , the analysis shows that both threading and energy calculations significantly contribute to the ability to distinguish the DNA-binding function among proteins with similar folds . There are 91 non-DNA-binding proteins with at least one significant template hit ( threading Z-score>6 ) , but they are from other SCOP folds that lack any known DNA-binding protein . These non-DNA-binding proteins may contain structural fragments similar to their significant template hits or may be falsely identified by threading . By applying the energy criteria , 82 of these 91 proteins were correctly filtered out as non-DNA-binding proteins . The energy calculations , therefore , serve to reduce the number of potential false positives generated by threading . The contribution of energy filtering can be illustrated through two examples from the NB3797 . The top ranked template hits by these two targets are significant with Z-scores over 20 , but these templates did not satisfy the energy criteria because of their high repulsive DNA-protein interaction energies . Both proteins are classified as non-DNA-binding . The first example is an inositol polyphosphate 5-phosphatase ( PDB 1i9yA ) , which hits a DNA repair protein APE1 ( 1dewB ) . They are evolutionarily related and belong to the same SCOP superfamily . However , they have very different selectivity for substrate , as the inositol polyphosphate 5-phosphatase is not known to bind DNA . The second example is λ lysozyme ( 1am7A ) , which hit an endonuclease ( 2fldA ) with a high Z-score . This seems to be a false positive by threading , since the target/template pair shares no apparent structural similarity and are not related . Nevertheless , the template did not pass energy screening . Overall , by applying energy filtration , the number of true/false positives decreases from 131/149 ( after threading ) to 100/17 , the numbers including results from all targets with official SCOP classification , as well as those unclassified . Thus , the filtration by energy improves the precision from 47% to 86% without dramatically compromising the sensitivity . In addition to function prediction , DBD-Threader also predicts structural models of DNA-binding domains from templates that provide the structural basis for function prediction . Furthermore , one may infer the functional sites directly from the template , once the functional and structural similarity between the template and the target is established . To demonstrate this point , we implemented a simple procedure in DBD-Threader that predicts DNA-binding protein residues from the top ranked template by those residues in the target aligned to DNA-binding residues in the template . In benchmark tests on DB179 , this procedure was conducted on 124 domains from 100 DNA-binding proteins predicted as positives by DBD-Threader at the optimal thresholds . The value of the MCC , which measures the degree of overlap between predicted binding residues and the true binding residues observed in the native ( experimental ) complex structures , is used to assess the accuracy of functional site prediction . As shown in Figure 3A , DBD-Threader performs well on both structural and functional site prediction . The mean Template Modeling score ( TM-score ) of the top-ranked structural models of the 124 DNA-binding domains with respect to their native structures is 0 . 65 , and 92% of these domains have a TM-score higher than 0 . 4 , which indicates significant structural similarity [15] . Similarly , 70% of these domain models have a backbone Cα RMSD of less than 6 . 5 Å from their native structures . Accordingly , the mean MCC of binding site predictions is generally satisfactory , being about 0 . 52 for all predicted DNA-binding domains and 0 . 54 for domains with a TM-score higher than 0 . 4 . As one may expect , the accuracy of binding site prediction is correlated with model quality . High quality models with a TM-score higher than 0 . 6 generally provide a high accuracy binding site prediction , yielding a mean MCC of 0 . 57 , whereas low quality models with a TM-score lower than 0 . 4 provide inferior binding site predictions with MCCs lower than 0 . 4 . We further analyzed the performance according to the SCOP superfamily association of these predicted domains , as shown in Table 3 . The analysis considers 84 predicted DNA-binding proteins that have an official SCOP assignment , which includes 100 domains detected by DBD-Threader and an additional 10 domains missed by DBD-Threader ( see SCOP superfamily prediction below ) . According to their SCOP classifications , the 100 detected domains are from 31 SCOP superfamilies . The performance of DBD-Threader is generally good across various SCOP superfamilies . 24 of 31 superfamilies have a mean TM-score/MCC higher than 0 . 4 . It appears that members of the winged helix superfamily have rather diverse DNA-binding sites . This is indicated by the mean MCC of 0 . 38 , despite the high quality of models that are obtained ( mean TM-score of 0 . 65 ) . The performance measures , sensitivity , specificity , accuracy and precision , were also calculated for each of 100 proteins including all DNA-binding domains . As shown in Figure 3B , for 61% of predicted DNA-binding proteins , good functional site predictions were obtained at a MCC higher than 0 . 50 . On average , a MCC of 0 . 53 , a sensitivity of 0 . 60 , a specificity of 0 . 93 , an accuracy of 0 . 86 and a precision of 0 . 64 were obtained . The results imply that DNA binding residues were identified with satisfactory accuracy in most cases . The homologous relationship between the target/template pairs identified by DBD-Threader was further validated using their SCOP superfamily classifications [43] . Here , we test the idea of inferring the SCOP superfamily identity of a predicted DNA-binding domain from its templates . Among 100 predicted DNA-binding proteins , we only consider those whose SCOP assignments have been officially assigned . The consideration leads to 84 proteins composed of 110 true domain assignments , which are then compared with the predictions by DBD-Threader . The predictions can be classified into four groups , as shown in Table 3 . The first group is 73 SCOP superfamily predictions that are consistent ( C ) with the official SCOP assignments . These correctly predicted domains are from 21 different SCOP superfamilies , including two of the most populated superfamilies , homeodomain-like and winged helix domains , with 16 and 11 correct predictions , respectively . These 27 domains consist of diverse members from 16 different SCOP families . The second group of predictions is 20 DNA-binding domains correctly identified as DNA-binding , but their SCOP superfamily classifications were un-annotated ( U ) because the corresponding templates have no official SCOP assignment . In 16/20 un-annotated cases , significant structural similarities between target/template pairs were found at a TM-score>0 . 5 , implying that most of these pairs likely belong to the same superfamily . The 16 cases that are un-annotated combined with the 73 consistent SCOP predictions lead to 89 cases , or 81% of 110 domains , that may be considered correct . The third group is comprised of ten missed ( M ) DNA-binding domains , which are from proteins with multiple DNA-binding domains . In these cases , DNA-binding function can be successfully predicted by identifying some but not all of its DNA-binding domains . The fourth group of predictions are from the seven cases where the SCOP superfamily predictions are inconsistent ( I ) with the true SCOP assignment . Inspection of these predictions suggests potential functional homology in 5/7 cases . Two are presented in detail below . In the first example , the target protein is the DNA-binding domain of PhoB , a transcription activator from E . coli [44] . According to SCOP , this domain belongs to the superfamily named C-terminal effector domain of the bipartite response regulators . DBD-Threader predicts that the domain belongs to the superfamily of winged helix DNA-binding domains based on its top ranked template , the Zα domain of an enzyme ADAR1 ( Adenosine Deaminase Acting on RNA ) from human [45] . Although ADAR1 is best known as an RNA binding protein , it is also known to bind Z-DNA with its Zα domain , as shown in multiple crystal structures of ADAR1/DNA complexes [45] , [46] . In addition , the DNA-binding ability of Zα has been used to detect stable Z-DNA segments in the human genome [47] , and has been linked to a new functional role of ADAR1 as a sensor of immunoreactive DNA [48] . Despite the difference in SCOP superfamily classification , the target and the template share a similar structural motif , with a high TM-score of 0 . 69 , as shown in Figure 4A . In fact , both structures are members of the same superfamily of winged helix domains according to CATH , a hierarchical classification of protein domain structures [49] . In addition , both DNA-binding domains have similar DNA-binding sites , which include six residues from a α helix and a β hairpin ( Figure 4A ) . The significant structural similarity and the overlap of the DNA-binding sites suggest that these two domains might have remote homology , despite the lack of sequence similarity . Thus , we have an interesting case of the PhoB domain being correctly assigned as DNA binding through the matching to an RNA binding protein that is also known to bind DNA . In the second example , the target is the N-terminal domain from a eukaryotic DNA polymerase , Pol β [50] . The target hits a significant template from an archaeal endonuclease XPF , whose structure is composed of two heterogeneous domains [51] . As shown in Figure 4B , the target domain from Pol β was aligned to the N-terminal domain of XPF with significant structural similarity , having a TM-score of 0 . 48 , and considerable overlap of DNA-binding residues , despite the fact that the two domains have different superfamily classifications in SCOP . The structural and functional site analysis suggests that the two domains may have a remote relationship . To demonstrate that DBD-Threader is a useful tool for automatic function annotation , we applied DBD-Threader to 18 , 621 unique protein sequences from the human genome . The method made positive predictions for 1 , 654 ( 8 . 9% ) proteins ( see Methods for availability ) . Our predictions are compared to the GO annotations for the human genome [52] in Figure 5 . According to the GO molecular function annotations , all human proteins can be classified into four sets: DB−1 , 744 proteins annotated as DNA-binding , UB−1 , 573 proteins not explicitly annotated as DNA-binding but annotated with a molecular function likely implicating DNA-binding , such as transcription factor activity , NB−10 , 616 proteins with at least one molecular function annotation and not in either DB or UB , and UK−4 , 688 proteins with unknown molecular function . While the vast majority of entries in DB are classified based on electronic annotations , we collected a small subset of DB , named DB EXP , in which the DNA-binding function has been verified for each member in direct experimental assay . This DB EXP set consists of 69 sequences . DBD-Threader detected at least one significant structural template for 56 of them and correctly predicted 54 as DNA-binding . Similarly , when applied to the DB set , DBD-Threader found at least one significant template for 1 , 235 sequences and predicted 1 , 179 ( 95% ) of them as DNA-binding proteins . Notably , when applied to the UB set , DBD-Threader predicted 325 DNA-binding proteins . Among these UB positives , 256 and 51 have transcription factor activity and RNA-binding activity according to their GO annotations , respectively . These proteins likely possess DNA-binding function as well . While 89% of the positives are from either DB or UB , very few positives , 72 ( 0 . 68% ) , are from the NB set . This result is expected , since the chance that a protein has both DNA-binding and an unrelated molecular function is small . Despite the fact that 298 targets from NB hit a significant structural template that binds DNA with Z-score>6 , 75% of them are filtered out by the interaction energy criterion . These negatives likely possess a fold similar to a DNA-binding domain , but they do not carry out the same function . Furthermore , DBD-Threader predicts 78 DNA-binding proteins among previously uncharacterized sequences . These predictions provide potentially interesting targets for further experimental validation . A total of 6 , 896 DNA-binding domains from 59 SCOP superfamilies were located by DBD-Threader in the sequences of 1 , 654 positives . The top twenty most populated SCOP superfamilies of predicted DNA-binding domains are listed in Table 4 . Notably , zinc-fingers appear in about 41% ( 674 ) of predicted DNA-binding proteins , and this particular superfamily dominates the domain predictions at 80% of the total ( 5 , 504 ) . The second and third most common SCOP superfamilies are homeodomain-like and winged helix domains , which are found in 263 and 143 sequences , respectively . Many DNA-binding proteins , particularly zinc-fingers , contain two or more DNA-binding domains . Moreover , it is not uncommon that a sequence encodes DNA-binding domains from different SCOP superfamilies . In our annotations , we found 175 such cases . Our predictions are compared with Pfam predictions , which are based on Hidden Markov Models ( HMMs ) [53] in Table 4 . The results of Pfam predictions were obtained from the UniProt knowledge base . For an objective comparison , we consider Pfam families defined for DNA-binding proteins from our template library . The Pfam definitions of these template structures were initially obtained from the PDB . These were then manually curated to ensure that the definitions correspond to the DNA-binding domains . This led to 179 Pfam families that likely include all DNA-binding proteins with known atomic complex structures , but not those with unknown structures or with only DNA-unbound structural forms . Using the SCOP definitions of the templates , we are able to assign these Pfam families to 69 SCOP superfamilies . Overall , Pfam found 7 , 162 significant domain matches in 1 , 591 proteins from the same sequence set scanned by DBD-Threader . The numbers are consistent with the 6 , 896/1 , 654 domains/proteins predicted independently by DBD-Threader . As shown in Table 4 , the results of the top three most populated SCOP superfamilies are comparable between these two methods . About 80%/85% , 4 . 5%/4 . 3% , and 2 . 1%/1 . 4% of predicted domains are zinc finger , homeodomain , and winged helix proteins by DBD-Threader/Pfam , respectively . The zinc finger proteins dominate both predictions , and over 95% of predicted zinc finger proteins were positively hit by both methods . Despite the similarity , however , about 26% of zinc finger domains detected by Pfam are predicted as negatives by DBD-Threader . One interesting question is whether these domains have evolved their function from DNA-binding to have other roles not involving DNA-binding . Although the vast majority of these proteins have not been experimentally studied , we found a few potential examples of such zinc finger domains with experimental evidence from the literature ( see Case Studies below ) . Moreover , we noticed that DBD-Threader predictions are more diverse in terms of the number of SCOP superfamilies detected ( 59 vs . 48 ) and are about three times more sensitive than Pfam in assigning putative DNA binding ability to functionally uncharacterized proteins ( 78 positive hits vs . 20 ) . Predictions of DBD-Threader are further compared with PSI-BLAST results in Figure 6 . For each target , we identified the lowest PSI-BLAST E-value of all sequence alignments with all templates . In Figure 6A , the distributions of the lowest PSI-BLAST E-values are given for both positives and negatives predicted by DBD-Threader . One can immediately recognize that most positives share significant sequence similarity with a known DNA-binding domain . About 79% ( 1 , 314 ) of positives hit a significant template at a PSI-BLAST E-value<10−20 . In contrast to positives , only 0 . 3% ( 53 ) of negatives fall into this significant E-value regime . Using the GO annotations , we found that 78% of the 1 , 314 positives belong to the DB set , while only 49% of the 53 predicted negatives belong to DB . On the other hand , the overwhelmingly majority of negatives ( 16 , 642 ) are found within the regime where the E-value is higher than 10−3 . However , DBD-Threader managed to predict 136 positives in this regime , despite low/no sequence similarity . Analysis of their GO annotations found that 13% ( 18 ) of positives belong to DB , the ratio is over four times 2 . 9% ( 476 ) , the rate of negatives classified as DB in the same E-value regime . The comparison suggests that DBD-Threader considerably enriches the predictions of true positives compared to PSI-BLAST . DBD-Threader can make a strong prediction without apparent sequence similarity . This is illustrated through an application to the origin recognition complex subunit 6 ( Orc6 ) , which is a component of the heterohexameric origin recognition complex ( ORC ) . The main function of ORC is to initiate DNA replication , which necessitates DNA-protein interactions [54] . It has been shown experimentally that Orc6 of Drosophila melanogaster binds to DNA [55] . Human Orc6 has a statistically significant sequence similarity to Drosophila Orc6 ( PSI-BLAST E-value = 10−24 ) , though the global sequence identity is relatively low at 30% over ∼240 AAs . It is not clear , however , whether human Orc6 has a similar DNA binding function [55] , [56] . The sequence of human Orc6 was assessed by DBD-Threader , which predicted two DNA-binding domains in the N-terminal region ( residues 1–202 ) , based on a significant hit to the transcription factor TFIIB at a Z-score of 26 and an energy value of −9 . 3 . By contrast , neither PSI-BLAST nor Pfam can detect a significant template from our library , which is not surprising given that there is no apparent sequence similarity between TFIIB and Orc6 . Although the structure of Orc6 has not been experimentally solved , our prediction agrees with a structural model of Drosophila Orc6 that was recently proposed [57] . In addition , point mutations of Ser72 and Lys76 , two residues located within a putative DNA-binding helix-turn-helix motif and conserved between human and Drosophila , abolish the DNA-binding ability of Drosophila Orc6 [55] . It is well-known that function inference based on sequence or structural comparison , even at a statistically significant level of similarity , can be misleading [8] , [20] . By applying the energy based filter , DBD-Threader can reduce false positives generated from structural or sequence similarity comparison . This is illustrated through a second example , the barrier-to-autointegration factor-like ( BAF-L ) protein , whose sequence is about 40% identical to that of BAF , a known DNA-binding protein [58] . The homologous relationship was detected by PSI-BLAST ( E-value<10−46 ) , Pfam ( E-value<10−48 ) , and DBD-Threader ( Z-score = 35 ) . The GO annotations of BAF-L include DNA-binding function , probably inferred from BAL based on sequence similarity . However , using the energy filter , DBD-Threader predicts that BAF-L is not a DNA-binding protein due to its repulsive DNA-protein interaction energy . The prediction is supported by an experimental study which suggests that the functional role of BAF-L is not DNA-binding [59] . Instead , it is proposed to be a regulator of BAF through dimerization with BAF . The prediction is also supported by the fact that most residues involving DNA-binding of BAL are not conserved in BAF-L . Particularly interesting are the classic ( C2H2/C2HC type ) zinc finger domains found in 41% of predicted DNA-binding proteins . The classic zinc finger domain is one of most abundant protein domains encoded in the human genome . According to the domain annotations in the UniProt knowledge base , these are 6 , 873 C2H2/C2HC zinc finger domain matches in 751 protein sequences of the 18 , 621 sequences scanned by DBD-Threader . The vast majority ( 92% ) of these sequences contain multiple zinc finger domains . An interesting question is what functions these domains perform . If one assigns DNA-binding function according to sequence similarity detected by PSI-BLAST or Pfam , all zinc finger domains detected would be assigned as DNA-binding . Although the classic zinc finger domains originally discovered are DNA-binding domains of many transcription factors , recent studies have demonstrated that they can play a functional role through protein-protein interactions ( see reviews , [60] , [61] ) . While DNA-protein and protein-protein interactions are not necessarily mutually exclusive , it is possible that some zinc finger domains play a role involving only protein-protein interactions . About 91% of zinc finger domains annotated in UniProt were detected during threading , and 22% of these significant threading hits were assessed as negatives according to the energy calculations by DBD-Threader . Although there are inevitably false positives/negatives , we speculate that most of these negatives have acquired a functional role that does not involve DNA-binding but other biological interactions , such as protein-protein interactions . To further examine our hypothesis , we compiled from the review in [60] a list of 18 zinc finger domains likely involved only in protein-protein interactions , as shown in Table 5 . These domains , collected from six human sequences , are all experimentally well characterized . Note that we excluded domains with known DNA-binding function from these sequences . If the predictions by DBD-Threader were random , one would expect that a true negative is predicted at a success rate of 22% . Assuming that all 18 domains we collected are true negatives , we further test the null hypothesis that DBD-Threader predicts non-DNA-binding zinc finger domains at a success rate of 22% or less . Among the 18 domains , DBD-Threader predicts 4 positives and 14 negatives , which yields a significant p-value ( 7 . 6×10−7 ) in a one-tailed binomial test . Therefore , we rejected the null hypothesis . The result suggests that the predictions by DBD-Threader are statistically highly significant . Lastly , we examine an intricate example from Table 5 in the transcription factor OAZ ( Olf1/EBF-associated zinc finger protein , also known as ZNF423 ) . This is a 1284 AA long sequence composed of 30 zinc-fingers distributed in several clusters ( Figure 7 ) . The homology of OAZ to other well-characterized zinc finger proteins , such as Zif268 and TFIIIA , were readily established by both PSI-BLAST and DBD-Threader . Significant hits with PSI-BLAST E-values<10−20 and threading Z-scores>15 cover virtually all zinc-finger repeats of OAZ . However , evaluation of the DNA-protein interaction energy by DBD-Threader suggests that only fingers 2 to 6 are DNA-binding , whereas other fingers do not carry out this function due to their highly repulsive energy values ( typically E>10 ) . The prediction is in agreement with two independent experimental studies [62] , [63] . In the former study , the protein was partitioned into six clusters , and the DNA-binding activity of each was assessed with SELEX . Only the cluster containing fingers 2-5 was found to be DNA-binding [62] . The second study was performed on rat OAZ , the ortholog nearly identical ( ∼96% ) to its human counterpart . Consistently , the DNA-binding region was mapped within the first seven fingers of OAZ [63] . In addition , both studies identified the same consensus DNA sequence recognized by these fingers . Among other zinc fingers , it was suggested that the three C-terminal zinc-fingers are essential for the interactions between OAZ and another transcription factor Olf-1/EBF , which regulates olfactory gene expression in rat [63] , [64] . Another study reported that zinc-fingers 14 to 19 mediate the interaction with transcription factors Smad1 and Smad4 , and that zinc-fingers 9 to 13 bind BMP ( bone morphogenetic proteins ) target gene promoters together with Smads [65] . The latter result that fingers 9–13 bind DNA apparently disagrees with the prediction by DBD-Threader . One possible explanation for the discrepancy is that zinc-fingers 9–13 of OAZ may adopt an atypical DNA-binding mode not present in our template library . This is supported by the observation that zinc-fingers 9–13 have unusually long ( >15 AAs ) linkers between them ( Figure 7 ) , whereas other structurally known DNA-binding zinc-finger proteins have shorter linkers , typically six residues , connecting their fingers . In summary , OAZ plays a central role in two distinct processes involving BMP signaling and olfactory neurogenesis , and its multi-functional role is fulfilled by different zinc fingers . While sequence or structure similarity alone cannot distinguish the functional roles of zinc-fingers , which may interact with DNA or other proteins , DBD-Threader provides a means to assess the DNA-binding preference of individual zinc-finger domains .
Previously , threading-based methods were proposed for predicting protein-protein and protein-ligand interactions [39] , [40] . In this study , DBD-Threader extends this idea to the prediction of DNA-binding function . The method employs two key functional discriminating features: fold similarity and DNA-binding propensity . Given a target , sequence threading is used to identify a template that has a similar fold to the target . Compared with standard sequence comparison methods , such as PSI-BLAST , threading is more sensitive in detecting homology , especially when the sequence identity is lower than 30% [41] . However , since threading itself does not differentiate functional roles among sequences with a similar fold , this can give rise to a considerable number of false positives . To reduce the number of false positives , the DNA-protein interaction energy is calculated to assess whether the target preferentially interacts with DNA . In our approach , DBD-Threader uses a statistical pair potential , which has been successfully implemented in our previous application ( DBD-Hunter ) in predicting DNA-binding function given the native protein's structure [19] . Overall , DBD-Threader achieves better performance than approaches using only sequence homology . In benchmark tests on ∼4000 proteins , DBD-Threader is about 15% to 25% higher in sensitivity than PSI-BLAST at the same false positive rate of less than 1% , using templates that share less than 30% sequence identity with the targets . The optimal performance of DBD-Threader has a MCC of 0 . 68 , better than the MCC of 0 . 61 of PROSPECTOR and 0 . 55 of PSI-BLAST , and is comparable to the performance of DBD-Hunter where the experimental structure of the target is required . There exist quite a few template-free methods for predicting DNA-binding function [27] , [28] , [30] , [31] or DNA-binding protein residues [32]–[34] , [36] , [37] , the latter class of methods require the information that the protein is known to be DNA-binding . Most of these methods use machine-learning techniques , which provide no structural and limited biological insights . While these template-free approaches have the potential to predict the DNA-binding sites of a novel fold , their accuracy is generally lower than template-based methods [19] , [37] , and their performance has not been tested in large-scale benchmarks . DBD-Threader , as a template-based method , provides not only function prediction , but also structural insights into the predicted function by identifying the DNA-binding domains and associated DNA-contacting protein residues . In benchmark tests using templates with less than 30% sequence identity to the target , the backbone RMSDs of the top-ranked structural models are within 6 . 5 Å of their native structures for 70% of predicted DNA-binding domains . In addition , the mean sensitivity and specificity of binding site predictions is 60% and 93% among predicted DNA-binding proteins , whose DNA-binding domains have been correctly identified in terms of SCOP superfamily in most cases . The main disadvantage of a template-based approach is that it cannot predict DNA-binding function/sites for structures not present in the template library . In addition , one should generally not expect a high-level of detailed binding-site conservation between a template/target pair at low sequence identity , though the success of DBD-Threader suggests that it tends to identify functionally related template/target pairs , whose DNA-binding sites are significantly similar in most cases . In the post-genomic era , there is a pressing need for accurate , automatic function annotation tools . DBD-Threader , implemented as a fully automated method , contributes to such a task . This is illustrated in the application of DBD-Threader to the human genome . The method predicts 1 , 654 DNA-binding proteins among ∼19 , 000 unique sequences from human . Comparing the results of DBD-Threader to their existing GO annotations , about 68% of the positives by DBD-Threader agree . Most of the remaining predictions have a GO annotation related to DNA-binding , such as transcription factor activity . Therefore , they very likely play a DNA-binding role . Moreover , DBD-Threader predicts a few protein sequences among uncharacterized sequences as DNA-binding . These can serve as candidates for further experimental examination . The predicted DNA-binding proteins from the human genome contain 6 , 896 DNA-binding domains from 59 SCOP superfamilies . The vast majority of these predicted DNA-binding domains are cross-validated by other sequence annotation methods , such as Pfam annotations . The largest population of DNA-binding proteins is the zinc-finger proteins , which are about 41% of predicted DNA-binding proteins . Interestingly , 22% of detected zinc finger domains yield negative results based on the DNA-protein interaction energy assessment . Case studies of these zinc finger domains suggest that they likely perform other biological functions , such as protein-protein interactions , but not direct DNA-binding . Function prediction from protein sequences is a challenging problem . Since proteins are evolving , they can acquire new functions and/or lose old ones . With respect to DNA-binding , a possible scenario is that it evolves to become a regulator of DNA-binding through interactions with other DNA-binding proteins , instead of directly participating in DNA-binding . While such evolution is biologically very interesting , it creates problems for approaches to function inference based on sequence similarity alone , such as those based on PSI-BLAST or HMMs . By assessing DNA-binding propensity through use of the DNA-protein interaction energy , DBD-Threader can help to discriminate DNA-binding from other functional roles , thus improving the overall quality of the predictions . Application of the method generates not only potentially interesting positives , but also negatives evolved from direct DNA-binding . Through this study , we identified 22% of zinc finger domains annotated in the human genome as such negatives with DBD-Threader .
All datasets listed below , the statistical potential parameters , prediction results on the human genome , and a web-server implementation of DBD-Threader are freely available at http://cssb . biology . gatech . edu/skolnick/files/ . The method DBD-Threader has three main modules: sequence threading , domain partition , and function prediction . Sequence threading was conducted using the in-house program PROSPECTOR [41] . The purpose of threading is to examine whether a target sequence encodes a structural fold similar to any structurally known DNA-binding domain . Specifically , the target sequence is threaded sequentially against two template libraries . The first library is a regular template library composed of ∼8000 protein structures , which share less than 35% global sequence identity among each other; the second library is composed of DNA-binding domains described above . The target is first threaded against the regular template library , generating statistically more robust mean and standard deviation of threading scores than threading directly on the much smaller library of DNA-binding domains . Then , the mean and standard deviation are used to calculate the Z-scores when threading templates of the DNA-binding domain library are used . Note that in the benchmark test on DB179/NB3797 , we excluded all templates with more than 30% sequence identity from both threading libraries for any target . In the application to the human genome , the exclusion rule was eliminated . For each pair of target/template proteins , a corresponding Z-score is calculated as where S is the score associated with the best alignment between the pair , and quantity in angle brackets denotes the mean of the quantity over all entries in the regular template library . Based on our benchmark results , we consider templates with Z-scores>6 as significant hits , which are then ranked according to their Z-scores . Since most DNA-binding proteins are composed of multiple domains , it is necessary to locate the domain ( s ) that directly fulfill DNA-binding function . To this end , an iterative clustering procedure was implemented to partition domains of the target sequence based on significant template hits . Clustering is required because a DNA-binding protein may contain multiple DNA-binding domains , which can hit different sets of templates . Initially , the top Z-score-ranked template is chosen as the clustering seed , and all significant templates having more than 50% overlap with respect to the seed are moved to this cluster , and excluded from subsequent clustering . After this process , if there is any template left , the highest ranked template remaining is used as a new clustering seed , and this clustering procedure is repeated until no template is left . The clustering is used to consolidate redundant templates that hit the same sequence region , and a domain can then be defined according to the alignment of a seed to the target . The threading and partition procedures are iterated for any sequence region without a significant hit that is longer than 40 amino acids , until no new domain is found . This iterative procedure can reduce missing hits to domain repeats , e . g . , zinc finger clusters , because threading itself only returns the most significant alignment from each template in each round . For function prediction , we evaluate the DNA-protein interaction energy and use it to assess DNA-binding propensity . Here , we consider only significant templates hits whose DNA-protein contacts have been obtained beforehand using the experimentally determined DNA-protein complex structures . The contacts between the target and a corresponding template DNA are inferred by replacing original template protein residues with aligned target residues . The protein-DNA interaction energy is then calculated using these contacts and a statistical pairwise potential developed previously [19] . Negative and positive energy values indicate attractive and repulsive interactions , respectively . A target is predicted to be a DNA-binding protein if at least one template yields an energy value below a specified threshold , and non-DNA-binding if no template satisfies the energy criterion . Finally , the SCOP superfamily domain assignment is inferred from the highest Z-score-ranked template that satisfies the energy criteria , and corresponding DNA-binding residues are also transferred from this template . The SCOP superfamily prediction will be skipped if the top template does not have official SCOP classification . The optimal energy threshold values determined in benchmark tests on DB179/NB3797 are shown in Table 6 . Depending on the threading Z-score of their templates , the targets fall into two regimes: Medium ( 20≥Z-score>6 ) and Easy ( Z-score>20 ) . In each regime , we select an optimal energy threshold that gives the highest MCC of predictions on DB179 and NB3797 . As one expected , a more permissive energy value is obtained for the Easy targets . In the benchmarks , the ROC and PR curves of DBD-Threader were obtained by varying the energy threshold for templates in the Medium regime , and use the optimal energy threshold for templates in the Easy regime . The optimized values were adopted in the application to the human genome . DBD-Threader was compared with three alternative approaches: DBD-Hunter [19] , PROSPECTOR [41] , and PSI-BLAST [11] . To ensure fair comparison , the same template library and benchmark sets DB179/NB3797 were employed . In case of DBD-Hunter , structures of targets were used as input , and the results obtained with optimized parameters are reported . When applying PROSPECTOR , the threading Z-score was used as the criterion for predictions . A target protein is classified as DNA-binding if it hits a template with a Z-score higher than a specified threshold and as a non-DNA-binding otherwise . When applying PSI-BLAST ( version 2 . 2 . 17 ) , two position specific sequence profiles were derived separately for each target using two libraries: the NCBI-NR protein sequence library ( the Jul 2007 release ) , and the UniProt sequence library ( UniRef100 version 15 . 5 ) . Each profile was obtained using up to four rounds of scanning the respective libraries . We tested up to twenty rounds of iterations for profile derivation and found that four rounds gave the best performance in our benchmark tests . An inclusion E-value threshold of 0 . 001 and default values for other arguments were employed . For each profile generated , a final PSI-BLAST run was performed on the sequence library of the DNA-binding protein templates . If a target hits a template with an E-value higher than the specified threshold , then the target is classified as being a DNA-binding protein; otherwise , it is classified as a non-DNA-binding protein . For each target in the benchmark tests , its homologs with global sequence identity >30% were excluded from the template library of DNA-binding proteins . Note that the exclusion rule was not applied during the derivation of the PSI-BLAST profiles , and we allow all sequence hits for building the profiles . In each prediction scenario , the numbers of true positives , false positives , true negatives and false negatives are designated as TP , FP , TN , and FN , respectively . In case of DNA-binding function prediction , a TP refers to a protein sequence correctly predicted as DNA-binding protein; in case of DNA-binding site prediction , TP refers to an amino acid correctly assigned as a DNA-binding residue . Performance measures are defined as the following:
|
DNA-binding proteins represent only a small fraction of proteins encoded in genomes , yet they play a critical role in a variety of biological activities . Identifying these proteins and understanding how they function are important issues . The structures of solved DNA protein complexes of different protein families provide an invaluable knowledge base not only for understanding DNA-protein interactions , but also for developing methods that predict whether or not a protein binds DNA . While such methods are useful , they require an experimental structure as input . To overcome this obstacle , we have developed a threading-based method for the prediction of DNA-binding domains and associated DNA-binding protein residues from protein sequence . The method has higher accuracy in large scale benchmarking than methods based on sequence similarity alone . Application to the human proteome identified potential targets of not only previously unknown DNA-binding proteins , but also of biologically interesting ones that are related to , yet evolved from , DNA-binding proteins .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"computational",
"biology/protein",
"structure",
"prediction",
"biophysics/biomacromolecule-ligand",
"interactions",
"computational",
"biology/macromolecular",
"sequence",
"analysis",
"computational",
"biology/protein",
"homology",
"detection"
] |
2009
|
A Threading-Based Method for the Prediction of DNA-Binding Proteins with Application to the Human Genome
|
The soil transmitted helminths are a group of parasitic worms responsible for extensive morbidity in many of the world’s most economically depressed locations . With growing emphasis on disease mapping and eradication , the availability of accurate and cost-effective diagnostic measures is of paramount importance to global control and elimination efforts . While real-time PCR-based molecular detection assays have shown great promise , to date , these assays have utilized sub-optimal targets . By performing next-generation sequencing-based repeat analyses , we have identified high copy-number , non-coding DNA sequences from a series of soil transmitted pathogens . We have used these repetitive DNA elements as targets in the development of novel , multi-parallel , PCR-based diagnostic assays . Utilizing next-generation sequencing and the Galaxy-based RepeatExplorer web server , we performed repeat DNA analysis on five species of soil transmitted helminths ( Necator americanus , Ancylostoma duodenale , Trichuris trichiura , Ascaris lumbricoides , and Strongyloides stercoralis ) . Employing high copy-number , non-coding repeat DNA sequences as targets , novel real-time PCR assays were designed , and assays were tested against established molecular detection methods . Each assay provided consistent detection of genomic DNA at quantities of 2 fg or less , demonstrated species-specificity , and showed an improved limit of detection over the existing , proven PCR-based assay . The utilization of next-generation sequencing-based repeat DNA analysis methodologies for the identification of molecular diagnostic targets has the ability to improve assay species-specificity and limits of detection . By exploiting such high copy-number repeat sequences , the assays described here will facilitate soil transmitted helminth diagnostic efforts . We recommend similar analyses when designing PCR-based diagnostic tests for the detection of other eukaryotic pathogens .
Estimated to infect more than one quarter of the world’s total population , the soil transmitted helminths ( STH ) are responsible for profound morbidity and nutritional insufficiency [1] . Concentrated in the world’s most impoverished locations , the results of widespread infection on economic capacity are equally burdensome . Yet despite the scope of such disease , and continuing efforts to improve treatment programs and integration strategies , reliable and accurate diagnosis of STH infections remains difficult , and resulting prevalence estimates remain imprecise [1–2] . In recent years , the interest in molecular diagnostic methods for the detection of gastrointestinal helminths has grown substantially . Largely , this escalation in interest has occurred in parallel with the belief that standard microscopy-based methodologies for the examination of stool samples are sub-optimal , leading to underrepresentation of infection [3–5] . Further complicating matters , rates of STH egg/larval excretion have been shown to vary considerably within sequentially collected stool samples originating from a single infected individual [6–7] . This variability in egg/larval count can result in false negative samples , particularly when non-amplification-based diagnostic methodologies are utilized [7] . Such underrepresentation of disease complicates programmatic efforts , making the accurate assessment of the effects of intervention difficult , and frequently leaving low-level infections undiagnosed [5 , 8–9] . Additionally , microscopy-based diagnostic methods have been linked with pathogen misidentification due to the morphological similarities that exist between species [5 , 10] . Because of such concerns , a number of conventional and real-time PCR-based assays have been developed with the objective of improving both species-specificity and limits of detection [4 , 11–17] . These assays have proven valuable , and as global efforts to estimate the burden of disease caused by the soil transmitted helminths ( STHs ) continue to increase , the number of studies incorporating such assays has risen in response [3 , 5 , 9 , 18–21] . To date , the target sequences for such assays have been ribosomal internal transcribed spacer ( ITS ) sequences , 18S or ribosomal subunit sequences , or mitochondrial genes such as cytochrome oxidase I ( COI ) [4 , 11–14] . Ribosomal sequences have been selected as diagnostic targets because they are typically found as easily identified moderate copy number tandem repeats in nucleated organisms [22–25] . Similarly , multiple copies of mitochondrial targets are found in the vast majority of eukaryotic cells [26] , making them attractive target choices . However , while effective , such diagnostic targets are often sub-optimal . This is particularly true in the case of nematodes and other multi-cellular organisms where species-specific , highly repetitive DNA elements frequently make up a substantial portion of the genome , and are often present at copy-numbers exceeding 1 , 000 per haploid genome [27–29] . Due to such overrepresentation , non-coding repetitive sequence elements have become the targets of choice for many PCR-based diagnostic assays for the detection of various helminth species [30–31] . However , the identification of such repeats has historically been complicated and labor intensive . This identification has relied on techniques such as restriction endonuclease digestion of genomic DNA , followed by gel electrophoresis and Sanger DNA sequencing or polyacrylamide slab gel sequencing [32–34] . However , the advent of next-generation sequencing ( NGS ) technologies and associated informatics tools has expedited the search for highly repetitive sequence elements [35–39] , and greater confidence can be placed in the accuracy of the results of such searches . Furthermore , as ribosomal and mitochondrial sequences tend to demonstrate high degrees of conservation between species , species-specificity of detection is also improved through the targeting of unique , highly-divergent , non-coding repeat DNA elements . Here we describe the development of a multi-parallel real-time PCR assay for the detection of five species of soil transmitted helminths ( Necator americanus , Ancylostoma duodenale , Trichuris trichiura , Strongyloides stercoralis , and Ascaris lumbricoides ) . Using NGS-generated sequence data and the Galaxy-based RepeatExplorer computational pipeline [38–39] , we have searched the genomes of each organism for highly repetitive , non-coding DNA elements in order to identify diagnostic targets capable of providing optimal limits of detection and species-specificity of detection . Using these targets to design small-volume , multi-parallel tests [4] , we have created a platform that provides cost-minimizing implementation of only those assays appropriate for a specific geographic region based upon the infections present . While performing multiplex assays may provide labor and time savings in locations where many parasites are co-endemic , such assays result in considerable waste when used in areas harboring only one or a few of the target species . In such settings , the “pick-and-choose” nature of multi-parallel assays minimizes reagent waste , and by improving upon limits of detection , the species-specific platform we describe here should facilitate improved STH monitoring and mapping efforts . Since NGS-based repeat analyses allow for the selection of the most efficacious target sequences , this approach to assay design should be applied to the development of additional diagnostics tests for other eukaryotic pathogens .
For isolation of genomic DNA from N . americanus , A . duodenale , and T . trichiura , extractions were performed on cryopreserved adult worms in accordance with the “SWDNA1” protocol available on the Filarial Research Reagent Resource Center website ( http://www . filariasiscenter . org/parasite-resources/Protocols/materials-1/ ) . For N . americanus and A . duodenale , DNA extractions were conducted using a pool of approximately 10 adult worms . Both hookworm species belonged to strains originating in China . In the case of T . trichiura , extraction was performed using a single adult female worm of Ugandan origin . For S . stercoralis and A . lumbricoides , previously extracted genomic DNA was received from collaborators . S . stercoralis DNA was obtained from laboratory-reared worms originating from Pennsylvania , USA , and A . lumbricoides DNA was isolated from worms obtained from Ecuador . For each parasite analyzed , raw sequencing reads were uploaded to the Galaxy-based RepeatExplorer web server [39] . Reads were processed according to the workflow in Fig 1 , enabling the identification of high copy-number repeat DNA sequences for each organism . Promising repeat families were further analyzed using the Nucleotide BLAST tool ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) available from the National Center for Biotechnology Information ( NCBI ) . Results from each organism were screened for repetitive DNA elements found to have high degrees of homology with elements of the human genome , common bacteria of the human microbiome , or other parasitic organisms likely to be found within the human gut . Had such sequences been identified as among the most repetitive , they would have been eliminated from further consideration as they would be expected to cause species-specificity challenges during downstream PCR assay development . However , no such conserved highly repetitive elements were identified . Following screening , sequences from each organism , putatively determined to be among the most highly repetitive , were utilized for further assay development ( Fig 2 ) . Candidate primer and probe pairings for each organism , excluding A . lumbricoides , were designed using the PrimerQuest online tool ( Integrated DNA Technologies , Coralville , IA ) , utilizing the default parameters for probe-based qPCR . The putative species-specificity of each primer pair was further examined using Primer-BLAST software ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) . In the case of S . stercoralis the highest copy-number repeat ( as determined by RepeatExplorer ) was not selected as a target sequence , due to design difficulties associated with the extreme A-T richness of the repeat ( A-T % = 80 . 25 ) . As a result , a second repeat analysis was performed , selecting only for sequence reads with > 30% G-C content , and a second candidate sequence was selected based on these results . In the case of A . lumbricoides , RepeatExplorer analyses of two different sequencing runs performed from two distinct libraries both resulted in the identification of ribosomal and mitochondrial sequences as the most highly repetitive . For this reason , sequences from an existing , proven , primer and probe set targeting the ITS1 region were selected for further analysis [14 , 16] . With the exception of the previously published A . lumbricoides probe , all probes were labeled with a 6FAM fluorophore at the 5’ end , and were double quenched using the internal quencher ZEN and 3IABkFQ ( IOWA BLACK ) at the 3’ end ( Integrated DNA Technologies ) . This fluorophore-quencher combination was chosen as comparative testing of each probe revealed improved Ct values and greater ΔRn values using this chemistry when compared to typical TAMRA quenching ( Fig 3 ) . Primer and probe sets for each organism can be found in Table 1 .
A panel of 79 blindly-coded patient samples , obtained in Timor-Leste as part of a previously described study [42] , was tested using the newly described multi-parallel Smith assays , as well as the previously described , multiplex real-time PCR detection methodology ( QIMR assay ) ( Table 2 , S1 Table ) . As samples were patient-obtained and no true “gold standard” exists for the detection of the various STH infections examined here , it is difficult to definitively determine whether increased sample positivity is a result of improved assay detection limits or non-specific , off-target amplification . For this reason , the comparative performances of each assay were assessed through calculations of positive , negative , and overall agreement [43] . For the detection of N . americanus , a positive agreement ( PA ) of 100% and a negative agreement ( NA ) of 61% were calculated . This resulted in an overall agreement ( PO ) of 85% ( Kappa 0 . 658 ) . Use of the Smith assay resulted in the detection of 60 positive samples , while the QIMR assay resulted in the detection of 48 positives . All 48 QIMR-positive samples were among the 60 positive samples detected using the Smith methodology . For the detection of A . lumbricoides , a PA of 100% , an NA of 82% , and a PO of 91% ( Kappa 0 . 822 ) were seen . The Smith assay for A . lumbricoides detection resulted in the identification of 47 positive samples , while the corresponding QIMR assay resulted in 40 positives . Again , all 40 QIMR-positive samples were among the 47 Smith-positive samples which were identified . Detection of Trichuris gave a PA of 71% , an NA of 88% and a PO of 85% ( Kappa 0 . 580 ) . Sample examination using the Smith assay identified 18 positive extracts , while examination with the QIMR assay identified 14 positives . However , only 10 positives were common to both assays , with 8 samples identified as positive only by the Smith assay , and 4 samples demonstrating the presence of parasite DNA using only the QIMR methodology . Amplification in control reactions demonstrated that the QIMR assay , but not the Smith assay , would provide for the detection of the closely related parasite Trichuris vulpis , a whipworm species common to canines , but also known to cause zoonotic infection [48–49] . As Trichuris ssp . including T . vulpis , Trichuris suis , and Trichuris ovis have a wide geographic distribution with increased prevalence in tropical and sub-tropical locations [50–51] , the four QIMR-positive , Smith-negative samples were sequenced to determine the identity of the Trichuris species present within these samples . BLAST analysis indicated that two of the samples contained DNA from the ruminant parasite T . ovis ( E values = 0 . 0 ) . Unfortunately , two independent trials failed to produce usable sequence for the remaining two samples , after which both sample stocks had been exhausted , making further examination impossible . Examination of all 79 samples for the presence of S . stercoralis resulted in the detection of only a single positive sample . This single sample was identified using both the Smith and QIMR assays . Sample examination for the presence of Ancylostoma resulted in the identification of 22 Ancylostoma ssp . positive samples using the QIMR methodology . However , not a single A . duodenale-positive sample was identified using the Smith assay . As the zoonotic parasite Ancylostoma ceylanicum has been suspected of causing human infection in Timor-Leste [52] , a previously described , semi-nested PCR-RFLP assay was employed to discriminate infection with A . duodenale from infection with A . ceylanicum [47] . In this assay , an MvaI restriction digest of PCR product is indicative of the presence of A . ceylanicum , while digestion with Psp1406I is indicative of A . duodenale . 21 of the 22 Ancylostoma ssp . positive samples were digested by MvaI , identifying the infections as A . ceylanicum in origin . Two independent PCR trials ( four replicates ) failed to amplify the remaining Ancylostoma ssp . -positive sample , preventing a definitive determination of the identity of the parasite in that sample . Because a sizeable panel of field-collected samples was analyzed using the two different real-time PCR methodologies discussed here , a comparison of Ct values was conducted for all samples testing positive for a given parasite by both the Smith and QIMR methods ( S1 Table ) . All 10 samples demonstrating positive results for T . trichiura when tested by both assays showed lower Ct values using the Smith methodology ( mean difference in Ct value = 7 . 86 +/- 2 . 46 ) . Examination for N . americanus resulted in a similar pattern , with all 48 samples testing positive by both methodologies possessing lower Ct values when tested using the Smith assay ( mean difference in Ct value = 4 . 94 +/- 1 . 22 ) . In the case of A . lumbricoides , Ct values were lower using the QIMR methodology for 38 of 40 samples demonstrating positive results for both assays . However , at 0 . 896 +/- 0 . 767 , the mean difference in Ct values was low . For S . stercoralis testing , only a single positive sample was identified . This sample possessed a lower Ct value when tested using the Smith assay . As no samples tested positive for Ancylostoma using the Smith assay ( QIMR-positive samples were demonstrated to be A . ceylanicum ) , a Ct comparison could not be made .
In light of their impact on global health , the importance of optimal and species-specific diagnostic methods for the detection of soil transmitted helminths cannot be overestimated . While current molecular assays making use of ribosomal and mitochondrial targets have vastly improved the diagnosis of STH infection , these targets are frequently sub-optimal , potentially leaving low-level infections undiagnosed . Furthermore , such sequences may lack the species-specificity required to discriminate between different species of the same genus . In contrast , assays targeting high copy-number repetitive sequences improve upon assay detection limits , as many eukaryotic pathogens contain large numbers of such non-coding repeat DNA elements . Accordingly , by coupling the high throughput nature of NGS with the Galaxy-based RepeatExplorer computational pipeline , a cost effective , accurate , and expedited methodology for the identification of high copy-number repeat DNA elements was developed . Through the design of real-time PCR primer/probe pairings that uniquely target such repetitive sequences in a species-specific manner , diagnostic accuracy and limits of detection are improved dramatically when compared with microscopy-based diagnostic techniques and PCR-based diagnostics targeting mitochondrial or ribosomal sequences . Utilizing this strategy , we have successfully identified novel target sequences for the detection of N . americanus , A . duodenale , T . trichiura , and S . stercoralis . Furthermore , we have demonstrated the consistent detection of genomic DNA from each target organism at quantities of 2 fg or less , and have presented evidence to suggest improved limits of detection and species-specificity relative to an established and validated PCR diagnostic methodology [Llewellyn , 2016] . Although further testing utilizing “spiked” samples containing known quantities of eggs/larvae is currently underway , 2 fg of DNA is far less than the quantity present within a single fertilized egg or L1 larvae of each species [53–55] ( Table 3 ) . In principle , we have therefore demonstrated the potential of these assays to detect a single egg within a tested patient stool sample . While the high copy-number nature of non-coding repetitive sequence elements makes them attractive diagnostic targets , such elements also frequently demonstrate rapid evolutionary divergence [56–57] . This divergence increases the diagnostic appeal of these sequences , as divergence reduces the risk for non-specific , off-target amplification , a characteristic essential for the development of species-specific PCR assays capable of discriminating between closely related organisms . Accordingly , while additional testing against genomic DNA from a growing panel of closely related parasites will continue to be used to evaluate the species-specificity of each selected primer/probe set , we have successfully demonstrated that each Smith assay does not amplify off-target templates from any other parasite species included within this multi-parallel platform . Furthermore , by employing a semi-nested PCR-RFLP tool , we were able to successfully demonstrate that our assay for the detection of A . duodenale does not amplify the closely related parasite A . ceylanicum . In contrast , the previously published primer/probe set employed for comparative testing was unable to distinguish between these two species , resulting in consistent off-target amplification of A . ceylanicum DNA . Similarly , while our T . trichiura assay failed to amplify four samples containing genetic material from Trichuris ssp . , the comparative QIMR assay again demonstrated non-specific , off-target amplification for at least two of these samples , as sequence analysis demonstrated the presence of DNA from the ruminant parasite T . ovis . Taken together , these findings support the notion that improved assay species-specificity results from non-coding , repeat-based PCR assay design . Of note , to our knowledge , this is the first example of T . ovis potentially serving as a causative agent of zoonotic infection . However , as sheep are considered a major agricultural commodity of Timor-Leste [58] , the possibility exists that individuals testing positive for T . ovis may have ingested intestinal material from an animal harboring infection , making it conceivable that the T . ovis DNA present was not the result of zoonotic infection . Given that T . ovis is not known to cause human infection , further exploration of this possible zoonosis is warranted . Attempting to design a non-coding , repetitive DNA sequence-based assay for the species-specific detection of A . lumbricoides presented a unique set of challenges . A . lumbricoides , like many species of Ascaridae , discards large portions of its highly repetitive , non-coding genomic DNA during embryonic development . This process , known as chromosome diminution , eliminates the presence of such DNA elements from post-embryonic somatic cells [59–61] . Presumably for this reason , two separate repeat analyses , performed on two distinct library preparations , failed to identify any repetitive sequences with copy numbers greater than ribosomal and mitochondrial targets . Accordingly , a previously described primer/probe set targeting the ITS1 ribosomal region was chosen for inclusion in our multi-parallel platform [14 , 16] . In order to improve diagnostics for this parasite , further analysis of A . lumbricoides using DNA extracted from eggs alone ( before chromosome diminution ) will be undertaken . In addition to the potential detection limit improvements and species-specificity gains realized when diagnostically targeting non-coding repetitive DNA sequences , designing multi-parallel assays provides another unique set of advantages over previous design strategies [4] . By reducing the number of tests required , multiplex assays can provide labor and reagent savings over alternative diagnostic measures when used in environments that harbor the full complement of organisms targeted by the assay [62–63] . However , as the geographic distribution of STH species is not uniform , the use of multi-parallel assays makes it possible to select only the assays appropriate for a given location , reducing primer/probe costs associated with testing for unnecessary targets [4] . By running these assays as “small-volume” 7 μl reactions , reagent use is minimized , resulting in cost savings . Furthermore , as multi-parallel reactions are run independently , this enables the development of new assays for new pathogens and their subsequent addition to the testing platform without the complex re-optimization of assay conditions required for multiplex PCR assays . While reagent costs associated with performing molecular diagnostic testing are higher than costs associated with conducting traditional microscopy-based diagnostics , expenses associated with molecular techniques are declining as improved reagents and enzymes have allowed reaction volumes to decrease , minimizing reagent needs [4 , 64] . Furthermore , reagent improvements have increased the practicality of sample pooling , a practice already adopted by many tropical disease surveillance and diagnostic efforts [65–69] . Such pooling allows for cost-reducing high-throughput screening of stool samples [70–71] . Thus , while the total cost associated with performing a duplicate Kato-Katz thick smear under field conditions has been estimated at $2 . 06 [72] and we estimate the total cost associated with the duplicate testing a single stool sample using all five multi-parallel assays to be approximately $10 , the pooling of as few as five samples would render small volume , multi-parallel PCR testing more cost effective than Kato-Katz testing . Furthermore , molecular diagnostic accuracy and reliability provide increased clarity of results [64] , allowing for the implementation of more informed and effective treatment and control strategies . Such improvements in efficiency result in greater programmatic gains , drastically reducing long-term costs and expenses of control or elimination programs . One profound shortcoming which hampers STH diagnostic development is the lack of a reliable gold standard for detection [8] . While still used in many clinical , mapping , and research efforts , microscopy-based methodologies are known to lack both adequate limits of detection and species-specificity of detection [3–5 , 10 , 64] . Similarly , while currently available molecular methods have greatly improved upon many of the shortcomings inherent to microscopy , the use of sub-optimal ribosomal or mitochondrial targets possessing relatively high degrees of conservation can result in both false-negative , and off-target , false-positive results . Thus , a gold standard of detection is sorely needed . Unfortunately , without a definitive method for assigning positive/negative status to an unknown sample , distinguishing improved limits of detection from false-positive amplification can be difficult . Nonetheless , comparative assay testing remains an important aspect of designing any diagnostic test . As such , we believe the evaluation of Timor-Leste patient samples presented in this paper provides strong evidence for improved limits of detection when utilizing the newly described Smith assays . While strain-specific genetic differences arising within divergent geographic isolates could present detection challenges , testing on a limited number of patient-derived samples from Argentina and Ethiopia aimed at providing evidence for the global applicability of these multi-parallel assays is currently underway . Additional studies to further validate these assays on a variety of geographic isolates will continue . In all instances , and for all parasites excluding Ancylostoma and Trichuris ( where off-target amplification of A . ceylanicum and T . ovis by the QIMR assay was demonstrated ) , each Timor-Leste patient sample that provided a positive QIMR assay result also demonstrated positivity with the corresponding Smith assay . Furthermore , all N . americanus , T . trichiura , and S . stercoralis samples that were positive by both assays exhibited lower Ct values for the Smith assay results . These findings strongly suggest improved limits of detection for the Smith assays , and support our contention that samples returning Smith assay positive results , but QIMR assay negative results , are likely low-level positives escaping detection by the sub-optimal PCR platform . This conclusion is further supported by the finding that the Smith assays do not show off-target amplification of any other STH parasites , human DNA or E . coli DNA . As both the QIMR and Smith assays for the detection of A . lumbricoides make use of the same previously published primer/probe combination [14 , 16] , comparative assay testing for this parasite provided results which were more difficult to interpret . As increased reaction volumes are known to frequently improve detection limits for an assay , likely due to the large volume nature of the QIMR assay ( 25 μl vs . 7 μl for Smith ) , 38 of 40 samples returning positive results for both testing platforms demonstrated lower Ct values when examined using the QIMR method . Interestingly , despite this tendency for QIMR testing to result in lower Ct values , seven samples identified as positive using the Smith assay were found to be QIMR-negative . In contrast , not a single sample was found to be QIMR-positive and Smith-negative . As the QIMR assays are multiplexed , one explanation for this apparent contradiction is that the multiplex methodology failed to detect A . lumbricoides in a subset of samples that were positive for multiple STH parasites ( S1 Table ) . Such failures are known to occur in multiplex reactions , particularly when primer concentrations are suboptimal , as reagents are utilized for the amplification of a more prevalent target , preventing the amplification of the lower copy-number target sequences within the sample [73] . Alternatively , while the results of our assay specificity testing present compelling evidence to the contrary , the possibility of false positive amplification cannot be definitively ruled out . Non-coding repetitive DNA elements are found in nearly all eukaryotic organisms . Such sequences are typically highly divergent , and frequently exist in high copy-number . These characteristics make them ideal molecular diagnostic targets , particularly for the detection of pathogens such as the STHs , which remain an underdiagnosed , poorly mapped global health concern . By applying next-generation sequencing technology to the challenge of repeat DNA discovery , we have designed highly specific multi-parallel PCR assays with improved limits of detection over existing diagnostic platforms . We believe that these assays will greatly aid in the global efforts to map STH infection , facilitating accurate disease prevalence estimates . Furthermore , we intend to apply this approach to molecular target discovery of other parasitic organisms and NTDs , as optimal limits of detection and species-specificity of detection are vital to all diagnostic efforts . This is particularly true when implementing diagnostics in climates of decreasing disease prevalence . Accordingly , as NTD elimination efforts continue to progress , optimized assays will play an increasingly critical role in the detection of sporadic and focal infections and the monitoring for disease recrudescence .
|
With a growing emphasis on the mapping and elimination of soil transmitted helminth ( STH ) infections , the need for optimal and specific diagnostic methods is increasing . While PCR-based diagnostic methods for the detection of these parasitic organisms exist , these assays make use of sub-optimal target sequences . By designing assays that target non-coding , high copy-number repetitive sequences , both the limit of detection and species-specificity of detection improve . Using next-generation sequencing technology , we have identified high copy-number repeats for a series of STH species responsible for the greatest burden of disease . Using these repetitive sequences as targets in the design of novel real-time PCR assays , we have improved both the limits of detection and species-specificity of detection , and we have demonstrated this improved detection by testing these assays against an established PCR-based diagnostic methodology . Accordingly , these assays should facilitate mapping and monitoring efforts , and the generalized application of this approach to assay design should improve detection efforts for other eukaryotic pathogens .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"sequencing",
"techniques",
"invertebrates",
"animals",
"necator",
"americanus",
"ascaris",
"ascaris",
"lumbricoides",
"molecular",
"biology",
"techniques",
"strongyloides",
"stercoralis",
"research",
"and",
"analysis",
"methods",
"ancylostoma",
"sequence",
"analysis",
"strongyloides",
"trichuris",
"artificial",
"gene",
"amplification",
"and",
"extension",
"necator",
"repeated",
"sequences",
"molecular",
"biology",
"dna",
"sequence",
"analysis",
"polymerase",
"chain",
"reaction",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"genomics",
"organisms"
] |
2016
|
Improved PCR-Based Detection of Soil Transmitted Helminth Infections Using a Next-Generation Sequencing Approach to Assay Design
|
Large-scale white matter pathways crisscrossing the cortex create a complex pattern of connectivity that underlies human cognitive function . Generative mechanisms for this architecture have been difficult to identify in part because little is known in general about mechanistic drivers of structured networks . Here we contrast network properties derived from diffusion spectrum imaging data of the human brain with 13 synthetic network models chosen to probe the roles of physical network embedding and temporal network growth . We characterize both the empirical and synthetic networks using familiar graph metrics , but presented here in a more complete statistical form , as scatter plots and distributions , to reveal the full range of variability of each measure across scales in the network . We focus specifically on the degree distribution , degree assortativity , hierarchy , topological Rentian scaling , and topological fractal scaling—in addition to several summary statistics , including the mean clustering coefficient , the shortest path-length , and the network diameter . The models are investigated in a progressive , branching sequence , aimed at capturing different elements thought to be important in the brain , and range from simple random and regular networks , to models that incorporate specific growth rules and constraints . We find that synthetic models that constrain the network nodes to be physically embedded in anatomical brain regions tend to produce distributions that are most similar to the corresponding measurements for the brain . We also find that network models hardcoded to display one network property ( e . g . , assortativity ) do not in general simultaneously display a second ( e . g . , hierarchy ) . This relative independence of network properties suggests that multiple neurobiological mechanisms might be at play in the development of human brain network architecture . Together , the network models that we develop and employ provide a potentially useful starting point for the statistical inference of brain network structure from neuroimaging data .
Increasing resolution of noninvasive neuroimaging methods for quantifying structural brain organization in humans has inspired a great deal of theoretical activity [1]–[4] , aimed at developing methods to understand , diagnose , and predict aspects of human development and behavior based on underlying organizational principles deduced from these measurements [5]–[7] . Ultimately , the brain is a network , composed of neuronal cell bodies residing in cortical grey matter regions , joined by axons , protected by myelin . Diffusion-weighted magnetic resonance imaging methods trace these white matter connections , based on the diffusion of water molecules through the axonal fiber bundles . While resolution has not reached the level of individual neurons and axons , these methods lead to reliable estimates of the density of connections between regions and fiber path lengths . The result is a weighted adjacency matrix , with a size and complexity that increases with the resolution of the measurements [8] , [9] . The immense complexity of this data makes it difficult to directly deduce the underlying mechanisms that may lead to fundamental patterns of organization and development in the brain [10] . As a result , comparison studies with synthetic network models , employing quantitative graph statistics to reduce the data to a smaller number of diagnostics , have provided valuable insights [11]–[15] . These models and statistics provide a vehicle to compare neuroimaging data with corresponding measurements for well-characterized network null models . However , the methods are still in development [16]–[18] , and vulnerable to the loss of critical information through oversimplification of complex , structured data sets , by restricting comparisons to coarse measurements that ignore variability [10] , [19] , [20] . Two critical questions motivate development of network methodologies for the brain . The first question focuses on predictive statistics: Are there graph metrics that may ultimately be useful in parsing individual differences and diagnosing diseases ? Comparing empirical brain data to benchmark null models can establish the statistical significance of a topological property [21]–[23] , and normalizing a topological property by its null model surrogate can be a useful preprocessing step prior to the determination of statistical differences in brain network structure between groups [16] . The second question focuses on network characteristics from a fundamental , development and evolutionary perspective: What organizational principles underlie growth in the human brain ? Here comparing empirical brain data to simplified model networks that have been created to capture some aspect of , for example , neurodevelopmental growth rules [24] , neuronal functions [11] , or physiological constraints [25] may aid in developing a mechanistic understanding of the brain's network architecture ( e . g . , [26]–[28] ) . Both efforts require a basic understanding of the topological similarities and differences between synthetic networks and empirical data . In this paper , we perform a sequence of detailed , topological comparisons between empirical brain networks obtained from diffusion imaging data and 13 synthetic network models ( see Table 1 ) . The models are investigated in a tree-like branching order , beginning with the simplest , random or regular graphs , and progressively adding complexity and constraints ( see Figure 1 ) . The objective of this investigation is to determine , in a controlled , synthetic setting , the impact of network properties on the topological measurements . Our goal is not to create a definitive network model of the brain , but to gain an intuition for structural drivers of network statistics and to create a battery of null models to be used in statistical comparisons of brain networks . At the coarsest level in the model hierarchy , we distinguish between synthetic networks that are constructed purely based on rules for connectivity between nodes ( non-embedded ) , and those that constrain nodes to reside in anatomical brain regions ( embedded ) ( see Figure 1 ) . While non-embedded models are frequently used for statistical inference , recent evidence has suggested that physical , embedding constraints may have important implications for the topology of the brain's large-scale anatomical connectivity [2] , [8] , [22] , [26]–[29] . By examining both non-embedded and embedded models , we hope our results will help to guide the use , development , and understanding of more biologically realistic models for both statistical and mechanistic purposes [23] , [30] . A second important classification of the synthetic models in our study separates those obtained from static ensembles with fixed statistical properties and those generated using mechanistic growth rules ( see Figure 1 ) . While algorithms for generating networks based on static sampling and growth rules ultimately both produce ensembles of fixed graphs for our comparison with data , additional constraints imposed by underlying growth rules may facilitate understanding of mechanisms for development and evolution in the brain as well as other biological and technological networks . To compare the models with brain data , we employ a particular subset of the many network diagnostics that have been proposed as measures of network topology [31] , specifically chosen to highlight the regional variability and multiscale nature of network architecture . Many network diagnostics can be described as summary diagnostics , in which a property of the network organization is reduced to a single diagnostic number . Examples include average path length and average clustering coefficient . However , the comparison of summary diagnostics between real and model networks can be difficult to interpret [32] because they often hide the granularity at which biological interpretations can be made . To maximize the potential for a mechanistic understanding , we instead study diagnostics that provide distributions , visualized and analyzed by two-dimensional curves or scatter plots where the regional variability of network structure is readily apparent . The following four diagnostic relationships are obtained from a distribution of values over network nodes or topological scales: hierarchy [33] , degree assortativity [34] , topological Rentian scaling [35] , [36] , and the topological fractal dimension [37] . Each of these inherently relational properties has previously been investigated in the context of anatomical brain networks in humans [28] , [38] , [39] . In this paper , we use them to examine the differences between empirically derived anatomical brain networks and synthetic network models .
We utilize previously published diffusion spectrum imaging data [39] to examine the network structure of anatomical connectivity between cortical regions in the human brain . In this data , the direct pathways between N = 998 cortical regions of interest are estimated using deterministic white matter tractography in 5 healthy human participants [39] . This procedure results in an N×N weighted undirected adjacency matrix W representing the network , with elements Wij indicating the ( normalized ) number of streamlines connecting region i to region j ( see Figure 2 ) . The organization of white matter tracts can be examined at two distinct levels of detail: topological and weighted . Studies of the topological organization of brain anatomy focus on understanding the presence or absence of white matter tracts between regions [26]–[28] , while studies of the weighted organization focus on understanding the strength of white matter connectivity between those regions . In this paper , we explore the topological organization of white matter connectivity between cortical regions . In future work we plan to build additional constraints into our models that will enable a comparison of model and empirical weighted networks . To study topological organization , we construct the binary adjacency matrix A in which the element Aij is equal to 1 if the employed tractography algorithm identifies any tracts ( of any strength ) linking region i with region j ( i . e . , ) . In this data [39] , the adjacency matrix A is relatively sparse , resulting in a network density of , where is the total number of connections present . This estimate of brain network sparsity is consistent with estimates extracted from other similar data sets of comparable network size [8] , [40] . Given the potential variability in the topological organization of networks extracted from different individuals [8] , [41]–[44] , we report results for one individual in the main manuscript and describe the consistency of these results across subjects in the Supplementary Materials . We also briefly note that while extremely rich , this data set also has its limitations . In particular , the development of high resolution imaging methods and robust tractography algorithms to resolve crossing fibers are fast-evolving areas of research . Novel imaging techniques have for example recently identified the existence of 90-degree turns in white matter tracts [45] , a biological marker that we are not sensitive to in our data . We measure four network properties including degree assortativity , hierarchy , Rentian scaling , and topological fractal dimension as well as several summary diagnostics , as reported in Table 2 . All computational and basic statistical operations ( such as t-tests and correlations ) were implemented using MATLAB ( 2009b , The MathWorks Inc . , Natick , MA ) software . Graph diagnostics were estimated using a combination of in-house software , the Brain Connectivity Toolbox [60] , and the MATLAB Boost Graph Library ( http://www . stanford . edu/~dgleich/programs/ ) . To perform the recursive topological partitioning employed in the examination of topological Rentian scaling , we used the software tool hMETIS [61] . Several of the network models that we investigate include one or more tunable parameters affecting the details of the generated graphs . These include the Barabási-Albert , affinity , and hybrid distance growth models . To compare these network models to the data , we optimized parameter values to minimize the difference between the model network and the empirical brain network . Specifically , we used the Nelder-Mead simplex method , which is a derivative-free optimization method , that minimizes the value of a difference metric between the two networks . We chose to let be the sum of the absolute relative difference of nine of the network characteristics reported in Table 2 ( clustering coefficient C , path length P , diameter D , degree assortativity r , hierarchical parameter β , topological Rentian exponent pT , topological fractal dimension dB , modularity Q , and number of communities #com ) . Alternative choices for the difference metric could weight some network characteristics to a greater or lesser degree than others . However , because we do not a priori have a rubric by which to determine the biological relevance of a single network diagnostic in comparison to others , we chose not to utilize such a weighting scheme .
We begin by comparing the network organization of the brain's anatomical connectivity with that of 8 network models whose structure is not a priori constrained to accommodate a physical embedding of the nodes in cortical areas . ( In the next subsection , we will examine 5 embedded network models . ) The non-embedded network models include an Erdös-Rényi graph , a configuration model with the same degree distribution as the empirical network , a ring lattice graph , a modular small-world graph , a fractal hierarchical graph , a Gaussian drop-off graph , a Barabási-Albert graph , and an affinity graph ( see Figure 2 for associated example adjacency matrices for these graphs and Table 1 for abbreviations of model names ) . These models range from disordered to ordered ( e . g . , the Erdös-Rényi and regular lattice models ) with a range of mesoscale organization for intermediate cases ( e . g . , modular small-world and fractal hierarchical models ) which influence the network diagnostics , and ( dis ) similarities to corresponding measurements for the brain . The non-embedded models described in the previous section necessarily ignore a fundamental property of the brain: its embedding in physical space . Spatial constraints likely play an important role in determining the topological properties of brain graphs [22] , [26]–[29] . In this section , we explore the topological properties of spatially embedded graphs in which the probability of connecting any two nodes in the network depends on the Euclidean distance between them [79] . We explore the same topological diagnostics as we did in the previous section: degree distribution , assortativity , hierarchy , and diagnostics estimating the topological dimension of the network . As a whole , we find that spatially embedded models capture more topological features of the empirical networks than models that lack the physical embedding constraint . To clarify the distinction between embedded and non-embedded network models , it is necessary to highlight the differences between topological and physical notions of space . Many topological models ( such as the Barabási-Albert model ) are often described in ways that utilize notions of “local” connections . However , this concept of locality is present in a purely topological sense and not in a geographical sense . Topological models such as the Barabási-Albert model are not derived from spatial embeddings in or and therefore the nodes of these networks do not have spatial positions and the edges of these networks do not have physical lengths . The nonequivalence of topological and geographic structure is illustrated by the fact that a network topology ( e . g . , BA ) can either remain non-embedded or can be embedded into Euclidean space ( e . g . , ) in many different ways: in some embeddings , the topological distance between nodes could correlate with the physical distance between nodes , but in other embeddings one need not observe such a correlation . While the previous section described topological and non-embedded models , in this section we focus on networks that have been embedded into Euclidean space .
While the details of this set of analyses are of course important , we can also propose a set of integrated insights into the biological underpinnings of structural brain network organization based on the collective results extracted from these models . First , the fact that models hard-coded to display one topological property are unlikely to also display a second topological property suggests that the processes of neurodevelopment have not been selected by evolutionary drivers to optimize a single topological variable . Such a suggestion is intuitively plausible: What mechanisms exist to isolate and optimize single topological properties in the complicated cellular milieu of a developing organism ? Evidence from evolution and development instead suggest that the neuronal systems in living organisms are constrained by energy and metabolic concerns [83] . While energetic concerns may subsequently translate into constraints on topological network architectures [12] , [25] , [28] , topological features are unlikely to be the singular driving mechanism of evolution . Supposing that energetic concerns play a role in guiding network connectivity in large-scale brain structure , how might these concerns manifest themselves in the observed network organization of a single organism at a single point in time ? One possibility is that such constraints would impact on the physical length of connections since long connections are arguably more costly to both develop and maintain [12] , [25] , [28] . Consistent with this possibility , we observe that models that penalize physical length of connections ( embedded models ) tend to be more similar to the empirical data than models that hard-code specific topological properties ( non-embedded models ) . This gross result , robust to individual variation in different model parameters , supports the view that biological physics may be a more fundamental driver of structural brain architecture than network topology . However , we also note that simple distance models remain unable to capture all of the intricacies of the observed network architecture . While there is certainly room to construct more complicated physical models , it is also arguable that additional biophysical constraints are playing a secondary but influential role . A key feature of networked neuronal systems is their development over time , which displays complicated maturation-dependent trajectories [84]–[86] . It is therefore intuitively plausible that growth processes pose unique constraints on network development that cannot be captured by static physical distances alone . Indeed , we observe that 2 of the 3 models that display most similarity to the empirical network structure are growing models ( the affinity model and the hybrid distance growth model , which we define for the first time in this paper ) , suggesting that principles underlying the time evolution of network structures is critical . If true , this result uncovers a major gap in current network models of neuronal systems: namely , a sequence of models of increasing complexity that account for both physical constraints and growth processes on final ( adult ) network architecture . We speculate that such models , which obey principles of both physics and time , will be best able to capture observed empirical brain network structure . Model interpretations aside , it is important to emphasize that this work has a complementary purpose: to provide researchers with mathematical null models to inform statistical inference . The pragmatic uses of these models fall under two broad categories: ( i ) the use of a single model and ( ii ) the use of the full model battery . Single models can be used to address the question “How different are my two sets of networks in property y beyond that expected by their differences in property x . ” For example , one might have a group of networks from a clinical population and a group of networks from a control population . The two groups might differ in both their degree distribution and their clustering coefficient . However , one would like to test whether their difference in clustering coefficient is more than expected given their difference in degree distribution . That is , one would like to isolate the independent contribution of one network parameter to the phenotype of the disease . The statistical test one could then employ is to compare the clustering coefficient of the empirical networks in one group ( normalized by the clustering coefficient of the associated configuration models , which control for degree distribution ) to the clustering coefficient of the empirical networks in the other group ( again normalized by the clustering coefficient of the associated configuration models ) . Such a test directly determines whether the clustering coefficient is more different between the two groups than expected given the differences in their degree distributions . While we have used the clustering coefficient and degree distribution for simplicities sake in this argument , all other ( potentially more complicated ) pairs of properties can be examined similarly ( e . g . , hierarchical structure , preferential attachment , modular structure , wiring properties , etc . ) . In addition to single models , model batteries can be used to probe more general questions of group differences between sets of networks , for example from clinical and control populations . In some group comparisons , it is possible to observe marginally significant group differences in many network properties but to not observe any single network property that is affected drastically in isolation . In such cases , it is useful to report a comprehensive statistical test that encompasses these findings , rather than report a series of separate t-tests . In this context , model batteries can be extremely useful because they can provide response functions ( such as the summed relative difference from data , illustrated for a single individual in Figure 11A ) that indicate the differences between the data and the model battery . Different individuals can have different response functions ( as illustrated in Figure 11B ) , as can different groups . To directly compare these functions between groups , one can use a branch of statistics known as functional data analysis ( for a relevant textbook see [87] and for an application in network neuroimaging see [88] ) . Model batteries therefore complement network diagnostics in providing measurable statistics that can be used to identify subtle differences in network architecture between groups . In the following sections we discuss the details of each model more fully and relate our results to prior work . We conclude with a description of model interpretations , future directions , and methodological limitations . We probe non-embedded models with differing amounts and types of structure . While the Erdös-Rényi model provides an important benchmark with a random topology , it bears little resemblance to the brain network . Although a homogeneous random distribution of links has been suggested to characterize the small-scale structure of neuron-to-neuron connections [89] , [90] , the large-scale structure of human and animal brains instead displays heterogeneous connectivity [67] . Perhaps one of the simplest measures of this heterogeneity is found in the degree distribution , which displays a predominance of low degree nodes and a long tail of high degree nodes . In comparing the degree distribution of the brain to that obtained from a BA model , it is clear that this tail , however , is not well-fit by a power-law , a finding consistent with previous reports in brain anatomy [21] , [38] and function [15] , [91] . However , by matching the empirical data , for example using a configuration model with the same degree distribution , we note that we do not automatically uncover higher order structures like assortativity , suggesting that the degree distribution provides only limited insight into the forces constraining brain network development . Several decades ago , neuroanatomists observed that the pattern of connections in several animal brains displayed a combination of both densely clustered areas and long range projects between distant areas [92]–[95] . The regular lattice and Gaussian drop-off models are able to capture these densely connected structures but fail to capture the extent of long-range connectivity observed in the brain . The small-world modular and fractal hierarchical models contain both properties: dense local connectivity and long-range interactions . The fractal hierarchical model has the added benefit of containing nested structures , which have been implicated in the heterogeneity of neuronal ensemble activity [11] and in the separation and integration of information processing across multiple frequency bands [96] . Moreover , hierarchical modular structure has been identified in organization of white matter streamlines in human diffusion weighted imaging data [8] , [72] , [74] and implicated in neurobiological phenomena [11] , [75] , [76] . None of the non-embedded models discussed earlier in this section simultaneously provide a heterogeneous degree distribution , degree assortativity , hierarchical topology , and realistic topological dimensions . Such a “No Free Lunch” rule is perhaps unsurprising , in that a network that is developed to directly obtain one property typically fails to also display a second property . This result suggests that the topological properties that we explore here are in some sense independent from one another . It is , however , important to clarify that the interpretation of our findings in light of the observed correlations between network diagnostic values themselves , estimated over different networks or models ( see previous literature , e . g . , [97] , [98] , and results for the current data presented in Figure S4 in the Supplementary Materials ) , that suggest the need for methods to identify distinguishing properties among networks [56] , [99] . The two sets of observations can be brought together by realizing that while classes of networks ( e . g . , brain networks ) might display correlated network diagnostics values , these relationships need not be expected theoretically from any randomly chosen set of networks . Indeed , networks can be segregated into families based on the profile of interdependence between network diagnostic values [100] . Finally , in our affinity model , we hard-code both degree assortativity and a continuous hierarchical topology , rather than the discrete hierarchy employed in nested models like the fractal hierarchical model examined here . Interestingly , however , and in contrast to the other non-embedded models , we simultaneously obtain a heterogeneous degree distribution , and similar estimates of the topological dimension . This model fits multiple properties of brain networks that were not explicitly included in the construction of the network model , but are nevertheless a consequence of a three-parameter fit in the specific affinity model selected . The affinity model therefore serves as a promising candidate as both a generative model and statistical null model of brain organization . In an effort to include additional biological constraints , we also explore several models that employ information regarding either the physical placement of network nodes or that place constraints on the Euclidean lengths of network edges . In general , this set of networks outperforms most of the non-embedded network models that we studied , demonstrating that the brain is highly spatially organized and supporting the notion that physical constraints might play important roles in brain network development and structure [8] , [25]–[29] , [90] , [101] , [102] . It is important to preface the discussion of our results by mentioning the fact that the properties of empirically derived brain networks display a heterogeneity that could at least in part stem from the peculiar physical properties of the organ . Brains are symmetric objects , with the two hemispheres being connected with one another via tracts in the corpus callosum and via subcortical structures . This separation allows for a very different topology within a hemisphere than between hemispheres . Moreover , cortical areas ( gray matter ) form a shell around the outer edges of the brain while their connections ( white matter ) compose the inner volume . Finally , brain areas are inherently heterogeneous in physical volume , making their distances from one another far from homogeneous . While the morphology of the brain constrains its potential topological properties , evidence also suggests that the lengths of tracts connecting brain areas follow a heavy tailed distribution , with short tracts being relatively common and long tracts being relatively rare [26] , [27] . These findings are in concert with the idea that energy efficiency—to develop , maintain , and use neuronal wiring—remains a critical factor in brain evolution and development [29] , [103] . In this study , we begin with a random geometric model , whose nodes are placed uniformly at random in a volume but whose edges selectively link nodes that are nearby in physical space . In light of the simplicity of this model , it is somewhat surprising that we obtain such good agreement with the empirical degree distribution , the presence of assortativity , and the presence of a hierarchical topology . In the minimally wired graph we employ a similar connection rule but also fix node placement to be identical to that in the empirical brain network , following previous studies [28] . However , neither of these two models are able to capture the extent of long-distance connections observed in the empirical data . By employing the distance drop-off model , we can fix a connection probability that varies with distance , rather than simply a connection threshold . This connection probability , however , is not enough to provide a realistically broad degree distribution . Our distance drop-off growth model combines the strengths of each of these models by laying down a set of seed edges uniformly at random in a volume and then iteratively adding edges between pairs of nodes according to a probability that falls off with inter-node distance . The resulting degree distribution and assortativity properties are the best match to the empirical data of the models that we studied . A hybrid between the minimally wired model and the distance drop-off growth model does not perform significantly better in matching these properties and shows a hierarchical structure that is more pronounced than the data . Importantly , the embedded network models examined here are purposely simplistic . While arbitrarily more complex models could be constructed , our goal was to isolate individual drivers of topology and probe their relationship to observed network diagnostics . Other studies of interest in relation to these findings include those that explore the effects of geometric folding [90] , radial surface architectures [102] , and the effects of wiring minimization on functional networks [25] . While the construction of network models is genuinely critical in providing null tests for statistical inference of brain structure from data , this avenue of research also has the potential to provide key insights into the neurobiological mechanisms of brain development and function if performed with appropriate caution . In light of this second use , we note that several of the network models discussed in this paper employ rules that are reminiscent of—or even directly inspired by—known biological phenomena . For example , physical models that place constraints on the length of connections in Euclidean space are consistent with the known distribution of connection lengths in the brain and the modern understanding of metabolic constraints on the development , maintenance , and use of long wires [26]–[29] , [101] , [103] . However , even topological constraints that link nodes that have similar sets of neighbors can be interpreted as favoring links between neurons or regions that share similar excitatory input [25] . As an example , our affinity model hard-codes two inter-node relationships . First , nodes with a similar degree are more likely to be connected to one another by an edge , leading to degree assortativity throughout the network . This behavior can be thought of as a mathematical representation of the intuitive principle of spatial homophily: large neurons with expansive projections ( e . g . , pyramidal or basket cells ) are more likely to connect to one another because they densely innervate tissue over large distances . Network assortativity can also stem from the temporal homophily that occurs during development: neurons that migrate over longer distances during development are more likely to come into contact with—and therefore generate a synapse with—one another than neurons that migrate over shorter distances . The second topological relationship hard-coded into the affinity model is the prevalence of clustering in local neighborhoods , a property consistent with physical constraints on network development . As neurons develop , it is intuitively more likely for them to create synapses with neighboring neurons than non-neighboring neurons , thereby closing topological loops in close geographic proximity . While we have only provided a few examples here , links between topological rules and biological phenomena provide potentially critical neurophysiological context for the development and assessment of synthetic network models . The perspective that we have taken in choosing synthetic network models is one of parsimonious pragmatism . We seek to identify models with simplistic construction rules or growth mechanisms to isolate topological ( non-embedded ) and physical ( embedded ) drivers of network topology . One alternative perspective would be to begin with a certain graph topology ( for example , an Erdős-Rényi graph ) , and iteratively rewire edges to maximize or minimize a network diagnostic or set of network diagnostics [25] . However , this approach requires prior hypotheses about which network diagnostics are most relevant for brain network development , a choice that is complicated by the observed correlations between such diagnostics [97] . Another approach is to employ exponential random graph models [16] , [19] , [104] , which provide a means to generate ensembles of networks with a given set of network properties but do not provide a means to isolate mechanistic drivers of those network properties . A third approach is to construct a mechanistic model based on particle-particle collisions , which might serve as a physical analogy to the biological phenomena of neuronal migration through chemical gradients [105] , [106] . In each of these cases , a perennial question remains: at what spatial scale should we construct these models to gain the most insight into the relevant biology ? Important future directions could include the development of multiscale growth models , enabling us to bridge the scales between neuronal mechanisms and large-scale structure . There remain important limitations to our work . In particular , we have focused on understanding the ( binary ) topology of brain network architecture rather than its weighted connection strengths . Our choice was informed by three factors: 1 ) An understanding of the relationship between synthetic network models and brain network topology could be useful for informing a similar investigation into network geometry , 2 ) In these particular networks , node degree ( binary ) and node strength ( weighted by the number of streamlines ) are strongly correlated ( Pearson's correlation coefficient , ) and therefore topology serves as a proxy for weighted connectivity , and 3 ) The choice of how to weight the edges in an anatomical network derived from diffusion imaging is an open one [107] , and therefore investigations independent of these choices are particularly useful . Network models constitute necessarily simplified representations of often very complex systems . The 13 synthetic network models we study in this work could be extended to include additional physical features of the human brain . For example , a key constraint on brain morphology and connectivity lies in the organ's bilateral symmetry . This symmetry in brain structure is evident in the distribution of anatomical connectivity in the brain networks examined in this study: pairs of homologous regions are more than 3 times more likely to be connected to one another than pairs of non-homologous regions . As described in [39] , each of the 998 regions used in the parcellation is affiliated with one of 66 anatomical parcels defined based on surface reconstruction performed in Freesurfer . We calculated the average density of connections between all of the regions in one anatomical parcel and all of the regions in another anatomical parcel . In this way , we obtain a pairwise density of connectivity between all 66 anatomical parcels . The average density of connections between homologous regions is 15 . 22% and the average density of connections between non-homologous regions is 4 . 05% . The topological ramifications of this symmetry are not well understood . Moreover , in simple network models , emphasis is placed on characterizing the patterns of network edges while the characteristics of individual nodes ( apart from their connectivity ) are examined to a lesser degree [108] . The development of more complicated models that account for feature vectors of brain region properties could provide additional insights into neurophysiological phenomena . Indeed , quantifying the relationship between a brain region's connectivity and its functional or anatomical properties is a critical goal of network neuroscience . Initial forays into this area have demonstrated that topological properties of a brain region ( node degree ) can be linked to neurophysiological properties ( prevalence of amyloid-beta deposition ) [109] , suggesting the utility of network approaches in providing mechanistic hypotheses regarding disease attributes . In this paper , we have examined the mechanistic drivers of network topologies by employing and developing a range of synthetic network models governed by both topological ( non-embedded ) and physical ( embedded ) rules and comparing them to empirically derived brain networks . These tools may prove useful in the statistical inference of anatomical brain network structure from neuroimaging data . Future efforts can further build on these findings to identify neurobiologically relevant mechanisms for healthy brain architecture and its alteration in disease states .
|
White matter tracts crisscrossing the human cortex are linked in a complex pattern that constrains human thought and behavior . Why the human brain displays the complex pattern that it does is a fascinating open question . Progress in uncovering generative mechanisms for this architecture requires greater knowledge about mechanistic drivers of anatomical networks . Here we contrast network properties derived from images of the human brain with 13 synthetic network models investigated in a progressive , branching sequence , chosen to probe the roles of physical embedding and temporal growth . We characterize both the empirical and synthetic networks using network diagnostics presented here in statistical form , as scatter plots and distributions , to reveal the full range of variability of each measure . We find that synthetic models that constrain the network nodes to be physically embedded in anatomical brain regions tend to produce distributions that are most similar to the corresponding measurements for the brain . We also find that network models hardcoded to display one network property do not in general simultaneously display a second , suggesting that multiple neurobiological mechanisms drive human brain network development . The network models that we develop and employ enable statistical inference of brain network structure from neuroimaging data .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"statistical",
"mechanics",
"nervous",
"system",
"applied",
"mathematics",
"neuroscience",
"systems",
"science",
"mathematics",
"statistics",
"(mathematics)",
"computational",
"neuroscience",
"neuroimaging",
"complex",
"systems",
"computer",
"and",
"information",
"sciences",
"connectomics",
"systems",
"biology",
"physics",
"neuroanatomy",
"anatomy",
"interdisciplinary",
"physics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"statistical",
"methods"
] |
2014
|
Resolving Structural Variability in Network Models and the Brain
|
Glucose is the preferred carbon and energy source in prokaryotes , unicellular eukaryotes , and metazoans . However , excess of glucose has been associated with several diseases , including diabetes and the less understood process of aging . On the contrary , limiting glucose ( i . e . , calorie restriction ) slows aging and age-related diseases in most species . Understanding the mechanism by which glucose limits life span is therefore important for any attempt to control aging and age-related diseases . Here , we use the yeast Schizosaccharomyces pombe as a model to study the regulation of chronological life span by glucose . Growth of S . pombe at a reduced concentration of glucose increased life span and oxidative stress resistance as reported before for many other organisms . Surprisingly , loss of the Git3 glucose receptor , a G protein-coupled receptor , also increased life span in conditions where glucose consumption was not affected . These results suggest a role for glucose-signaling pathways in life span regulation . In agreement , constitutive activation of the Gα subunit acting downstream of Git3 accelerated aging in S . pombe and inhibited the effects of calorie restriction . A similar pro-aging effect of glucose was documented in mutants of hexokinase , which cannot metabolize glucose and , therefore , are exposed to constitutive glucose signaling . The pro-aging effect of glucose signaling on life span correlated with an increase in reactive oxygen species and a decrease in oxidative stress resistance and respiration rate . Likewise , the anti-aging effect of both calorie restriction and the Δgit3 mutation was accompanied by increased respiration and lower reactive oxygen species production . Altogether , our data suggest an important role for glucose signaling through the Git3/PKA pathway to regulate S . pombe life span .
Glucose is the major carbon source entering the metabolic pathways . Glucose ultimately generates ATP to supply the energy necessary for the cell biosynthetic and functional demands . Substantial evidences support the idea that excess glucose acts as a pro-aging and pathogenic factor [1] , [2] . Consistently , lowering glucose intake in a calorie restriction diet increases life span in many species , from yeasts to mammals [3] , [4] . Research carried out in Saccharomyces cerevisiae has been fruitful to unravel the role of nutrient sensing in longevity . Mutations blocking the action of genes controlling nutrient- signaling pathways increase replicative life span ( RLS ) , defined as the number of times a mother yeast cell produces a daughter cell [5]–[10] . For instance , genetic deletion of PKA signaling via Gpr1 or Gpa2 genes resulted in the extension of RLS [11] . Likewise , nutrient-signaling shortens chronological life span ( CLS ) , the time a yeast population remains viable in stationary phase [12]–[15] . In other words , nutrient-signaling pathways have a pro-aging effect in budding yeast . So far , mutations found to increase life span in S . cerevisiae map to genes that respond to multiple nutrients , such as the PKA , Sch9 and Tor pathways [16] . Glucose is the major source of calories for yeast . Experimentally , calorie restriction ( CR ) is achieved by reducing the concentration of glucose in S . cerevisiae cultures . Under these conditions , yeast cells exhibit an increase in both their replicative life span , and their CLS . It is therefore possible that nutrients , and more particularly the glucose-signaling pathway , are major regulators of the effects of calorie restriction on aging . In yeast , the connection between nutrient sensing and mitochondrial activity has been depicted in different contexts . This regulation of mitochondria allows yeast to adapt its energy metabolism to the available nutrients , and is crucial for the control of longevity [16] . Several genetic studies demonstrate that forcing S . cerevisiae to use respiration instead of fermentation induces a gain in both chronological and replicative life span [17]–[20] . To summarize , the activity of nutrient-signaling pathways seem to promote aging by inhibiting both stress resistance and respiration . However , the predominance and the interdependence of each of these two functions , metabolic changes and signaling in the control of longevity are still nebulous . Our laboratory introduced Schizosaccharomyces pombe as a model organism for the study of chronological aging [21] . The use of this particular yeast is justified by the differences existing with budding yeast in traits that can potentially affect longevity . Both have been referred as Crabtree-positive yeast because of their capability to repress mitochondrial respiration in favour of glycolysis when glucose is abundantly available [22] . Nevertheless in fission yeast , the Crabtree effect is less pronounced than in S . cerevisiae since the inhibition of oxygen consumption by glucose is smaller; in other words S . pombe maintains a higher respiration rate in the presence of glucose [22] . Consistently , it is hard to isolate respiratory-deficient cells ( petite ) in S . pombe [23] , [24] , while these mutants occur spontaneously in S . cerevisiae . Furthermore , S . pombe differs from S . cerevisiae because of its lack of glyoxylate cycle that makes it inefficient in ethanol consumption as carbon source [25] , [26] . Finally , fission yeast is also distinguishable in its mitochondrial inheritance which is mediated by microtubules like in higher eukaryotes [27] . In the present study , we wished to determine whether glucose metabolism or extracellular glucose signaling is responsible for the regulation of life span . We found that environmental glucose decreases CLS in S . pombe in a dose-dependent manner , and this effect is mimicked in cells lacking the glucose receptor Git3p , a G protein-coupled receptor ( GPCR ) which signals the presence of glucose in the medium through a cAMP/PKA pathway [28] , [29] . Consistently , the constitutive activation of the Gα subunit of the G protein-coupled to the glucose receptor significantly decreases CLS . Deletion in the Git3/PKA signaling is characterized by higher oxidative stress defense , respiration and mitochondrial membrane potential; the same features observed in CR . Interestingly , CR has no effect either on stress defense or longevity in the strain constitutively activated for Git3/PKA ( by mutational activation of the Gα subunit ) , although it still enhances respiration . Knockout of S . pombe hexokinase genes ( hxk1 and hxk2 ) , which are required to channel extracellular glucose into glycolysis , does not extend CLS in S . pombe . On the contrary , these mutant yeast strains accumulated glucose in the medium , exhibited increased glucose signaling and accelerated aging . Reduction of extracellular glucose or mutation of the glucose receptor Git3p rescued their aging phenotype . Altogether , our data suggest that glucose signaling constitutes the main pathway in the pro-aging effect of glucose in fission yeast .
To study the effects of glucose concentration on CLS , wild-type S . pombe cells were cultured in rich medium with different concentrations of glucose . Survival was assessed by counting colony forming units ( CFU ) as a function of time , after cells entered stationary phase [21] . Decreasing the concentration of glucose from 2% to 0 . 05% resulted in a dose-dependent extension of chronological life span ( Figure 1A ) . Cultures with higher glucose concentration exhibited a premature appearance of aged-cell phenotype upon entering the stationary phase . This phenotype is characterized by a shrunken shape and oversized vacuoles ( Figure 1B ) . DNA content analysis by flow cytometry revealed that cells cultured in glucose 2% and 0 . 2% displayed a typical G2 cell-cycle arrest in stationary phase ( data not shown ) . Moreover , the cells had similar doubling times at different glucose concentrations during the exponential growth phase of the culture ( Figure S1 ) . Aging in yeast results in part from cellular damage due to the accumulation of reactive oxygen species ( ROS ) [30] . In agreement , cells cultured in higher glucose concentrations accumulated more ROS than cells grown at lower glucose concentrations , as shown by staining with dihydrorhodamine 123 and dihydroethidium ( Figure 1C and 1D and Figure S2 ) . On the other hand , culturing cells in SMC medium lacking glucose and containing glycerol as carbon source increased chronological longevity up to tenfold longer than in 2% glucose ( Figure 1E ) . Glycerol as sole carbon source forces the cell metabolism toward mitochondrial respiration , as evidenced by the diminished growth rate and the rise in oxygen consumption [31] . Altogether , these results confirm that the relationship between nutrition and longevity in S . pombe is similar to that observed in other model organisms . A number of mechanisms could account for the pro-aging effects of glucose in S . pombe . For instance , the effect of glucose on aging could be due to extracellular glucose sensing ( signaling pathway ) or through intracellular glucose effects including glucose metabolism and cytoplasmic glucose sensing . To distinguish between them , we studied a strain deleted for the git3+ cytoplasmic membrane glucose receptor gene . S . pombe cells lacking this receptor ( Δgit3 ) exhibited extension of their CLS ( Figure 2A ) , suggesting that the pro-aging effects of glucose depends , at least in part , on the activation of a signaling pathway initiated by this receptor . To further confirm this idea , we used a constitutively active Gα subunit ( Gpa2R176Hp ) that acts downstream of Git3p in the glucose-signaling pathway . Gpa2R176Hp constitutively activates the PKA kinase independently of the presence of glucose by promoting the synthesis of a high level of cAMP [28] , [32] . As expected , cells expressing this activated Gα protein displayed a significantly reduced CLS ( Figure 2A ) . To confirm that Δgit3 and gpa2R176H cells have decreased and increased glucose signaling , respectively , we took advantage of the fact that in S . pombe , glucose represses the transcription of the fructose-1 , 6-bisphosphatase fbp1+ gene via PKA activation [33] , [34] . We used an fbp1-driven lacZ reporter integrated in the S . pombe genome to measure fbp1 transcription [35] , 36 . Thus , the β-galactosidase activity inversely reflects the level of PKA activation in this glucose-sensing pathway . As expected , at late logarithmic phase , deletion of git3+ increased expression of this reporter , while the gpa2R176H mutation reduced its expression ( Figure 2B ) . Consistently , culturing WT cells in low glucose conditions also increased fbp1-lacZ expression ( Figure 2B ) . Since chronological aging in yeast is linked to the accumulation of ROS [21] , [30] , [37] , we next measured the levels of ROS in Δgit3 and gpa2R176H cells . As expected , deletion of the glucose receptor reduced ROS levels , while the constitutive activation of the glucose-signaling pathway increased ROS levels ( Figure 2C ) . Although these results suggest that glucose signaling regulates aging independently of glucose utilization , it is possible that loss of the glucose-signaling pathway reduces glucose metabolism in this mutant . Indeed , the PKA pathway is known to control glucose intake via the regulation of hexose transporters responsible for glucose import in S . cerevisiae [38] , [39] . We thus measured glucose consumption and found that mutations affecting the glucose-signaling pathway did not change the rate of glucose consumption ( Figure S3 ) . In conclusion , these results suggest that the glucose-signaling pathway controls chronological aging independently of glucose intake and utilization . Experimentally , the intervention referred as calorie restriction ( CR ) is achieved by reducing the calorie intake of an organism and represents the most effective way to increase life span [3] . This phenomenon has been verified in almost all species studied , from yeast to mammals [7] , [40] including non-human primates [41] . CR improves general health and delays the inception of many late-onset diseases in a variety of organisms [42] . In S . cerevisiae , calorie restriction is implemented by culturing the cells on low glucose concentrations [5] , [15] , [43] . Above , we showed that culturing S . pombe in low glucose decreases glucose signaling , and demonstrated that mutations affecting this signaling pathway increase the life span of S . pombe when cultured on high glucose concentration . Increased respiration correlates with longevity in yeast [17] , [20] , and mammals [4] . In yeast , low glucose availability leads to a switch of the pyruvate metabolism from fermentation toward mitochondrial tricarboxylic acid cycle and respiration [26] , [44] . To determine if mutations in the glucose-signaling pathway affect respiration in S . pombe , we measured the oxygen consumption of long-lived Δgit3 and short-lived gpa2R176H cells . In high glucose , we observed that Δgit3 cells display a higher level of oxygen consumption as compared to that of WT ( Figure 3A ) . The effect of respiration on the mitochondrial membrane potential ( Δψm ) was determined using the DiOC6 dye and showed that the Δψm in stationary phase cells was higher in Δgit3 compared to WT cells ( Figure S4 ) . Interestingly , WT cells cultured in 0 . 2% glucose exhibited a higher level of oxygen consumption than the Δgit3 mutant in 2% glucose ( Figure 3A ) , and an increased mitochondrial membrane potential ( Δψm ) in early exponential phase ( Figure S4 ) . This could explain why CR is slightly more efficient than the git3+ deletion in extending CLS ( Figure 2A ) . Higher respiration was concomitant with a better growth on respiration medium ( glycerol 3% ) of both , WT cells previously grown on CR conditions and Δgit3 cells grown on either normal or CR conditions ( Figure 3B ) . In addition to glucose repression , the participation of the PKA/cAMP-mediated signaling pathway in mitochondrial functions has been suggested in budding yeast [45] , [46] . Our data and the observation that pka1+ deletion increased respiration as well ( not shown ) support the involvement of Git3/PKA in the regulation of mitochondrial functions in fission yeast . To investigate if Git3/PKA is the only pathway regulating the metabolic switch toward mitochondrial respiration during CR , we subjected Δgit3 cells to CR and measured respiration . We observed an increase in respiration when Δgit3 mutation was combined to CR compared to Δgit3 cells in 2% glucose ( Figure 3A and C ) . Hence , CR can increase respiration by mechanisms independent of the glucose receptor Git3 . On the other hand , we observed that the activated Gpa2R176Hp prevents the full activation of respiration induced by CR ( Figure 3D ) . These results are supported by the observation that gpa2R176H cells did not grow on respiration medium ( glycerol ) ( Figure 3B ) . Moreover , as oxygen consumption of WT and Δgit3 cells was 30% higher than gpa2R176H cells in CR , we observed that the Δψm of Δgit3 and WT cells in stationary phase remained higher than gpa2R176H cells ( Figure S4 ) . Altogether these data suggest that CR and reduced glucose signaling are not equivalent , and that Git3/PKA is involved in the control of respiration . Guarente and colleagues proposed that in yeast , CR increases life span by increasing respiration but not oxidative stress resistance [17] . However , another study from Kaeberlein and collaborators contradicted these data . They showed that reducing glucose levels increased replicative life span in respiratory-deficient yeast [47] . We showed above that both CR and the Δgit3 mutation increase respiration , while the gpa2R176H mutation decreased the effect of CR on respiration . Combining CR and the Δgit3 mutation did not increase respiration over the values with CR alone . However , the survival of Δgit3 cells was higher on CR than that of WT cells ( Figure 4A ) . On the other hand , CR did not increase the respiration rate of the strain expressing activated Gpa2R176Hp as this intervention did in WT cells ( Figure 3D ) , and this defect could partially explain its short life span ( Figure 4B ) . To investigate further whether the additive effect of CR and loss of Git3p signaling involves respiration , cells were cultured in 20 µM of antimycine A , an inhibitor of complex III of the mitochondrial electron transport chain , that creates a leakage of electrons [48] and increases ROS production . Glucose restriction and git3+ deletion together increased longevity in this high-ROS context ( Figure S5 ) . This suggests that low glucose signaling cooperates with other effects of CR acting downstream of ROS production , perhaps stimulating ROS defense mechanisms . Together , the data suggests that CR and reduced glucose signaling are not equivalent and these manipulations can actually cooperate to increase life span by a mechanism different than an increase in respiration . To investigate whether resistance to oxidative stress could explain the longevity effects of CR and git3+ deletion , we next study the effects of several pro-oxidants molecules on WT and mutants S . pombe cells grown at high or low glucose concentrations . First , CR and , to a lesser extent , loss of the Git3p GPCR increased hydrogen peroxide and menadione resistance ( Figure 4C ) . Moreover , CR strengthened the already high stress resistance of Δgit3 cells ( Figure 4C ) . On the other hand , the resistance to both hydrogen peroxide and menadione treatment in gpa2R176H cells was significantly lower than in WT ( Figure 4D ) . This stress sensitivity could also explain the very short CLS of this mutant in both high and low glucose . We also measured the levels of cytosolic Cu/Zn-superoxide dismutase ( SOD1 ) and mitochondrial Mn-SOD ( SOD2 ) by quantitative PCR ( Figure 4E ) . The importance of these two enzymes for long-term survival was demonstrated in budding yeast cultured in high glucose concentration [49] . No significant differences of expression were seen , neither in SOD2 ( Figure 4E ) nor in glutathione peroxidase ( Gpx1 , data not shown ) for all the mutants and growth conditions tested . Interestingly , WT cells on CR showed no increased expression of SOD1 or SOD2 despite their very high oxidative stress resistance ( Figure 2C ) . On the other hand , the deletion of glucose receptor increased significantly SOD1 expression . This correlates with the gain of oxidative stress resistance of this strain ( Figure 4C ) . An unexpected three to six time rise of SOD1 transcript was observed in the gpa2R176H mutant , even if this strain displayed a very weak oxidative stress resistance . This could be the consequence of a feedback mechanism attributable to the very high production of ROS in this strain ( Figure 2C ) . Although further studies on the mechanisms of stress resistance will be necessary , our data clearly shows that glucose signaling regulates SOD1 expression in S . pombe . Since SOD1 is not regulated by CR , it may be part of the mechanism by which the git3+ deletion cooperates with CR to increase the resistance to oxidative stress and life span . In yeast , hexokinase 2 is responsible for channeling glucose into metabolic pathways by catalyzing phosphorylation of this sugar . It also has a function in glucose signaling in S . cerevisiae by promoting the down-regulation of glucose-repressed genes [50] , [51] . Mutants of hexokinase do not influence CLS but increase replicative life span in S . cerevisiae [17] , [43] . In fission yeast , the glucose phosphorylation activity is provided by two hexokinases ( Hxk1p and Hxk2p ) , but the main enzymatic activity is due to hexokinase 2 [50] . Loss of Hxk1p has no significant phenotype ( data not shown ) and loss of Hxk2p dramatically decreases the growth rate in glucose [50] . The double knockout of both hxk1+ and hxk2+ is not viable on glucose [50] . To determine if hexokinase affects CLS in S . pombe , we first measured the life span of an S . pombe Δhxk2 deletion strain . Unlike in S . cerevisiae , we observed a significant decrease in CLS in this strain ( Figure 5A ) . We have shown above that glucose signaling mediates pro-aging effects in S . pombe . Therefore , we reasoned that defective glucose utilization in the Δhxk2 strain could result in an accumulation of intracellular glucose followed by the inhibition of glucose import . Moreover in S . cerevisiae , glucose can be re-exported in the extracellular medium by the hxt hexose transporter [52] . In turn , the high extracellular glucose concentration would lead to increase the duration of glucose signaling . To test this hypothesis , we first measured the glucose concentration in the medium during the growth of both wild-type and Δhxk2 cells . We found that glucose levels remained high in the Δhxk2 culture as compared to WT , and that this strain has a very slow growth rate ( Figure 5B ) . Congruently with this observation , at early stationary phase Δhxk2 cells exhibited increased glucose signaling in comparison with control cells , as represented by the drop of fbp1-lacZ reporter expression ( Figure 5C ) . In budding yeast , hexokinase activity has been involved in glucose-signaling pathways during exponential phase [53] . Our results do not contradict , but support those observations since the Δhxk2 mutant has a defect in glucose signaling in exponential phase when compared to wild type , as indicated by elevated fbp1-lacZ expression ( data not shown ) . However , the Δhxk2 mutant reaches stationary phase with glucose in the medium and its short life span was completely rescued when cultured in 0 . 2% glucose . As expected , culturing Δhxk2 cells in low glucose resulted in a two-fold increase in β-galactosidase activity indicating an increase in fbp1-lacZ reporter expression . This shows a reduction in signaling through the Git3/PKA pathway ( Figure 5A and 5C ) . Taken together , the results are consistent with the model that an increase in glucose signaling via the Git3/PKA pathway accelerates aging in Δhxk2 mutants . The strain Δhxk2 still has the hexokinase 1 activity permitting glucose metabolism ( Figure 5B ) [50] . To confirm the importance of the pro-aging effect of glucose signaling isolated from the effect of glucose utilization as energy source , we constructed a double knockout of both hexokinases in fission yeast ( hxk1+ and hxk2+ ) . These two mutations should prevent glucose from entering glycolysis and the pentose phosphate pathway . However , so far attempts to obtain this double mutant has been unsuccessful [50] . It was concluded that Δhxk1 Δhxk2 strain is not viable on glucose . To circumvent this problem , we complemented Δhxk2 with a plasmid expressing Hxk2p ( pREP41_hxk2+ ) and crossed it with a Δhxk1 strain . After sporulation of the diploid , we selected for offspring containing both Δhxk1 and Δhxk2 deletions and the plasmid pREP41_ hxk2+ . Then we allowed the strain to lose the hxk2+ plasmid in a medium containing only glycerol as carbon source and picked clones without plasmid . Because we obtained viable double mutants , we concluded that hexokinase activity and possibly glucose phosphorylation was required for sporulation but not for survival in S . pombe . Using the same approach , the triple knockout , Δhxk1 Δhxk2 Δgit3 was created . These mutants Δhxk1 Δhxk2 and Δhxk1 Δhxk2 Δgit3 could not grow when switched on plates with only glucose as carbon source . After at least ten days of incubation however , in some plates we observed for both strains clones that grew on glucose at a frequency between 10−6 to 10−7 ( data not shown ) . The appearance of such clones was attributed to genetic reversion due to the nature of hexokinase 2 knockout that was created by insertion of a marker rather than complete suppression of the open reading frame [50] . Mutants Δhxk1 Δhxk2 and Δhxk1 Δhxk2 Δgit3 were grown in glycerol as a carbon source for around two divisions with a doubling time of around ten hours . At this point , they were spotted on plates containing glycerol plus glucose ( Figure 6A ) . The double mutant Δhxk1 Δhxk2 did not grow on glycerol with 2% glucose , but it did on glycerol with 0 . 2% glucose . This result is consistent with the idea that the absence of hexokinase activity leads to a sustained and toxic glucose signaling . In agreement , the impaired growth of the double hexokinase mutant on glycerol plus glucose 2% was completely restored by a deletion in the glucose receptor git3+ ( Figure 6A ) . To assess whether the increase in glucose signaling of the double hexokinase knockout decreases the viability in stationary phase ( chronological aging ) , we added 2% glucose to liquid cultures at late exponential phase ( OD595 5–6 ) . Then , viability was evaluated as a function of time by counting the number of living cells per mL ( Figure 6B ) . After glucose addition , cultures with and without glucose needed two-day incubation to reach saturation corresponding to OD595 13 to 16 . The Δhxk1 Δhxk2 double deletion mutant exposed to 2% glucose displayed striking loss of viability 24 hours after glucose addition in comparison to cultures with no added glucose ( Figure 6B ) . This loss of viability was prevented by CR ( 0 . 2% glucose ) or by deletion of git3+ ( Figure 6B ) . To further characterize the loss of viability induced by glucose in double hexokinase knockout strains , we stained yeast cells with Phloxin B , a dye accumulated by dead cells . We found a high proportion of stained cells ( 30% ) at 18 hours after glucose addition in comparison to 5% in control cells ( Figure 6C ) . Notably , Phloxin B stained cells were longer and displayed oversized vacuoles , a typical phenotype of aging in yeast ( Figure 6C ) . ROS production was evaluated 36 hours after glucose addition by flow cytometry with DHE staining . A considerable number of DHE stained Δhxk1 Δhxk2 cells was observed in the culture with 2% glucose ( Figure 6C ) . Again , Phloxin B staining and the increase in ROS were prevented by CR ( glucose 0 . 2% ) or deletion of git3+ ( Figures 6B and C ) . Our results show that glucose signaling via the Git3p GPCR is required for the pro-aging effects of glucose in S . pombe and is sufficient to mediate detrimental effects even in the absence of glucose consumption .
Excessive glucose signaling has been associated with humans diseases such as diabetes , as well as with the less understood process of aging [54] . Several mechanisms have been proposed for the harmful effects of glucose . Glucose can be directly toxic to cell components because it can promote non-enzymatic glycosylation and the accumulation of advanced glycation end products ( AGE ) which impair cellular functions [55] , [56] . Excess glucose metabolism can also be deleterious because glucose oxidation increases the source of electrons to the mitochondrial respiratory chain in the form of NADH . In cells with a very active glucose metabolism , excess electrons can promote the generation of deleterious ROS if there is no matching increase in the efficiency of electron transport [54] , [57] . Glucose and/or nutrient-signaling pathways also control life span in various species including yeast [15] . The data raise the question about the relative contribution of signaling and metabolism to the regulation of life span [58] . Here we examined this question in S . pombe using mutants of the Git3/PKA glucose-signaling pathway . In this pathway , PKA kinase is activated by glucose signaling through the Git3p G protein-coupled receptor ( GPCR ) , which results in the Gα subunit ( Gpa2p ) -mediated activation of adenylate cyclase [29] as represented in Figure 7 . This , in turn , produces a linear increase in cAMP levels . The cAMP is bound by the Cgs1 regulatory subunit of Pka1 kinase , activating PKA . The consequence is a re-localization of PKA to the nucleus followed by the inhibition of the Rst2 transcription factor , an increase in stress sensitivity and a decrease in cell survival [21] , [33] , [59] . We previously demonstrated the importance of cAMP/PKA pathway in regulating S . pombe aging by showing that knocking out the only catalytic subunit of the PKA complex results in increased chronological life span as well as enhanced stress resistance [21] . However , other nutrient-signaling pathways may activate PKA complex in yeast , so the specific role of glucose signaling in the longevity of S . pombe was unknown . We show here that low glucose levels increase CLS in S . pombe , a typical CR response . Also , mutants with a defective Git3/PKA pathway have an increased life span , a normal glucose consumption rate , and only a slightly reduced growth rate . The reverse is also true . High glucose concentration , acting through the Git3/PKA pathway , promotes aging and decreases stress defense and respiration . Likewise a constitutively active Gα subunit , normally coupled to the Git3p GPCR , mimics the effects of high glucose even in low glucose . Further support for a role of glucose signaling in the control of CLS in S . pombe was obtained by studying hexokinase deletion strains . These mutants die prematurely in stationary phase concomitant with prolonged stimulation of Git3/PKA signaling . Since cells without hexokinase cannot metabolize glucose , these results suggest that sustained glucose signaling , caused by the excess of extracellular glucose that remains in the medium of hexokinase mutants , promotes aging in S . pombe . The loss of Git3p GPCR blocks the detrimental effects of glucose in double hexokinase mutant . This suggests that glucose exerts a strong pro-aging effect via the Git3/PKA signaling pathway . Notably , the premature death of double hexokinase mutant due to high glucose is concomitant with an accumulation of ROS . It is remarkable that the effect of deleting hexokinases differs between S . pombe and S . cerevisiae . In the budding yeast , deletion of all major hexokinases ( glucokinase , hexokinase 1 and 2 ) impairs cAMP production and activation of the PKA pathway [60] . Conflicting with these data , we show that in S . pombe , hexokinase mutants die prematurely due to sustained signaling through this pathway . Careful examination of our results also reveals that hexokinase mutants have a defective PKA pathway during the exponential phase of the cultures . However , these mutants in S . pombe enter stationary phase with high concentrations of glucose in the medium and a continual activity of the Git3/PKA pathway that is responsible for their premature aging . In contrast , hexokinase mutants in S . cerevisiae have longer RLS and a normal CLS [8] , [43] . This apparent discrepancy could be the result of particular differences in the glucose-signaling pathways and energy metabolism between S . pombe and S . cerevisiae . For instance , the regulation of glycolysis is different between these two yeasts . S . cerevisiae growth on glucose is sensitive to trehalose biosynthesis whereas S . pombe is not [61] . Despite the very significant role of Git3/PKA pathway in the pro-aging effect observed in the double hexokinase mutant , our work showed that the signal from the Git3p GPCR dependent pathway is not the only regulator of all the effects on aging due to glucose . First , in minimal medium completed ( SDC ) , lowering glucose concentration had no effect on longevity ( data not shown ) . Nevertheless , glucose decreased longevity when S . pombe were grown in synthetic medium based on yeast nitrogen base [26] . Other nutrient limitation is suspected to affect PKA-regulated processes . For instance , conjugation efficiency is controlled in both Git3/PKA cAMP-dependent manner and in a cAMP/PKA independent manner sensitive to medium composition [62] . This PKA-independent nutrient sensing could mimic the effect of glucose restriction in rich medium and may explain why the life span of S . pombe grown in SDC is not affected by glucose . This explanation is consistent with the fact that the respiration rate in 2% glucose is higher in synthetic medium than in rich medium ( complex medium ) [26] . Another indication that the signal from the Git3/PKA pathway is not the only one to control the rate of aging is provided by the glucose receptor mutant ( Δgit3 ) . This deletion strain still responds to CR with a higher oxidative stress resistance , lower ROS levels and an increased survival ( Figure 4C , 2C and 4A ) . In agreement , studies in Caenorhabditis elegans [63] , Drosophila [64] and mice show that disabling the insulin/IGF-1 signaling pathway can cooperate with CR to increase longevity [65]–[72] . What are the other possible mechanisms by which glucose could accelerate aging in S . pombe ? In budding yeast , glucose activates the glucose repression pathway , which is regulated by the AMP-activated protein kinase ( AMPK ) Snf1p complex [53] , [73] . So the Git3/PKA-independent effect of glucose could be explained by the activation of AMPK complex which affects aging in yeast and metazoans [74] , [75] . Our S . pombe hexokinase deletion mutants are expected to be defective in this pathway [53] , [76] but they still age prematurely when grown in high glucose concentrations , suggesting that glucose repression is not involved in the pro-aging effects of glucose . Conversely , we could not discard the possibility that some pro-aging effects of glucose are mediated by the non-enzymatic glycosylation of proteins by glucose . Nevertheless , altogether our data point toward a regulation of longevity primarily via the glucose signaling through Git3/PKA pathway , raising the question about the underlying mechanisms . Although further work is required to discover the mechanisms by which glucose signaling accelerates aging in S . pombe , our current evidence points to the mitochondria as the target of glucose signals . First , CR ( low glucose ) in wild type yeast enhances respiration and mitochondrial membrane potential , prevents ROS production and improves oxidative stress defense . Second , Δgit3 cells have a similar phenotype and , in addition this strain displays a higher expression of cytosolic superoxide dismutase in stationary phase . These could explain the additional longevity of Δgit3 cultured under CR conditions ( Figure 7 ) . In yeast , the cAMP/PKA glucose sensing pathway possibly represents the ancestor pathway of insulin/IGF-1 signaling in multicellular eukaryotes . This pathway signals the presence of glucose , the preferred energy source . It also controls stress resistance , growth rate and sexual development , modifies mitochondrial metabolism , and ultimately controls life span as we have shown in this study . Similar to our observations for the Git3/PKA pathway in fission yeast , a decrease in the insulin/IGF-1 signal increases longevity as a function of CR in mammals . The extent to which dietary restriction may actually be effective in humans is still unknown . Our results also show that CR and loss of the Git3p GPCR cooperate to increase life span . This suggests that if this pathway is conserved in higher organisms , its inhibition may lead to an anti-aging treatment not relying on strict diets with a limited caloric content as used in animal research . Interestingly , inhibition of cAMP synthesis by the knockout of the type 5 adenylyl cyclase ( AC5 ) gene induced Raf/MEK/ERK-dependent stress resistance and lengthened life span in mice [77] . The effect of reducing glucose signaling in S . pombe also results in a decreased cAMP synthesis in response to glucose , because Git3p , via Gpa2αp , activates adenylate cyclase [32] . Since CR and inhibition of glucose signaling cooperate to extend life span in S . pombe , it would be interesting to combine agents that reduce cAMP synthesis or reduce PKA activity with CR in mammals . In conclusion , our work with S . pombe highlights the importance of glucose-signaling pathways and oxidative stress resistance in aging . Given the importance of glucose as a central metabolite , it is surprising that the pathway for glucose sensing existing in S . pombe has not been found yet in mammals . Whether a glucose receptor contributes to these signaling pathways in metazoans remains to be demonstrated . Our data together with the interesting phenotype of the AC5 KO mice provide the rationale for further inquiry into glucose sensing pathways in mammals .
This study was conducted according to the principles expressed in the Declaration of Helsinki . MM refers to Edinburgh Minimal Medium [78] complemented by adenine , uracil , leucine and/or histidine 75 mg . L−1 ( A , U , L , H ) . SMC refers to synthetic medium complemented and is composed of MM plus adenine , uracil , leucine and/or histidine 444 mg L−1 ( A , U , L , H ) . Its composition is described in a previous study [21] . The same medium with glycerol 3% ethanol 0 . 2% for carbon source was named SMC glycerol . Mating and sporulation were carried out on MEA plates ( bacto malt extract 3% , glucose 0 . 4% , pH 5 . 5 , supplemented by adenine , histidine , uracil , leucine 225 mg . L−1 each ) . Yeast extract complete medium ( YEC ) , was made of yeast extract 5 g . L−1 ( BD , Difco ) supplemented with 222 mg . L−1 of adenine , uracil , leucine and histidine , and glucose 2% unless otherwise specified . All cultures were incubated at 30°C in a rotating incubator shaker at 250 rpm ( New Brunswick instrument ) . Growth curves represent the average of three independent cultures . Morphological analysis of wild type cells in YEC glucose 0 . 05% , 0 . 2% , 0 . 5% or 2% was done in 10 mL cultures in 50 mL conic tubes with air-permeable cap grown overnight . Early log phase refers to OD595 0 . 5 . The strains used in this work are all described in supplementary Table 1 . Wild type refers to strain SP14000 , except for Figure 2B , 5C , S4 in which it refers to FWP87 . The gpa2R176H ( RWP1 ) [32] Δgit3 deletion ( CHP984 ) [28] , Δhxk1 and Δhxk2 deletions ( CJM387 , CJM389 ) [50] alleles were previously described . The double Δhxk1 Δhxk2 mutant was constructed as follows . Δhxk2 ( CJM389 ) was transformed with a plasmid bearing the hxk2+ ORF , previously amplified by PCR and inserted into the SalI site of pREP41 ( pREP41_Hxk2 ) . PCR primes sequences will be provided upon request . The Δhxk2 pREP41_Hxk2 strain ( SP14405 ) was mated with Δhxk1 ( CJM387 ) . Corresponding diploids were sporulated in MES media and spores hxk1::ura4+ hxk2::his3+ harbouring the pREP41_Hxk2 plasmid were selected on MMA media . Haploids were grown to saturation in liquid SMC supplemented with adenine and leucine 222 mg . L−1 and with glycerol 3% , ethanol 0 . 2% as carbon sources . Then , they were diluted in the same fresh medium and cultured a second time to saturation in order to force cells to lose pREP41_Hxk2 plasmid . At this point , clones without plasmids were selected on plates SMC AL glycerol . The loss of pREP41_Hxk2 plasmid was validated by verifying that these clones Δhxk1 Δhxk2 ( SP14483 and SP14493 ) cannot grow without leucine , the marker on the pREP41 plasmid . In addition , these clones cannot grow on SMC AL glucose 2% . The triple mutant Δhxk1 Δhxk2 Δgit3 was obtained by first constructing a Δhxk1 Δgit3 double knockout ( SP14373 ) after mating the single mutants Δhxk1 ( SP14313 ) and Δgit3 ( SP14105 ) . The resulting Δhxk1 Δgit3 strain was crossed with Δhxk2 pREP41_Hxk2 ( SP14405 ) and the haploid strain Δhxk1 Δhxk2 Δgit3 without plasmid was isolated as described previously for Δhxk1 Δhxk2 . Three independent cultures of each double hexokinase mutant Δhxk1 Δhxk2 and Δhxk1 Δhxk2 Δgit3 , were started in 75 mL ( 250 mL flask ) YEC glycerol and incubated 24 hours . At OD595 0 . 6 , 0 . 1 mL were harvested , washed in sterile water and serial diluted to be plated as drop test on solid YEC glycerol 3% ethanol 0 . 2% glucose 2% or 0 . 2% or no glucose . Plates were incubated 8 days at 30°C . The same 75 mL cultures were grew until OD595 5 to 6 and split in 3 times 25 mL cultures , one let with glycerol only , one complemented with glucose 2% , one with glucose 0 . 2% . These 18 cultures were then studied as described below . The frequency of cells able to recover the ability to grow on glucose in Δhxk1 Δhxk2 mutants was measured by plating serial dilution of 100 µL of a saturated culture of SP14383 and SP14393 on SMC AL glycerol and on SMC AL glucose and by counting colonies forming units . The average of the ratios of six independent clones of SP14383 and SP14393 was 1 . 6×10−6 . Because of the very low frequency of this event and the long time revertants take to grow and reach a significant number , we consider that these revertants did not influence our data . The protocol for CLS measurement by CFU counting has been described previously [21] except that the first estimation of the number of living cells was delayed . Cells that reached maximal density were harvested , serial diluted and plated 24 hours and 48 hours after the optical density was stable and maximal; the higher number of living cells from these two samples was considered as the beginning of CLS curve ( i . e . , survival 100% ) . Error bars represent standard deviation calculated from four cultures separated from a single initial culture at the end of exponential phase . Each assay was repeated at least three times . All CLS analysis were performed in YEC AULH 222 mg L−1 except in Figure 1E where the medium is SMC AUL 444 mg . L−1 glycerol 3% . For antimycine A treatment , cultures were started at OD595 0 . 2 with 20 µg . mL−1 antimycine A ( solubilized in ethanol 100% ) and CLS was measured as described above , except that cells entered stationary phase at a lower OD . Number of living cells per mL was calculated by plating dilutions of sample of the cultures as described above accepted that solid YEC glycerol was used . The concentration presented ( living cells/mL ) with standard deviation represents the average of three independent cultures . Survival analysis by Phloxin B staining was done according to a previous publication [21] , with the exception that the percentage of stained cell was obtained by counting under microscope after background subtraction . At least 500 cells were counted for each condition . Epifluorescence microscopy analyses were performed using an inverted Nikon Eclipse E800 microscope equipped with a Nikon_60 DIC H ( 1 . 4 NA ) lens and a Photometrics CoolSNAP fx CCD camera . Images were acquired using a motion-picture camera CCD CoolSnapFX 12 ( Photometrics , Tucson , AZ , USA ) bit and analysed with UIC Metamorph software ( Molecular Devices Corporation , Downington , PA , USA ) . The percent of ROS-positive cells was measured with dihydrorodhamine 123 ( DHR 123 , Sigma ) following a previously described protocol [21] . The fluorescence of this dye is activated by peroxynitrite and peroxide in the presence of peroxidase [79] . A total of 500 to 700 cells per culture were counted to determine the percentage of positively stained cells and standard deviations were calculated using three independent experiments . Staining by dihydroethidium ( DHE , Sigma ) is more specific to superoxide production [80] and was achieved as followed . 1 . 4×107 cells were collected and resuspended in 0 . 1 mL 1× PBS with DHE 50 µM and incubated 10 minutes at 30°C . The DHE solution was removed , cells were resuspended in 20 µL 1× PBS and deposited on a microscope slide with a thin layer of agarose 1% . Counting was done the same way than for DHR123 using a Cy3 filter . Flow cytometry analysis was performed following the protocol detailed in [21] excepted for cells grown in glycerol . They were sonicated 15 seconds with a Sonicator Dismembrator Fisher Scientific Model 100 set to 12 watts . FACS analysis was done using FL1 filter for DiOC6 dye and FL3 filter for DHE dye . Oxygen consumption was measured in cultures grown to cell concentrations between OD595 0 . 8 and 1 . 5 . Cells were cultured in YEC to a given OD , and then the culture was diluted in its own medium if OD595 was greater than 1 . 5 , or concentrated in its own medium by centrifugation if OD595 was less than 0 . 8 . The goal was to measure the respiration of cultures with similar concentrations and in the exact medium in which the samples were taken . 10 mL of culture , sometimes diluted or concentrated , was incubated with gentle agitation at 30°C and 5 mL was loaded in the measurement chamber at 30°C with agitation . The oxygen consumption was followed with a Clark electrode YSI model 53 oxygen monitor until all oxygen was consumed in the chamber . The calibration of the Clark electrode for the maximum oxygen concentration ( 100% ) was done on the air . The consumption was linear , the measure was recorded with a tracer Linear1100 and the slope was calculated for each sample . The result corresponding to the rate of respiration was normalized on the OD595 of the cells in the chamber and expressed in %O2 . min−1 . OD−1 . Glucose concentration was measured on the supernatants of cultures at different ODs following the protocol given in Quantichrom™ Glucose Assay Kit from BioAssay Systems® . The results presented are the averages of three independent cultures . β-galactosidase activity , expressed from the fbp1-lacZ reporter , was determined as previously described [36] except that cultures were grown in YEC to late exponential phase ( OD595 9 in glucose 2% , and OD595 2 in glucose 0 . 2% ) . CHP1229 was grown to only OD595 5 . 5 corresponding to the end of its exponential phase . Cells were cultured in YEC glucose 2% or 0 . 2% to stationary phase , and harvested 24 hours thereafter . Cultures were diluted to OD595 0 . 5 to 0 . 8 in water and submitted to various oxidative shocks at 30°C . These include 1 M H2O2 for 120 minutes; 0 . 75 M Menadione for 180 min; 0 . 5 M H2O2 for 30 min or 0 . 3 M Menadione for 90 minutes . Then , cells were washed twice with 1 mL water and serially diluted tenfold four times . Each dilution was spotted on YEC plates and incubated five days at 30°C . For re-growth on glycerol plates , cells were grown in YEC glucose 2% to OD595 0 . 5 and washed twice in water . Cells were serially diluted as described above and spotted on SMC AULH glycerol 3% . Mitochondrial membrane potential ( Δψm ) was measured with DiOC6 ( Molecular Probes ) . Yeast strains were grown over-night in 10 mL YEC using 50 mL tubes with half screw cap to allow gas exchange . Early exponential phase refers to cells harvested at OD595 from 0 . 7 to 1 . Late exponential phase refers to cultures at OD595 2 . 1 to 2 . 4 in glucose 0 . 2% and OD595 4 to 5 for cultures in 2% glucose . Stationary phase refers to cultures let 24 hours in the incubators after saturation , corresponding to OD595 2 . 4 to 2 . 7 in 0 . 2% glucose and OD595 6 to 7 in 2% glucose . Then , 1 . 4×107cells were collected , concentrated in 0 . 1 mL of their own medium and incubated in DiOC6 0 . 175 µM 15 minutes at 30°C . Next , 50 µL of this volume was diluted in 0 . 95 mL of 1× PBS and flow cytometry analysis was carried out as described above . Total RNA were reverse transcribed in a final volume of 100 µL using the High Capacity cDNA Reverse Transcription Kit with random primers ( Applied Biosystems , Foster City , CA ) as described by the manufacturer . Reverse transcribed samples were stored at −20°C . A reference RNA ( Human reference total RNA , Stratagene , Ca ) was also transcribed in cDNA . Gene expression level was determined using primer and probe sets provided upon request . PCR reactions for 384 well plate formats were performed using 2 µL of cDNA samples ( 50 ng ) , 5 µL of the Express qPCR SuperMix ( Invitrogen ) , 2 µM of each primer and 1 µM of the probe in a total volume of 10 µl . The ABI PRISM® 7900HT Sequence Detection System ( Applied Biosystems ) was used to detect the amplification level and was programmed FAST with an initial step of 3 minutes at 95°C , followed by 45 cycles of 5 seconds at 95°C and 30 seconds at 60°C . All reactions were run in triplicate and the average values were used for quantification . The relative quantification of target genes was determined using the ▵▵CT method . Briefly , the Ct ( threshold cycle ) values of target genes were normalized independently to endogenous control genes ( ▵CT = Ct target−Ct endogenous ) and compared with a calibrator ( WT 2% glucose sample C ) : ▵▵CT = ▵Ct Sample−▵Ct Calibrator . Relative expression ( RQ ) was calculated using the Sequence Detection System ( SDS ) 2 . 2 . 2 software ( Applied Biosystems ) and the formula is RQ = 2−▵▵CT . All gene expression ( RQ ) represents the average of three RQ from three independent experiments . Standard deviations were calculated with these three RQ . Two endogenous control genes were used: Top1+ and SPBC887 . 02; both selected to be highly and constitutively expressed during stationary phase . Similar results were obtained with both of them . Results showed were calculated with Top1+ . 5 mL of stationary phase culture ( day 1 ) was resuspended in 300 µL guanidinium isothiocyanate Solution ( Guanidinium Isothiocyanate 4 M , Sodium Citrate 25 mM , pH 7 . 0 , β-Mercaptoethanol 1 M ) in 2 mL screw cap tubes . 0 . 3 mL of RNase-free beads was added and vortexed 4 times 30 seconds with Bead Beater . All the homogenate was transferred to a 2 mL Phase Lock Tube ( PLG ) ( Qiagen ) . 26 µL Sodium Acetate 2 M ( pH 4 . 0 ) was added to the sample , cap the PLG tube and mix briefly . 260 µL water-saturated phenol was added to the sample , cap the PLG tube , and mix thoroughly . 75 µL Chloroform: Isoamyl Alcohol ( 49∶1 ) was added to the sample in the same PLG 2 ml tube and mix thoroughly by repeated gentle inversion . Incubate on ice for 15 minutes , and centrifuge at 13 , 000 rpm for 5 minutes in a microcentrifuge . The aqueous phase was transferred to a new pre-spin PLG 2 ml tube , 250 µL Phenol-Chloroform-Isoamyl Alcohol ( 50∶49∶1 ) was added and mixed thoroughly by repeated gentle inversion and centrifuge . In the same PLG tube , 250 µL Phenol-Chloroform-Isoamyl Alcohol ( 50∶49∶1 ) was added , then mixed and centrifuged . The resultant aqueous phase was collected; an equal volume of 100% Isopropanol was added , and mixed by repeated inversion . The solution was centrifuged at 13 000 rpm for 20 min at 4°C . The resultant supernatant was discarded and the pellet was washed 4 times with 200 µL 80% ethanol , using 2 minutes centrifugation to re-pellet the sample if necessary . The final wash was discarded and the pellet dried at room temperature . Finally , the pellet was dissolved in 100 µL RNase-free water and stored at −70°C . RNA integrity was checked on 1 . 5% agarose gel electrophoresis with RNA loading buffer ( Qiagen ) . The code in parenthesis refers to pombe genome project nomenclature git3+ ( SPCC1753 . 02c ) ; gpa2+/git8+ ( SPAC23H3 . 13c ) ; hxk1+ ( SPAC24H6 . 04 ) ; hxk2+ ( SPAC4F8 . 07c ) ; fbp1+ ( SPBC1198 . 14c ) ; sod1+ ( SPAC821 . 10c ) ; sod2+ ( SPAC1486 . 01 ) ; gpx1+ ( SPBC32F12 . 03c ) ; top1+ ( SPBC1703 . 14c ) ; unnamed a chloride channel ( SPBC1703 . 14c )
|
Lowering caloric intake by limiting glucose ( the preferred carbon and energy source ) increases life span in various species . Excess glucose can have deleterious effects , but it is not clear whether this is due to the caloric contribution of glucose or to some other effect . Glucose sensed by the cells activates signaling pathways that , in yeast , favor the metabolic machinery that makes energy ( glycolysis ) and cell growth . The sensing of glucose also reduces stress resistance and the ability to live long . Does glucose provoke a pro-aging effect as a result of its metabolic activity or by activating signaling pathways ? Here we addressed this question by studying the role of a glucose-signaling pathway in the life span of the fission yeast S . pombe . Genetic inactivation of the glucose-signaling pathway prolonged life span in this yeast , while its constitutive activation shortened it and blocked the longevity effects of calorie restriction . The pro-aging effects of glucose signaling correlated with a decrease in mitochondrial respiration and an increase in reactive oxygen species production . Moreover , a strain without glucose metabolism is still sensitive to detrimental effects of glucose due to signaling . Our work shows that glucose signaling through the glucose receptor GIT3 constitutes the main cause responsible for the pro-aging effects of glucose in fission yeast .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"biology/cellular",
"death",
"and",
"stress",
"responses",
"cell",
"biology/microbial",
"physiology",
"and",
"metabolism",
"cell",
"biology/gene",
"expression",
"cell",
"biology/cell",
"signaling"
] |
2009
|
Pro-Aging Effects of Glucose Signaling through a G Protein-Coupled Glucose Receptor in Fission Yeast
|
Stress granules ( SGs ) are membrane-less dynamic structures consisting of mRNA and protein aggregates that form rapidly in response to a wide range of environmental cellular stresses and viral infections . They act as storage sites for translationally silenced mRNAs under stress conditions . During viral infection , SG formation results in the modulation of innate antiviral immune responses , and several viruses have the ability to either promote or prevent SG assembly . Here , we show that rabies virus ( RABV ) induces SG formation in infected cells , as revealed by the detection of SG-marker proteins Ras GTPase-activating protein-binding protein 1 ( G3BP1 ) , T-cell intracellular antigen 1 ( TIA-1 ) and poly ( A ) -binding protein ( PABP ) in the RNA granules formed during viral infection . As shown by live cell imaging , RABV-induced SGs are highly dynamic structures that increase in number , grow in size by fusion events , and undergo assembly/disassembly cycles . Some SGs localize in close proximity to cytoplasmic viral factories , known as Negri bodies ( NBs ) . Three dimensional reconstructions reveal that both structures remain distinct even when they are in close contact . In addition , viral mRNAs synthesized in NBs accumulate in the SGs during viral infection , revealing material exchange between both compartments . Although RABV-induced SG formation is not affected in MEFs lacking TIA-1 , TIA-1 depletion promotes viral translation which results in an increase of viral replication indicating that TIA-1 has an antiviral effect . Inhibition of PKR expression significantly prevents RABV-SG formation and favors viral replication by increasing viral translation . This is correlated with a drastic inhibition of IFN-B gene expression indicating that SGs likely mediate an antiviral response which is however not sufficient to fully counteract RABV infection .
Viral infections initiate a number of cellular stress responses that modulate gene expression by affecting the regulation of cellular mRNA translation , localization and degradation , while promoting viral transcription , replication and translation [1] . One of the stress responses is the assembly of messenger ribonucleoprotein ( mRNP ) complexes into dynamic cytoplasmic structures known as stress granules ( SGs ) and processing bodies ( P bodies ) [2–5] . Viruses also modify cellular gene expression by initiating the transcriptional activation of type I interferon ( IFN ) genes and interferon-stimulated genes ( ISGs ) that mediate antiviral responses [6] . During viral infection , viral RNAs are recognized by different pattern recognition receptors ( PRR ) , such as RIG-I and MDA5 . This recognition triggers a series of events leading to the activation of protein kinase R ( PKR ) and the subsequent initiation of the SGs assembly [7–10] . Activated PKR mediates translation inhibition upon replication of many RNA viruses [7] by phosphorylating the eukaryotic initiation factor-2 regulatory subunit ( eIF2 α ) . Inhibition of eIF2 α activity interferes with the formation of eIF2-GTP-Met-tRNAi Met ternary complex required for the delivery of initiator Met-tRNAi to the 40S ribosomal subunit thereby stalling the translation initiation of most mRNAs [11] . Subsequent reduction of protein synthesis promotes cellular survival by limiting the consumption of energy and nutrients , and reallocating resources to the repair of cellular damages . During PKR-induced SGs formation , specific RNA-binding proteins with self-aggregating properties , such as ras GTPase-activating protein-binding protein 1 ( G3BP1 ) , T-cell intracellular antigen 1 ( TIA-1 ) , and TIA-1-related protein ( TIAR ) , recruit translationally inactive messenger ribonucleoproteins ( mRNPs ) and these complexes nucleate the formation of SGs [12] . During this process , the poly ( A ) -binding protein ( PABP ) is also sequestered into SGs [13 , 14] . After the stress removal , the mRNAs are released for translation on ribosomes or degradation in P-bodies [13 , 15 , 16] . It has become clear that a growing number of viruses , particularly RNA viruses , modulate RNA granule formation and function to maximize replication efficiency [1] . Different consequences of viral infection on these structures and their components have been described: they include induction of SGs ( which is most often transient ) , complete inhibition of SGs formation , and alternate SGs assembly and disassembly during the course of infection . Several mechanisms are used by viruses to interfere with SGs assembly . Influenza A virus blocks SGs formation throughout infection by expressing the NS1 protein , which suppresses IFN activation and serves as a potent PKR antagonist through its dsRNA binding activity [17] . Some positive RNA viruses encode proteases which cleave key SGs components leading to the disassembly of SGs [18] . Others divert SG factors into viral replication complexes to favor viral replication likely at the expense of SGs formation [19 , 20] . Among negative RNA viruses , vesicular stomatitis virus ( VSV ) produces SG–like structures containing TIA1 and TIAR that appear similar to transcription and replication inclusions as they contain viral replicative proteins and viral RNAs [21 , 22] . Thus , some viruses actively induce SG formation and utilize the stress response for their own benefit while some others inhibit SG formation , which suggests an antiviral function for these structures . Rabies virus ( RABV ) , the prototype of the Lyssavirus genus that , as VSV , belongs to the Rhabdoviridae family , causes a fatal disease that is associated with intense viral replication in the central nervous system . The viral genome consisting of single-stranded negative-sense RNA ( ~12kb ) , encodes five proteins , the nucleoprotein ( N ) , the phosphoprotein ( P ) , the matrix protein ( M ) , the glycoprotein ( G ) , and the polymerase ( L ) . The virus enters the host cell through the endosomal transport pathway via a low-pH-induced membrane fusion process catalyzed by G . Viral transcription and replication take place within Negri bodies ( NBs ) , which are cytoplasmic inclusions bodies formed during viral infection as viral factories [23] . During transcription , a positive–strand leader RNA and five capped and polyadenylated mRNAs are synthesized . The replication process yields nucleocapsids containing full-length antigenome-sense RNA which in turn serves as template for the synthesis of viral genomic RNA . After additional rounds of transcription and/or replication , neo-synthesized RNPs are transported to the cell membrane where they are assembled with the M and G proteins into virions , which are then released from the cell through the budding process . In this report , we show that during infection , RABV promotes the formation of cytoplasmic SGs that contain TIA-1 , G3BP1 and PABP . Using live cell imaging of infected cells , we also show that these structures are highly dynamic during the viral cycle . RABV-induced SGs are distinct , but close to Negri bodies as revealed by the physical proximity of both structures and by the transport of viral mRNA from NBs into SGs . The role of PKR and TIA-1 in SG formation , viral replication and IFN-B gene expression is also investigated . Our observations indicate that RABV induces PKR-dependent cell stress and innate immune responses .
Given the role of SGs in virus-induced stress responses , we investigated whether RABV infection induces the formation of SGs in infected cells . For this analysis , we used human and mouse neuroblastoma cell lines and primary neurons isolated from mice . Human glioblastoma U373-MG cells were infected with RABV ( CVS strain ) at a MOI of 3 for various times ( up to 24 h ) . Cells were then immuno-stained for G3BP1 and TIA-1 that are well-established SG-associated proteins used as SG markers [24] . Immunofluorescence microscopy revealed that , following infection , TIA-1 translocated from the nucleus to the cytoplasm and formed aggregates co-localizing with G3BP1 ( Fig 1A , first and second column ) , indicating that SGs are formed in RABV-infected cells . In contrast , SGs were not observed in uninfected cells over the time-course of the experiment . Co-immunostaining of infected cells with antibodies specific for RABV P protein that mainly localizes within the NBs revealed that NBs and SGs were distinct structures formed during viral infection ( Fig 1A ) . The kinetics of SG formation during viral infection showed that SGs were detectable at 6–8 h p . i . The number of infected cells containing SGs increased throughout the 24 h period of infection ( Fig 1B ) . Another SG-marker , poly ( A ) -binding protein ( PABP ) , colocalized with TIA-1-containing foci confirming the formation of SGs during RABV infection , although PABP was sometimes detected in NBs ( Fig 2A ) . SG induction was not specific to RABV-CVS strain , since two other RABV strains , PV and SAD B19 , also induced SG formation in infected cells ( Fig 2B ) . SGs were also observed in mouse neuroblastoma cells ( N2A cells ) and primary neurons infected by RABV ( Fig 2C ) , indicating that SG formation is a general process triggered upon RABV infection . To show that the stress response pathway of translation inhibition is activated following RABV infection , we then analyzed eIF2α phosphorylation status that is closely linked to SG formation and usually initiated by PKR activation in virus-infected cells [1 , 9 , 25] RABV enhanced eIF2α phosphorylation in infected cells . The level of phosphorylated eIF2α is already maximal at 8 h p . i ( around 5 fold ) and remained at the same level thereafter ( Fig 2D ) . This is consistent with the kinetics of SG formation ( Fig 1A ) . It has been reported that disruption of the microtubule network by drugs such as nocodazole or vinblastine affects SGs formation [26 , 27] . We therefore examined the role of microtubule network , in the formation of SGs following RABV infection . As shown in Fig 3 , depolymerization of microtubules with nocodazole , confirmed by anti-tubulin labelling , did not impair the formation of SGs in RABV-infected cells . These results indicated that the formation of SGs in RABV-infected cells did not require an intact microtubule network . To monitor the process of SGs assembly in living cells , U373-MG cells were transiently transfected with an expression plasmid for G3BP1-eGFP and infected for 14 h with the recombinant virus rCVSN2C-P-mcherry ( obtained as described in Material and Methods ) . In non-infected cells , G3BP1-GFP staining was diffuse and cytoplasmic , while , as expected , most of the infected cells contained SGs ( S1 Fig ) . In cells infected with the recombinant virus ( revealed by their inner red data punctuate structures ) , SGs formation was observed with G3BP1 relocalization in small cytoplasmic granules that increased in size over time ( Fig 4 ) . Analysis of infected cells by video-microcopy and time-lapse fluorescence ( images were acquired every 1 min during 7 h ) showed that SGs are highly dynamic structures which fuse together when in close contact ( Fig 4 , and S1 and S2 Movies in supplemental data ) . Some cells ( seven out of ten cells ) contained large SGs which were localized close to NBs and persisted for more than four hours ( Fig 4A ) . In other cells ( three out of ten cells ) , small size SGs , transiently formed at the vicinity of NBs , exhibited alternate cycles of assembly-disassembly , and finally gave rise to larger SGs ( Fig 4B ) . In both cases , RABV-induced SGs growing over time by fusion events upon contact exhibited characteristic liquid droplet behavior of non-membrane bound intracellular RNA granules [28–30] . We next explored whether viral protein synthesis is required for the maintenance of the SGs in infected cells . Cycloheximide ( CHX ) , a translational inhibitor of protein synthesis , was used in infected cells to examine the formation of SGs . CHX has been shown to be an inhibitor of SGs formation , since it traps mRNAs in polysomes by blocking translational elongation and thus preventing the formation and/or maintenance of SGs [14] . As expected , the SGs induced by sodium arsenite were disrupted when cells were incubated with CHX for 1 h or 3 h ( Fig 5A ) . In contrast , in RABV infected cells , no major change in the number and the size of SGs was observed at 16 h p . i after incubation of infected cells with CHX for an additional 1 h or 3 h or 6 h ( Fig 5B ) . This result indicated that the maintenance of RABV-induced SGs does not require viral and/or cellular translation . These data altogether revealed that RABV-induced SGs have specific features and are not canonical SGs . In most of RABV-infected cells , SGs and NBs are in close proximity ( Figs 1 and 2 ) . To analyze more precisely the spatial relationship between both structures , we performed 3D analysis on cells infected for 24 h . Cells were immuno-stained for G3BP1 and RABV P protein to detect SGs and NBs , respectively . This analysis shows that SGs and NBs are two distinct but juxtaposed structures ( Fig 6A ) . However , it is worth noting that some G3BP1-containing foci were enclosed inside the NBs , without any connection visible between these structures and the SGs that were surrounding the NBs . Moreover , these G3BP rich structures corresponded to NB areas where the P protein was less present ( Fig 6B ) , excluding co-localization of both proteins . These results showed that G3BP components were present both around the NBs and in specific areas embedded within the NBs . They also indicated that SGs and viral factories can be very close but still remain distinct structures . SGs are known to be RNA-rich granules [2] . As SGs and NBs are intimately associated , we explored whether viral RNA produced in NBs could localize to SGs in infected cells . We used 5-ethynyl uridine ( EU ) to perform short-term RNA labeling in presence of Actinomycin D ( Act D ) . As Act D inhibits cellular transcription , only viral RNA was labeled . U373-MG cells were infected at a MOI of 3 for 20 h , before treatment or not with Act D for 1 h . Cells were then incubated in the presence of EU for 45 min . After fixation and permeabilization , cells were treated for detection of EU incorporation into nascent RNA and were simultaneously immuno-stained with anti-G3BP1 and RABV P antibodies for detection of SGs and NBs , respectively , for analysis by confocal microscopy . Without Act D , cellular RNAs were mainly detected in the nucleus , as expected ( Fig 7A , upper panel ) . In presence of the drug , RABV RNAs were predominantly localized to NBs which are sites of viral transcription and replication ( Fig 7A , middle panel ) as previously shown [23] . In some viral factories , cell magnification revealed the presence of viral RNA as ponctate dots that sometimes corresponded to G3BP1-rich structures ( Fig 7A , lower panel ) . To analyze the fate of viral RNAs , we performed a pulse-chase analysis . Newly synthetized RNAs were labelled as described above . The cells were then incubated for 3 h in presence of Act D in new culture medium deprived of EU . In absence of Act D , EU labeling was essentially nuclear but RNA accumulation in the SGs was also detected ( Fig 7B , upper panel ) . In cells treated with Act D , viral RNAs were mostly detected in SGs located in close proximity to NBs although some RNA dots were still present inside viral factories apparently associated with G3BP1-rich structures ( Fig 7B , middle and lower panel ) . Taken together , these results indicated that viral RNAs were recruited from NBs to SGs . To identify viral RNAs recruited in SGs , we performed fluorescent in situ hybridization ( FISH ) analysis by using specific oligonucleotide probes detecting either viral N mRNAs or viral genomic RNA and a poly-A probe detecting both cellular and viral mRNAs . Infected cells were simultaneously prepared for FISH and immuno-stained with anti-G3BP and anti-P antibodies . As expected , SGs contained poly A+ mRNA that could have either cellular or viral origin . Viral mRNAs , such as the P mRNA , were detected in both NBs and SGs whereas viral genomic RNAs were localized exclusively in NBs ( Fig 8A ) . These results indicated that viral mRNAs , but not genomic RNA , were selectively transported into SGs . In order to determine whether the transport of viral RNA from NBs to SGs required microtubule network , infected cells were mock-treated or treated with nocodazole and viral RNA was detected by long-term EU labeling ( for 3hours in presence of ActD ) . In both ( mock or NCZ ) conditions , viral RNA accumulated in SGs which most often are in contact with NBs ( Fig 8B , arrowheads ) . Importantly , viral RNA was also detected in SGs which are far from NBs ( Fig 8B , arrows ) . This result indicated that the transport of vRNA from NBs to SGs is independent of microtubule network . TIA-1 , a major component of SGs has been described to be essential for their formation [31] . To investigate the role of TIA-1 in RABV replication , we analyzed the effect of down-regulation of TIA-1 by RNAi-mediated silencing . U373-MG cells were transfected with a pool of four siRNAs targeting TIA-1 ( 25 nM ) or with non-targeting control ( siScr ) for 48 h , then cells were infected . Cells were harvested at 24 h post-infection and lysates analyzed by western blot to determine the expression of TIA-1 , viral proteins and tubulin as control ( Fig 9A ) . TIA-1 has two isoforms ( TIA-1a 42 kDa and TIA-1b 40 kDa ) generated by alternative splicing [32] and differentially expressed depending on the cell type [33] . In U373-MG cells , isoform expression ratio was in favor of TIA-1a ( Fig 9A ) . As expected , TIA-1-targeting siRNAs inhibit TIA-1 protein expression in infected cells ( around 80% of inhibition ) . TIA-1 depletion had no major effect on cell viability ( >95% ) measured by trypan blue exclusion , and resulted in a significant but moderate increase ( around 2-fold ) of the amount of the viral P protein at 24 h p . i . ( Fig 9A ) , indicating that down regulation of TIA-1 expression resulted in an enhancement of viral protein expression . These results led us to examine RABV infection in TIA1-knockout murine embryonic fibroblasts ( TIA-1-/-MEFs ) compared to wild-type MEFs ( WT MEFs ) . Analysis of cell extracts by western blot confirmed the lack of TIA-1 expression in TIA1-knockout murine embryonic fibroblasts TIA-1-/- MEFs , whereas WT MEFs expressed the two TIA-1 isoforms in roughly a 1:1 ratio ( Fig 9B ) . In total absence of TIA-1 , the viral P protein expression was very efficiently enhanced ( around 20-fold ) , resulting in an increase of viral production ( of ~5-fold ) at 24 h p . i ( Fig 9B ) . In contrast , similar mRNA transcript levels of RABV P and N genes were detected in TIA-1-/- MEFs in comparison to control cells ( Fig 9C ) . These data revealed that TIA-1 has an antiviral effect affecting the translation of viral proteins without modifying viral transcription . We also analyzed whether the TIA-1 depletion impairs SGs formation . Strikingly , RABV-infected TIA-/- MEFs exhibited formation of SGs as efficiently as WT MEFs did , as shown by G3BP1 staining ( Fig 9D ) . Along this line , poly A+ mRNA and viral mRNA were detected in SGs formed in both cell lines ( S2 Fig ) . In addition , the kinetics of SGs formation during viral infection were similar in TIA-/- MEFs compared to WT MEFs . In both cell lines , the number of infected cells containing SGs increased throughout the 24 h period of infection ( Fig 9E ) . Furthermore , the SGs devoid of TIA-1 did not present any obvious morphological differences from the SGs formed in the WT MEFs . These results indicated that TIA-1 has little effect on the quantity of SGs and the kinetics of their formation . We also examined whether TIA-1 was involved in the formation of SGs in a non-infectious context . The wild-type and TIA-1-/- MEFs were treated with sodium arsenite to induce SGs formation . In both cell lines , G3BP1 formed similar aggregates ( S3 Fig ) . This result is in contrast with a previous study demonstrating that the formation of SGs is significantly impaired in TIA-1-/- MEFs [31] , but supports recent findings showing that the impairment of SGs formation requires the concomitant depletion of TIA-1 , TIAR and G3BP1 [34] . We next determined whether PKR is the kinase responsible for SGs formation during RABV infection . Down-regulation of PKR expression was performed by small interfering ( siRNA ) . U373-MG cells , previously transfected with a pool of four siRNAs targeting PKR ( 25 nM ) or with non-targeting control ( siScr ) for 48 h , were infected for 20h . Silencing efficiency and viral protein expression were assessed by western blot analysis . PKR-targeting siRNA , but not control siRNA , clearly inhibited PKR expression in infected cells ( around 50% to 70% of inhibition ) ( Fig 10A ) , PKR-silenced cells were then examined for the presence of SGs using indirect immunofluorescence for both SGs markers , TIA-1 and G3BP1 ( Fig 10B ) . In infected siScr-treated cells , we observed a punctuate G3BP1 and TIA-1 staining pattern which is characteristic of SGs assembly in infected cells ( see Fig 1A ) . In contrast , in infected siPKR-treated cells , G3BP1 remained mostly diffuse in the cytoplasm , TIA-1 localized in the nucleus , and , as a consequence , SGs were not detectable ( Fig 10B ) . In cells treated with sodium arsenite , which induces SGs formation via HRI kinase [31] , SG assembly was not impeded by siPKR treatment ( S4 Fig ) . Taken together , these results indicate , that PKR is specifically required for SGs formation following RABV infection . As SGs may play a role in antiviral responses , we have further analyzed the role of PKR depletion on RABV infection . A significant increase ( 4 to 5-fold ) both in viral P protein expression ( Fig 10A ) and viral production ( Fig 10C ) was observed in PKR depleted cells in the absence of SGs . The fact that more P was located outside NBs in siPKR-treated cells could be due to the increase of P protein concentration as P ( in contrast to N ) was not exclusively located in NBs . In contrast , PKR silencing did not affect the mRNA transcript levels of RABV N and P genes in comparison to cells transfected by non-targeting siRNA used as control ( Fig 10D ) . This indicated that PKR silencing led to increased translation of viral proteins without modifying viral transcription . These data demonstrated that PKR has an antiviral effect on RABV infection . PKR , besides its role in inhibition of cellular translation and SGs formation , may also be important for type I IFN gene induction ( essentially IFN-B gene ) and modulation of antiviral innate immune responses [35–37] . As PKR displayed an antiviral effect in RABV-infected cells , we tested whether this kinase participates in IFN-B gene regulation . IFN-B gene expression was determined by qRT-PCR at different times post-infection . IFN β mRNAs were detected at 6 h p . i and gradually increased up to 24 h p . i ( Fig 10E ) . In infected cells depleted for PKR , inhibition of PKR expression dramatically decreased the RABV-induced IFN-B gene expression ( Fig 10F ) . These results showed that silencing of PKR resulted both in full inhibition of SGs formation and inhibition of IFN-B gene induction . To further investigate a potential link between stress pathway and IFN pathway , we focused on the RIG-I-like receptors such as RIG I and MDA5 . Therefore , we analyzed their localization in RABV-infected cells by confocal analysis . MDA5 co-localized with TIA-1 in SGs ( Fig 10G ) whereas we detected no clear colocalization of RIG-I in SGs ( S5 Fig ) . Taken together , these results indicate that RABV infection initiates PKR-dependent cellular stress pathways leading to formation of SGs and initiation of antiviral innate immune responses .
The SGs are detectable as soon as 6 h post-infection and the number of infected cells containing SGs increases throughout the infection . Their formation is a general process observed for different RABV strains and in different cell types ( either neuronal or non-neuronal cells ) . SGs formed in RABV-infected cells are distinct from the canonical SGs induced under stress conditions . First , their morphology and size are heterogeneous with an irregular shape compared to the canonical SGs . Second , their formation and dynamic are independent of an intact microtubule network ( Fig 3 ) in contrast to canonical SGs which require microtubule integrity [26] . Microtubule-independent SGs formation has also been reported in the case of VSV-induced SGs which do not require an intact cytoskeleton for their formation [21] . Third , their maintenance during viral infection does not require viral and/or cellular translation , since cycloheximide treatment does not result in their disappearance in contrast to bona fide SGs ( Fig 5A and 5B ) . Similar results have been observed for the “antiviral SGs” induced by vaccinia virus infection [38] . Real time imaging of cells expressing GFP-G3BP1 and infected with recombinant fluorescent RABV reveals that SGs are highly dynamic structures . First , small granules are formed throughout the cytoplasm; then , they fuse to form larger structures . Our data show that RABV-induced SGs undergo different fates over time: they can persist in infected cells or exhibit assembly/disassembly cycles ( Fig 4 ) . Our results indicate that RABV-induced SGs behave as liquid droplets . They are spherical and they grow over time by fusion events upon contact . In addition , they are composed of RNA and RNA-binding proteins , which as G3BP1 , are enriched in intrinsically disordered and aggregation-prone domains and have been suggested to promote cytosolic phase transition [39] . Thus , our data are consistent with recent studies indicating that cytosolic liquid-phase transition is a general mechanism underlying the formation of RNA granules such as P bodies and SGs [28–30] . SGs induced by RABV are located close to cytoplasmic viral factories ( NBs ) in infected cells ( Fig 6 ) , similarly to antiviral SGs formed in vaccinia virus-infected cells [10 , 38] . This is not the case of VSV-induced SG-like structures which co-localize or even share the same structure with the viral factories [21] . For RABV , SGs and viral factories appear as distinct structures , even when they are in close contact with one another . Indeed , in some cells , the SG marker G3BP1 is located inside the NBs in specific areas from which viral P proteins are excluded . Pulse-chase experiments reveal that viral mRNAs , although synthesized within NBs [23] , are later detected in SGs . As shown by FISH analysis , both viral and cellular mRNAs accumulate in the SGs ( Figs 7 and 8A ) . In contrast , no viral genomic RNA is detected in RABV-induced SGs . These findings indicate that viral mRNAs are specifically transported from NBs to SGs . This transport does not require microtubule network as nocodazole treatment has no effect on the accumulation of viral mRNAs in SGs ( Fig 8B ) . However , we cannot exclude that a direct contact between NBs and SG is sufficient for the transfer of viral RNA in SGs . Indeed , both structures which are very dynamic might have been in contact at some point of infection . Since SGs have been proposed to be sites of mRNA sorting during translation inhibition , it is tempting to speculate that RABV mRNAs may be sequestered from the translation competent pools of mRNAs into SGs . Along this line , different species of viral RNA ( messenger or genomic RNA , as well as unusual viral RNA species formed during RNA synthesis ) have been detected in such compartments for positive or negative sense-RNA viruses [40–43] . The formation of bona fide SGs ( formed upon arsenite treatment of cells ) has been reported as being dependent on TIA-1 , TIAR and G3BP proteins [44 , 45] . Our results indicate that TIA-1 is not essential for the formation of arsenite-dependent SGs . Similarly , although TIA1 is recruited into SGs during viral infection , RABV-induced SGs do not require TIA-1 for their formation , as the SGs are formed efficiently in the absence of TIA-1 after infection of TIA-1-/- MEFs or si-TIA-1-treated U373-MG cells ( Fig 9D ) . This is correlated with recent findings showing that the impairment of SGs formation requires the concomitant depletion of TIA-1 , TIAR and G3BP [34] . Although TIA-1 depletion does not inhibit SGs formation , it results in a positive effect on viral gene expression and viral production ( Fig 9A and 9B ) in RABV infected cells demonstrating the antiviral effect of TIA-1 . This restrictive effect of TIA-1 has been described for VSV [21] and for several other viral infections [41] . It has been shown that TIA-1 inhibits VSV growth at the level of viral gene expression and/or replication [21] . We show here that TIA restricts RABV infection at the level of viral translation , emphasizing the antiviral function of TIA-1 . RABV-induced SGs are dependent on PKR expression as inhibition of PKR expression significantly prevents SGs formation ( Fig 10 ) . In this condition , viral protein expression and viral production increase indicating that PKR depletion favors viral replication ( Fig 10A and 10C ) . In contrast , viral mRNA levels are unchanged in the absence of PKR ( Fig 10D ) . These results are in accordance with the well-known role of PKR as an antiviral protein and a crucial sensor of viral infection whose activation results in phosphorylation of eIF2α and inhibition of translation initiation [11 , 46] . As inhibition of PKR expression also dramatically decreases the RABV-induced IFN-B gene expression ( Fig 10E ) , our results indicate that PKR has a pivotal role in the initiation of cellular stress pathways leading to formation of stress granules and antiviral innate immune responses . Multifaceted roles of PKR in the regulation of stress granule formation and virus-induced gene regulation have been documented in recent years [25 , 36 , 47–49] . PKR has also been reported to be located in virus-induced SGs where it colocalizes with viral RNA sensors RIG-I and MDA5 in order to promote their interaction with different forms of viral RNA , suggesting an antiviral role for SGs [8 , 49] . The inhibitory effect of PKR depletion on SGs formation , increased phosphorylation of eIF2α subunit in RABV-infected cells and concomitant decrease in IFN-B gene expression levels indicate that SG-dependent and PKR-mediated antiviral response is triggered during RABV infection . This is consistent with the presence of MDA5 in RABV-induced SGs ( Fig 10G ) suggesting that SGs may function as a scaffold for viral RNA recognition by RLRs . Our results altogether provide some evidence that RABV induces the formation of SG-like antiviral structures as shown for several viruses [8 , 38] . Viral infections have been shown to induce [42 , 50] or suppress SGs assembly [17 , 51] . The induction is often only transient and followed , at some point of the infection , by SGs disassembly [18 , 52 , 53] , apparition of new SGs modified in their composition [19 , 20] or subversion of the SGs components for replicative advantages [10 , 54] . Our data indicate that RABV induces the formation of SGs that persist at later times of infection . Although we did not get the conditions which totally inhibit SG formation ( except by PKR depletion as discussed above ) , our data indicate that RABV infection is efficient in SG-positive cells despite the described antiviral role of SGs . The sequestration of a pool of viral mRNAs in the SGs suggests that the virus uses the SGs to control the amount of viral transcripts to modulate viral transcription and replication . This may limit the cytopathic effect and consequently the cellular damage . However , we cannot exclude that the SG-mediated antiviral responses are counteracted by the virus . We have previously shown that RABV P protein interacts with STAT proteins , inhibits their translocation to the nucleus and prevents STAT-mediated transcription of ISG [55–57] . As IFN-induced PKR expression contributes to the amplification of PKR-mediated antiviral immunity , RABV may block PKR- and SG-mediated antiviral responses through P-STAT interactions . Other data link the stress response pathway to the interferon response by demonstrating localization of PRR in SGs following infection [9 , 25 , 43 , 58] . It has been very recently shown that the SGs induced by Newcastle disease virus , another member of the Mononegavirales order , recruit vRNA and RIG-I which trigger the induction of IFN cooperating in an efficient antiviral program [47] . Whether the ability of RABV to tolerate SG formation is due to anti-IFN effects of RABV proteins or to a beneficial role of SGs in viral infection requires further investigations .
BSR cells , cloned from BHK 21 ( baby hamster kidney ) were obtained from A . Flamand ( I2BC , Département de Virologie , former Laboratoire de Génétique des Virus , Gif , France ) , N2A cells ( mouse neuroblastoma ) and U373-MG cells ( human gliobastoma astrocytoma ) were purchased from the ATTC organization ( http://www . lgcstandards-atcc . org ) . All the cells were grown in Dulbeco’s modified eagle medium ( DMEM ) supplemented with 10% FCS ( fetal calf serum ) . Immortalized murine embryonic fibroblasts ( MEFs ) from wild-type ( WT ) and TIA1-knockout mice [59] were obtained from P . Anderson ( Harvard University ) . Mouse primary neurons were obtained from JM . Peyrin ( Université Pierre-et-Marie Curie , CNRS UMR 7102 , Paris ) as described [60] . Briefly , cortical cells were isolated from mouse embryos E16 embryos of Swiss mice ( Janvier Labs ) , and cultured for 5 days in complete neuronal culture medium DMEM glutamax ( Life Technologies , Inc . , Gaithersburg , MD , USA ) supplemented with serum-free Neurobasal ( Gibco ) and 2% B-27 supplement ( Gibco ) . The SADB19 , PV ( Pasteur Virus ) , and CVS ( Chalenge virus standard ) strains of rabies virus were grown in BSR cells . The plasmid encoding G3BP-eGFP , described by [18] was kindly provided by R . Lloyd ( Department of Molecular Virology and Microbiology , Baylor College of Medicine , Houston , USA ) . The full-length recombinant N2C ( prCVSN2C ) infectious clone was described previously [61] . The authentic P coding sequence was replaced with the P-mCherry fusion encoding sequence . The original full-length genomic plasmid was digested with AvrII and NruI restriction enzymes . Three overlapping fragments were amplified by PCR . The first one going from the AvrII site in the N gene to the end of the P coding sequence , the second one corresponding to the mCherry coding sequence and the third one going from the end of the P coding sequence to the NruI site in the G gene . The PCR products and the digested plasmid were assembled using Gibson Assembly kit ( New England Biolabs ) to obtain the resulting plasmid , prN2C-P-mCherry . Recombinant viruses were recovered as described previously [62 , 63] . Briefly , N2A cells ( 106 cells ) were transfected using lipofectamine 2000 ( Invitrogen ) with 0 . 85 μg of full-length prCVSN2C-P-mCherry , in addition to 0 . 4 μg pTIT-N , 0 . 2 μg pTIT-P , 0 . 2 μg pTIT-L and 0 . 15 μg pTIT-G , which encode respectively the N , P , L and G proteins of SAD-L16 rabies virus strain . These plasmids were cotransfected with 0 . 25 μg of a plasmid encoding the T7 RNA polymerase . Six days posttransfection , the supernatant was passaged on fresh N2A cells , and infectious recombinant viruses were detected three days later by the fluorescence of the P-mCherry protein . The rabbit polyclonal anti-P antibody was previously described [23] . The mouse monoclonal anti-G3BP-1 ( 2 F3 ) antibody was obtained from Sigma . The rabbit polyclonal anti-P antibody was previously described [23] . The rabbit anti-phospho eIF2α ( 04342 ) was obtained from Millipore . The Rabbit anti-MDA5 ( 33H12L34 ) was from Invitrogen , the rabbit anti-RIG-I ( AT111 ) was from Enzo life; the goat monoclonal anti-TIA-1 ( C20 ) and mouse anti-PABP ( 10E10 ) antibodies were from Santa-Cruz Biotechnology . Secondary fluorescent antibodies were purchased from Molecular Probes ( Alexa fluor 488- , 568- or 647- conjugated ) and Cell Signaling ( Fluor 800 or Fluor 680 conjugated ) . Nocodazole ( M1404 ) , actinomycin D ( A9415 ) , sodiumarsenite ( S7400 ) and cycloheximide ( C7698 ) were obtained from Sigma . Cells were fixed for 15 min with 4% PFA ( paraformaldehyde ) and permeabilized for 5 min with 0 . 1% Triton X-100 in PBS . Cells were incubated with the indicated primary antibodies for 1 h at RT , washed and incubated for 1h with Alexa fluor conjugated secondary antibodies . Following washing , cells were mounted with Vectashield ( Vector labs ) containing DAPI . Images were captured using a Leica SP8 confocal microscope ( 63X oil-immersion objective ) . For 3D reconstruction , confocal stacks were treated with Chimera Software and 3D rendering was carried out using the volume viewer tool in Chimera Software . To determine the number of stress granule-positive cells three wide-field 20X images were captured per experiment . Cells displaying ponctate immunofluorescent foci of G3BP-1 were considered as stress granule positive . Counterstaining with anti-P antibody was performed to discriminate between infected and non-infected cells . FISH was performed as previously described [23] , with some modifications and by using 5’-ATTO448 modified oligonucleotides ( Eurofins Genomics ) . The sequences of probes used to detect viral RNAs were previously described [23] . Messenger RNAs ( mRNAs ) were detected using 5’-ATTO488-oligo ( dT ) . Cells infected with RABV were fixed in 4% PFA for 15 min at room temperature , permeabilized 5 min with 0 . 1% TX-100 in PBS , and incubated with primary and secondary antibodies as described above . After dehydration in 70% RNase-free ethanol overnight , coverslips were rehydrated in 2X SSC ( 1X SSC: 0 . 15M NaCl , 0 . 015 M sodium citrate ) buffer , before prehybridization with 10 ng probe per coverslips in hybridization buffer ( 50% formamide ( Sigma ) , 10% dextran sulfate sodium salt ( Sigma ) , 20 μg/ml salmon sperm DNA ( Invitrogen ) , 2X SSC ) for 1 h at 60°C . Hybridization was carried out in hybridization buffer plus 10 ng probe per coverslips for 5 min at 60°C and 4 h at 37°C in the dark . Cells were washed in 2X SSC pre-warmed at 42°C and then fixed in 3 . 7% formaldehyde in PBS and mounted as described above . Infected cells ( 16 h p . i . ) were treated with 20 μM actinomycin D ( Act D ) for 1h , to inhibit cellular transcription , and then fed with 1 mM 5-ethynyl uridine ( EU; Invitrogen ) for 45 min . For pulse-chase analyses , the medium containing EU was replaced by new culture medium supplemented with Act D throughout the chase period . Cells were fixed using 4% PFA in PBS . EU labeling of cells was detected according to the manufacturer’s instructions ( Invitrogen , Click-it RNA imaging kits ) . After this step , cells were washed with PBS and incubated with antibodies as described above . U373 cells ( 2–4 105 ) transfected by siRNA and infected with RABV-CVS for 24 h were flash frozen in liquid nitrogen and stored at -80°C . Total RNA was extracted with Nucleospin RNA II kit ( Macherey Nagel , France ) and 1 μg of RNA was used for first strand cDNA synthesis by reverse transcription with AffinityScript QPCR cDNA synthesis kit ( Agilent Technologies , USA ) and oligo ( dT ) primers ( 200 ng ) . The quantitative real-time PCR was performed on the Mx3000P apparatus ( Stratagene , USA ) in a total volume of 20 μl containing the first strand cDNA template , 200 nM of each primer and 1x Mesa green QPCR master mix plus solution ( Eurogentec , France ) . Standard curves for Interferon-B gene ( IFN-B ) and the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) were obtained by using serial dilutions of human genomic DNA . The forward and reverse primers for mouse GADPH were: 5’-TCAACTACATGGTCTACATGTT-3’and 5’GGTCTCGCTCCTGGAAGAT-3’ , respectively . Forward and reverse primers for human IFN-B and GAPDH genes were 5’- GTC TCC TCC AAA TTG CTC TC ( f ) , 5’- ACA GGA GCT TCT GAC ACT GA ( r ) , 5’- ACA GCC TCA AGA TCA TCA GC ( f ) and 5’- TCT TCT GGG TGG CAG TGA T ( r ) , respectively . IFN-B mRNA levels were normalized to GAPDH expression levels that remained unaffected during viral infection . The plasmid pCW278 containing the cDNA of RABV-CVS-N2C genome was used to establish standard curves for quantification of viral N and P mRNAs and to calculate the amplification efficiency for each pair of primers used for RABV-N gene 5’- GCA GCA ATG CAG TTC TTT GA ( f ) , 5’- GTC AAT TCC ATG CCT CCT GT ( r ) , and RABV-P gene 5’- CTT GAG ATG GCC GAA GAG AC ( f ) , 5’- ACG ATT GGA ACA GGA GGT TG ( r ) , respectively . Each amplification reaction was carried out in triplicate with the following conditions: an initial denaturation at 95°C for 10 min , 40 cycles of 95°C for 10 s , 60°C for 30 s and 72°C for 10 s . The uniqueness and sizes of PCR products were checked by agarose gel electrophoresis . U373 cells mock-infected in parallel were used as controls . A pool of 4 siRNAs targeting PKR ( EIF2AK2 ) or TIA-1 were purchased from GE Healthcare ( ON-TARGET plus human EIF2AK2 siRNA SMART pool , ON-TARGET plus human TIA-1 siRNA SMART pool and ON-TARGET plus non-targeting pool ) . Cells were seeded at 5x104 per well in 24 well plates the day before . Transient transfections were performed using Dharmafect reagent according to the manufacturer’s instructions . A final siRNA concentration of 25 nM was used . A second transfection was performed 24 h after the first one with the same siRNA concentration . Cells were infected 24 h after the second transfection . Western-blot and immunofluorescence analysis were performed 24 h post-infection . The treatment with sodium arsenite ( 0;5mM ) was performed on non-infected cells 48 h post transfection . Cells were washed and re-suspended in PBS , lysed in hot Laemmli sample buffer and boiled for 5min . Proteins were separated by electrophoresis on 12% SDS-PAGE and transferred onto a nitrocellulose membrane . The membrane was blocked with 10% skimmed milk in TBS for 2 h and incubated overnight at 4°C with the corresponding antibodies . The blots were then washed extensively in TBS-0 . 5% Tween 20 and incubated for 1 h with Fluor 800-conjugated IgG or Fluor 680-conjugated IgG secondary antibody ( Cell Signaling ) at room temperature . After washing , the membranes were scanned with the Odyssey infrared imaging system ( LI-COR , Lincoln , NE ) at a wavelength of 700 or 800 nm . Protein spot levels were determined by using Image Studio software ( LI-COR ) . For live-cell imaging , U373 cells were seeded onto 35-mm micro-dishes ( Ibidi ) 24 h before transfection . Cells were transfected using Lipofectamine 2000 ( Invitrogen ) with a plasmid encoding G3BP-GFP . One hour after transfection cells were infected with CVS-N2C-P-mCherry rabies virus in DMEM FluoroBrite medium ( Invitrogen ) supplemented with 5% FCS . Live-cell time-lapse experiments were recovered with a Zeiss AxioObserver epifluorescence microscope ( 63X oil-immersion objective ) . Cells are maintained at 37°C and 5% CO2 during imaging .
|
Exposure of cells to environmental stresses , such as heat shock and viral infection , induces a cellular response leading to the formation of Stress Granules ( SGs ) composed of stalled translation initiation complexes ( RNA-binding proteins and mRNA ) . The subsequent inhibition of host translation participates to cell survival . Viruses modulate or interfere with SG formation to control viral replication and antiviral responses , but differences exist in the dynamics and outcome of the stress responses induced by various viruses . Our study shows that Rabies virus ( RABV ) induces the formation of SGs in infected cells . We combined different methods of advanced imaging techniques ( live-cell imaging , 3D analysis , FISH experiments ) to characterize for the first time these structures . SGs are highly dynamic structures that increase in size by fusion events , exhibit transient assembly or persist throughout infection . They localize close to viral factories , cytoplasmic structures characteristic of RABV infection involved in viral replication and transcription . Viral messenger RNAs , but not viral genomic RNA , are transported from the factories to SGs , indicating the communication between both compartments . In addition , we provide some evidence that RABV-induced cellular stress is dependent on double-stranded RNA-activated protein kinase ( PKR ) . Our data indicate that PKR also participates in innate immune responses through the induction of Interferon-B gene . Taken together , our results give an insight on new and important aspects of RABV infection and host antiviral stress responses .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"nuclear",
"staining",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"gene",
"regulation",
"pathogens",
"messenger",
"rna",
"immunology",
"microbiology",
"viral",
"structure",
"viruses",
"rna",
"viruses",
"antibodies",
"research",
"and",
"analysis",
"methods",
"rabies",
"virus",
"small",
"interfering",
"rnas",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"immune",
"system",
"proteins",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"viral",
"replication",
"biochemistry",
"rna",
"dapi",
"staining",
"lyssavirus",
"nucleic",
"acids",
"virology",
"physiology",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna",
"organisms"
] |
2016
|
Rabies Virus Infection Induces the Formation of Stress Granules Closely Connected to the Viral Factories
|
Voluntarily shifting attention to a location of the visual field improves the perception of events that occur there . Regions of frontal cortex are thought to provide the top-down control signal that initiates a shift of attention , but because of the temporal limitations of functional brain imaging , the timing and sequence of attentional-control operations remain unknown . We used a new analytical technique ( beamformer spatial filtering ) to reconstruct the anatomical sources of low-frequency brain waves in humans associated with attentional control across time . Following a signal to shift attention , control activity was seen in parietal cortex 100–200 ms before activity was seen in frontal cortex . Parietal cortex was then reactivated prior to anticipatory biasing of activity in occipital cortex . The magnitudes of early parietal activations were strongly predictive of the degree of attentional improvement in perceptual performance . These results show that parietal cortex , not frontal cortex , provides the initial signals to shift attention and indicate that top-down attentional control is not purely top down .
Shifting attention to the expected location of an impending visual stimulus will improve the perception of that stimulus once it occurs there [1] . This perceptual improvement is considered to be a consequence of attentional-control operations that are performed by frontal and parietal regions of the human brain [2 , 3] . According to the widely accepted top-down model of voluntary attentional control , neural activities in frontal and parietal regions control the deployment of attention in space and eventually modulate the excitability of neurons in sensory-specific areas , which are responsible for processing of the upcoming stimulus . Traditionally , it has been assumed that the frontal lobes initiate top-down attentional control , because regions in frontal cortex are involved in the executive control of other cognitive and motor operations [3] . This assumption has been built into neural models of attentional control , in which one-way pathways from frontal cortex to parietal cortex to low-level visual areas subserve the voluntary control of spatial attention ( Figure 1A ) [2] . However , there is still much debate about the precise sequence of activity in the fronto-parietal network . Some evidence has suggested that frontal cortex becomes active before parietal cortex [4] , while other evidence has suggested the opposite sequence [5 , 6] . This issue needs to be resolved in order to pin down the attentional control operations performed by the various regions in the network . For example , the latter sequence would suggest that parietal lobe is involved in the initiation of attentional control rather than the deployment or maintenance of attention in space , and thus necessitate a revision of current models of attentional control . A number of functional magnetic resonance imaging ( fMRI ) studies have confirmed the involvement of frontal and parietal lobes in the control of visual spatial attention [7–15] , but the changes in blood flow that give rise to the fMRI signal are too sluggish to investigate the time courses of activities within these brain areas ( however , attempts have been made to identify temporal order of activities using analytical techniques; see [16–18] ) . Advances in event-related fMRI have enabled researchers to separate attentional-control activity from subsequent attention effects on the neural responses to visual stimuli [11] . However , the hemodynamic response lasts for 10–20 s , whereas the neuro-cognitive operations involved in the control and deployment of attention in space each take only a fraction of a second [19] . Thus , the sequence of neural activations within the frontal-parietal network for attentional control cannot be elucidated with hemodynamic neuroimaging methods . By comparison , the scalp-recorded electroencephalogram ( EEG ) and event-related potentials ( ERPs ) triggered by sensory or cognitive processes reveal precisely the timing of brain activity associated with specific mental operations but traditionally have failed to provide precise information about the locations of active neurons . In both ERP and fMRI studies , the neural correlates of attentional control are often investigated by examining the neural activity elicited by a symbolic cue ( e . g . , an arrow ) that indicates which location to attend to in preparation for an upcoming target [20] . Typically , the neural responses between leftward-directing and rightward-directing cues are compared to one another to identify brain regions that are spatially selective for shifts of attention to particular locations [5 , 6 , 21–26] . Although this type of comparison has been useful for examining pre-target biasing in sensory areas , it has two important limitations with regards to identifying attentional control activity . First , not all of the spatially specific activities observed in the cue-target interval are related to attentional control . Some of these activities have been linked to low-level sensory responses elicited by the cue [14 , 24] , motor preparation [21] , saccadic suppression [23] , and other nonattentional processes . Second , this method cannot detect any activity that is associated with shifts of attention to both left and right locations , because such spatially nonspecific activity is subtracted away . If , for example , activity in the right parietal lobe controls shifts of attention to both left and right visual fields [27] , then that activity would go undetected . To better isolate activity related to the control of attention shifts , researchers have begun to compare activity associated with the presentation of attend cues to activity associated with the presentation of neutral cues that either provide no information about the location of the impending target ( i . e . , noninformative cues ) [6] or signify that the target will not occur ( i . e . , interpret cues ) [4 , 14 , 28] . This method controls for the presentation of the sensory cue stimuli and also permits the detection of both spatially specific and spatially nonspecific neural responses . ERP and fMRI studies using this isolation method have provided converging evidence for bilateral activity in frontal and parietal regions of cortex [4 , 6 , 14]; but unfortunately , the sequence of attentional control activities in these regions has remained unclear . One recent study that isolated attentional control with an interpret cue reported findings consistent with the top-down model of attentional control illustrated in Figure 1A using an fMRI-constrained dipole-modeling approach [4] . Neural sources of the grand-averaged attend-minus-interpret ERP difference waveforms were modeled with four dipoles placed at the coordinates of the bilateral frontal and parietal activations observed in a similar fMRI task [14] . The orientations of the dipoles were varied until the dipole model accounted for as much of the scalp-recorded ERP data as possible in the 400–1 , 900-ms time interval . The resulting fMRI-constrained model suggested that the left parietal source was active 200–300 ms after cue onset . Subsequent bilateral frontal source activity began 400 ms after cue onset and was sustained until target onset . Sustained bilateral activity was also seen in the parietal source waveforms beginning at 650 ms . Follow-up analyses suggested that the early left parietal source activity was not statistically significant; thus it was concluded that frontal cortex initiated attentional control about 400 ms after cue onset . However , the early parietal activity may have been obscured in three ways . First , the ERPs elicited by leftward and rightward directing cues were averaged together , thereby minimizing any spatially specific effects that might have occurred early in parietal cortex . Second , the analyses were not ideally designed to pick up small , transient ERP effects that may have occurred early in the cue-target interval . For example , differences between attend-cue ERPs and interpret-cue ERPs were analyzed statistically by measuring mean ERP amplitudes within consecutive 100-ms intervals that were not centered on any peaks in the attend-interpret difference waveforms . Moreover , the fMRI-constrained dipoles were not fit to the difference waveforms in the early ( 0–300 ms ) portion of the cue-target interval . Third , the fMRI-constrained dipoles may have been at suboptimal locations to pick up any early activity in the parietal lobes . Another study that isolated attentional control with a spatially noninformative cue reported findings that were inconsistent with the top-down model of attentional control illustrated in Figure 1A [6] . Bilateral activity was observed over frontal and parietal scalp sites , primarily at electrodes on the same side ( ipsilateral ) as the to-be-attended location , in the 300–450 ms time interval . This fronto-parietal activation was preceded by activity over the right parietal scalp at 250 ms , which suggests that right parietal cortex might initiate the sequence of attentional control . However , dipole source modeling of the isolated attentional control activity revealed sources in temporal , rather than parietal , cortices and was rejected as being physiologically implausible . Consequently , the neural sources of the early ERP activity seen over the parietal scalp remain unknown . In addition , some of the activities seen in the attend-neutral ERP difference waveforms may have reflected differences in overall arousal or motivation , because the attend cues and neutral cues were presented in separate tasks . Given the results of the two studies that isolated attentional control with neutral cues , it is possible that parietal , rather than frontal , cortex initiates attentional control in the spatial cueing paradigm . To date , however , the methodological and analytical procedures used to investigate the sequence of attentional control in the fronto-parietal network have been insufficient to verify this hypothesis . Here we capitalized on recent advances in EEG source reconstruction to clarify the timing and sequence of activity related to attentional control . We examined event-related changes in EEGs recorded from 11 participants during an attention-cueing task [20] , in which a cue presented at fixation indicated the likely location of an impending target ( Figure 2A ) . This task enabled us to separate the neural activities associated with the cue-induced orienting of attention from the subsequent effects of attention on target processing . In addition , we included a subset of trials on which the cue provided no information about the location of the upcoming target . By comparing activity elicited by these noninformative ( no-shift ) cues with the activity elicited by the informative ( shift ) cues , we were able to isolate further the neural activities associated with attentional control from those associated with the sensory processing of the cue itself . We reconstructed the neural sources of EEG attentional control activity using a beamformer spatial filtering method [29 , 30] . The beamformer approach has several advantages over the dipole modeling approach . First , the beamformer method does not require a priori determination of the number of neural sources that may be giving rise to the scalp-recorded electrical fields . Second , the beamformer method outputs a volumetric image of neural activity throughout the brain , thereby facilitating the comparison of our results with those obtained from previous fMRI studies . Third , the beamformer method can be used to reconstruct neural sources of EEG in specific frequency bands . This enabled us to focus on oscillatory activity that we hypothesized would be important for visualizing attentional control activity across the entire cortex . Prior studies have linked alpha band ( 8–14 Hz ) and gamma band ( >30 Hz ) oscillations to attention and perception [31] , but scalp-recorded oscillations in these frequency bands are primarily associated with the consequences of attention on activity in visual sensory areas [32–34] rather than the preceding attentional control operations in frontal and parietal cortices . To specifically examine attentional control activity , we opted to focus our beamformer analysis on the low-frequency theta band ( 4–7 Hz ) oscillations . Although there is little or no existing evidence linking theta band activity to attention , we hypothesized that focusing on theta band oscillations would enable us to visualize attentional control activity across the cortex , because theta band oscillations have the following properties: ( 1 ) they reflect long-range communications between distant brain areas [35]; ( 2 ) they are carrier frequencies for high-frequency oscillations that reflect communications between nearby neurons ( e . g . , within a region ) [36]; and ( 3 ) they have been previously linked to the working memory system [37] , which is known to overlap with the spatial attention system [38] . To maximize our ability to home in on the attentional control areas that were identified in previous fMRI studies , we included both the evoked ( phase-locked ) and induced ( non–phase-locked ) activities in the analysis , because both would contribute to the hemodynamic response measured with fMRI . We imaged neural sources of theta activity in each of 18 consecutive 50-ms intervals between cue and target . The reconstructed EEG source activities were then subjected to nonparametric statistical analyses [39] to determine which brain areas showed significant increases in activity associated with shifting attention . Based on previous electrophysiological studies , we made two predictions about the sequence of theta band activity during the voluntary control of visual attention . If voluntary attentional control is initiated in a completely top-down manner [4] , activity would be seen first in frontal cortex , then in parietal cortex . Alternatively , if attentional control is initiated in parietal regions [5 , 6] , activity should be seen first in parietal cortex and then in frontal cortex . Our results supported this latter hypothesis . Following a signal to shift attention , control activity was seen in parietal cortex 100–200 ms prior to activity in frontal cortex . Parietal cortex was then reactivated prior to anticipatory biasing of activity in occipital cortex .
Figure 2B displays surface-rendered maps of significant theta band activity for shift-up cues ( relative to noninformative cues ) in six representative time intervals . Activity associated with attentional control was observed in posterior brain areas during the first 300 ms following the appearance of the attention-directing cue . Initially , the activity was confined primarily to extrastriate regions of the occipital lobe , but by 200 ms , both the superior and inferior parietal lobes became active , and by 300 ms , the frontal lobes became active . Between 400 and 600 ms , the activity was confined to the inferior , middle , and superior frontal gyri . Following the activity in the frontal lobes , posterior parietal cortex became active for a second time ( 600–700 ms post-cue ) . During this second activation , activity was seen in the inferior , but not the superior , parietal lobule . This parietal activity was then followed by a second phase of activity in extrastriate visual cortex that extended along the middle and inferior occipital gyri into the inferior temporal lobes . To better characterize the spatio-temporal sequences of neural activities involved in attentional control , we plotted the normalized power changes in theta band activity for the shift-up cue relative to the noninformative cue across the entire cue-target interval in occipital , parietal , and frontal regions of interest ( ROIs ) ( Figure 3A ) . Activity in the inferior occipital gyrus ( IOG ) occurred in two phases , with an early peak at approximately 150–200 ms after the cue and a late phase that began approximately 600 ms after the cue and continued until the onset of the target stimulus . Activity in the inferior parietal lobule ( IPL ) showed a similar biphasic pattern . Notably , however , the first phase peaked later than in IOG , and the second phase peaked earlier . Activity in the superior parietal lobule ( SPL ) peaked early , around the same time as the initial peak activation in IPL , whereas activity in the middle frontal gyrus ( MFG ) peaked in the middle of the cue-target interval ( 300–600 ms post-cue ) . The sequence of peak activations across these ROIs suggests that an initial feed-forward sweep of activity sends information to executive control areas in frontal cortex , which then sends information back to lower areas . Similar patterns of attentional control activity were observed following shift-left and shift-right cues . In the case of shift-left and shift-right cues , however , some of the attention-related activity was lateralized ( i . e . , spatially specific ) . As shown in Figure 4 , initial occipital activity following these cues was observed predominantly in the hemisphere contralateral to the to-be-attended location ( i . e . , the right hemisphere for shift-left cues and the left hemisphere for shift-right cues ) . The early activity in SPL was bilateral , whereas the early activity in IPL was greater in the hemisphere ipsilateral to the to-be-attended location than in the hemisphere contralateral to the to-be-attended location . Subsequent activations in MFG and occipital cortex were also larger in the ipsilateral hemisphere , whereas the late activity in IPL was bilateral . The early occipital and parietal activations are inconsistent with current models of top-down attentional control , according to which the signal to shift attention originates in frontal cortex [2] . Because our informative cues differed from the noninformative cue in one important respect—they contained a specific color that was known in advance to be predictive of target location—it is possible that the early activity was associated with attentive processing of the cues rather than control of attention shifts to the cued locations . To evaluate this possibility , we performed a follow-up experiment in which informative and noninformative cues did not differ on the basis of a simple feature . Letters were used to cue attention to the left , upper-middle , and right locations ( L , U , and R , respectively ) as well as for the non-informative cue ( X ) . The results were almost identical to those obtained in the first experiment with the exception that no early occipital activity was observed ( Figure 3B ) . This shows that the early occipital activity seen in the main experiment reflected attentional processing of the cue but that the early parietal activity reflected control of attentional shifts to the cued location . To determine whether the activations in occipital , parietal , and frontal regions led to modulation of perceptual processing of the subsequent target , we examined correlations between the activation magnitudes and the attention effects on target discrimination accuracy ( Figure 5 ) . All peak activations in the ROI time-courses correlated significantly with performance ( rs > 0 . 78 ) , except the initial activation in occipital cortex ( Table 1 ) . The lack of significant correlation with early occipital activity bolsters the conclusion that the early occipital activity reflected attentional processing of the informative cue itself . The significant correlations only at the peaks of activity in the ROIs provide compelling evidence that the early parietal activations as well as the later frontal , parietal , and occipital activations reflect attentional control operations that enhance processing of the impending visual target . Taken together , these peak activations accounted for 93% of the variability in attention effects on target discrimination accuracy ( R = 0 . 97; R2 = 0 . 93; p < 0 . 006 ) . That is , the net activity within the attention-control areas identified here strongly predicts the level of attentional improvement in visual processing across participants .
The present study used a recently developed technique for localizing the neural sources of scalp-recorded EEG to investigate the time course of brain activity associated with voluntary control of visuospatial shifts of attention . Although converging lines of evidence have pointed to the involvement of the frontal and parietal lobes in attentional control , the sequence of activity within the fronto-parietal control network has remained unclear due to the poor temporal resolution of fMRI and the limitations of ERP dipole source modeling . A number of alternatives have been proposed , including an entirely top-down system wherein shifts of attention are initiated by executive control regions of the frontal cortex [4] and a system wherein shifts of attention are initiated by activity in posterior brain regions that precedes frontal lobe activity [6] . Recent findings have provided support for the top-down model proposing that the frontal lobes initiate the sequence of attentional operations involved in the voluntary control of visuospatial attention shifts ( Figure 1A ) . Our results , however , did not support this model . Instead , attentional-control activities in the parietal lobes were found to precede activity in the frontal lobes , which demonstrates that voluntary attentional control is not initiated solely by frontal cortex . Given that IPL was active twice and SPL was active only early on , the two regions appear to mediate different attentional-control operations . The combined early activity in parietal cortex likely reflects a signal to switch attention to a specific location that is sent to executive control structures in frontal cortex . Recent neuroimaging studies indicate that activity in SPL is associated with shifting attention in spatial [40] and nonspatial [41] visual tasks , as well as in auditory and audiovisual tasks [42 , 43] . On this basis , we believe that SPL supplies the initial signal to switch attention , whereas IPL supplies spatial information about the to-be-attended location . The spatially nonspecific ( bilateral ) activation of SPL coupled with the spatially specific ( predominantly ipsilateral ) activation of IPL early in the cue-target interval following shift-left and shift-right cues supports this interpretation . The late activity in IPL may reflect operations involved in the marking of the to-be-attended location [9] or the actual deployment of attention to that location [11] . The late IPL activity was not sustained until target onset; thus , it is unlikely to reflect operations involved in maintenance of attention at the cued location . The late activity in occipital and inferior temporal cortices began after the second phase of activity in IPL and was sustained until target onset . These areas are part of a ventral visual pathway that is involved in object processing and recognition [44] . Thus , the late occipito-temporal activity likely reflects anticipatory modulation of neuronal excitability in brain areas that would be responsible for processing sensory features of the upcoming target [45 , 46] . Following cues to shift attention to the nonlateralized location above fixation , attentional control activities in frontal and parietal areas as well as subsequent pre-target biasing in occipital cortex ( relative to the noninformative cue ) were largely bilateral . In contrast , attentional control activities in occipital , inferior parietal , and frontal cortices were lateralized following cues to shift attention to the left or right side of fixation . The spatially specific nature of the lateralized attentional control activity and subsequent pre-target biasing is in line with the lateralized organization of the primary visual pathways and is consistent with the observation of lateralized activity in ERP and fMRI studies examining activity following leftward and rightward-directing cues [4–6 , 11 , 14] . Increases in theta band activity were seen predominantly in cortical regions on the same side as the cued location , which suggests that this activity may be more closely associated with the anticipatory suppression of the to-be-ignored locations than the anticipatory enhancement of the to-be-attended location . The suppression of uncued locations has previously been linked to alpha band activity in this type of spatial cueing task [34] . The current results suggest that theta band activity also plays a role in the suppression of irrelevant information in order to maximize the attentional benefits for perception . Our main finding—that voluntary attentional control is initiated in parietal cortex—is inconsistent with data from a recent combined ERP-fMRI study that reported initial activity in frontal cortex [4] . This discrepancy may be due to differences in the methods used to model brain activity . The electrical neuroimaging approach employed here used a spatial filtering technique unconstrained by any previous results or a priori hypotheses about the number of activated brain regions or the locations of the activated regions , whereas the conventional ERP-fMRI approach models electrical activity with a few discrete ( dipolar ) sources constrained to be at locations of fMRI activations . In addition , the beamformer technique enabled us to reconstruct the distributed neural sources of all oscillatory activity in the theta band , rather than just the evoked activity that is observed in the ERP . By comparison , the combined ERP-fMRI method faces the potential problem that induced changes in post-synaptic neural potentials are not seen in ERP waveforms ( because they are not precisely phase-locked to events ) but are likely associated with changes in hemodynamic responses . Such differences between the physiological contributions to ERP and fMRI signals may lead to errors in estimating the locations of ERP sources , which would , in turn , lead to errors in estimating the timing of ERP source activities . Unfortunately , the combined ERP-fMRI approach also eliminates the opportunity to use the fMRI data to evaluate the validity of the ERP source model , because data from the two methods are integrated . We hypothesized that event-related changes in low-frequency theta band EEG oscillations would enable us to examine the spatial and temporal characteristics of activity in the voluntary attentional control network without any bias from previous fMRI results . To facilitate comparison of the present results with the results of recent fMRI studies of voluntary attentional control , we summarized the cortical sites of theta band activity across the entire cue-target interval in one image along with loci of fMRI activations [7–15] . This image , shown in Figure 6 , reveals clusters of activations in occipital , parietal , and frontal regions of cortex . Although our use of standard head models , MRIs , and electrode positions likely limit our accuracy in identifying precisely the regions where attentional control activity took place , the loci of the frontal , parietal , and occipital theta band sources dovetail nicely with the foci observed in previous fMRI studies . In light of this converging evidence , it is clear that these frontal , parietal , and occipital regions play important roles in the control of spatial attention . In addition , these results provide evidence for a link between low-frequency theta band oscillations and attentional processes that heretofore has not been explored in the literature . Our results show a clear link between low-frequency theta oscillations and attention . Prior studies have linked event-related changes in alpha and gamma band oscillations to attention and perception [32–34] , but to date , theta band activity has been most closely associated with learning and memory [35–37] . Our focus on theta band activity was motivated by the hypothesis that theta band oscillations are critical for long-range communications between distant brain regions [35] . From this view , any cognitive operation that requires communication between distant brain regions should involve changes in theta-band activity . However , these low-frequency oscillations overlap in space and time with oscillations in many other frequency bands and are even coupled with high-frequency oscillations ( e . g . , high gamma [36] ) . Thus , it is unlikely that activity in any particular frequency band—such as theta , alpha , or gamma—is fully responsible for the many different attentional control operations performed by the fronto-parietal network . Other frequency bands may show different sequences of activities ( i . e . , frontal activity preceding parietal ) , and it remains to be seen how sequences of activity in different frequency bands relate to different attentional control processes . It is possible that event-related changes in specific frequency bands relate to specific attentional control operations performed by a given brain region , but it is also possible that the dynamics of attentional control activity across the cortex are more closely linked to coupling between different frequency bands ( e . g . , between theta and high gamma ) . The electrical neuroimaging data provided here show that attentional control operations that follow the appearance of a symbolic spatial cue involve not just top-down signaling from frontal cortex but also an initial signaling from parietal cortex to indicate the need for an attention shift ( Figure 1B ) . Moreover , the magnitude of the early parietal activity accurately predicted behavior on the subsequent perceptual task , indicating the importance of this early activity for accurate target identification . While it is possible that the attention system may be flexible and display different sequences of parietal and frontal activations with varying task requirements , it is clear that models of top-down control that posit a one-way passage of information from frontal to parietal cortex are insufficient to explain the complexities of voluntary attentional control .
|
To extract important details about objects in the environment , people must focus their attention on a specific location in space at any given moment . Research using functional magnetic resonance imaging ( fMRI ) has suggested that regions of the frontal and parietal lobes work together to control our ability to direct attention to a specific location in space in preparation for an expected visual object . However , the sluggishness of the hemodynamic response has made it difficult to obtain information from fMRI about the timing of activity . Electroencephalography ( EEG ) has provided information about the timing of neural activity , but the limitations of traditional source estimation techniques have made it difficult to obtain information about the precise location in the brain that the EEG signals are coming from . Thus , the sequence of activities within this frontal-parietal network remains unclear . We used a recently developed electrical neuroimaging technique—called beamforming—to localize the neural generators of low-frequency electroencephalographic ( EEG ) signals , which enabled us to determine both the location and temporal sequence of activations in the brain during shifts of visuospatial attention . Our results indicate that low-frequency signals in parietal cortex provide the initial signal to shift attention .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion"
] |
[
"neuroscience"
] |
2008
|
Electrical Neuroimaging Reveals Timing of Attentional Control Activity in Human Brain
|
The bivalent killed oral cholera vaccine ( OCV ) provides 65% cumulative protection over five years . It remains unknown whether a boosting regimen can maintain protection in previously immunized populations . This study examines the immunogenicity and safety of an OCV regimen given five years following initial dosing . An open label controlled trial was conducted in 426 healthy Indian participants previously enrolled in a large efficacy trial . To assess whether an OCV regimen given after five years can elicit an antibody response equal to that of a primary series , we compared vibriocidal antibody titers in previously immunized participants receiving a two dose booster regimen to participants receiving a primary two dose immunization series . Among participants receiving a two dose primary series of OCV ( n = 186 ) , 69% ( 95% CI 62%-76% ) seroconverted . In the intervention arm ( n = 184 ) , 66% ( 95% CI 59%-73% ) seroconverted following a two dose boosting schedule given five years following the initial series . Following a single boosting dose , 71% ( 95% CI 64%-77% ) seroconverted . Children demonstrated 79% ( 95% CI 69%-86% ) and 82% ( 95% CI 73%-88% ) seroconversion after primary and boosting regimens , respectively . Administration of an OCV boosting regimen elicits an immune response similar to those receiving a primary series in endemic areas . Though a single boosting dose induces a strong immune response , further investigations are needed to measure if these findings translate to clinical protection .
Recent outbreaks in Haiti , Pakistan , and throughout the African continent , along with increased antimicrobial resistance and the heightening awareness of climate’s role upon the global burden have contributed to renewed interest in global cholera control . Though improved water and sanitation should continue to be the mainstays of cholera-prevention efforts , major improvements are a far off goal for much of the cholera-affected developing world . The notion that cholera epidemics are short lived are refuted by the fact that outbreaks have become more frequent , larger , and longer lasting , with case fatality rates higher than four percent [1] . Many countries with endemic disease either neglect or are unable to report cases greatly due to fears of the potential impact on their economy . With about 1 . 4 billion people at risk for cholera , an estimated 2 . 8 million cases , and 91 , 000 deaths occurring annually , common annual incidence estimates by the World Health Organization are likely conservative [2] . The disease has become more complicated in this pandemic since the emergence of the current O1 variant El Tor biotype due to concerns of heightened virulence [3] . These new organisms are better at surviving and more likely to result in asymptomatic carriage , meaning that infection may be introduced easier into a new area unknowingly , and once present , that area may well become a new cholera endemic zone [4] . Interest in oral cholera vaccine ( OCV ) has increased following demonstration of protective immunity via local , mucosally secreted intestinal antibodies [5] . A large cluster randomized , double blind , placebo controlled trial was conducted in the cholera endemic urban slums of Kolkata , India in late 2006 to evaluate the protection offered by the killed bivalent OCV . Though vibriocidal titers wane by one year after dosing , a cumulative vaccine protective efficacy of 65% has been measured over five years [6 , 7] . Vaccine efficacy was much lower in children one to five years of age ( 42% ) . Though not completely understood , reasons for this finding may include interference with pre-existing maternal antibodies , underlying co-existing enteric infections , mucosal damage following enteropathy , and malnutrition [8] . However , it bears special mention that more cases were prevented by vaccination ( 10 . 5/1000 ) in the younger age group ( 1–5 years ) , compared with older age groups ( 5 . 5/1000 in 5–15 years and 3 . 1/1000 in ≥15 years ) . Significant protection of unvaccinated individuals has been demonstrated in areas of modest vaccine coverage [9] . Mathematical models based on these data suggest when vaccinating over half of the population in cholera endemic areas , incidence can be reduced by 93% due to the vaccine’s ability to induce herd protection [10] . The WHO recommends that immunization with the safe , efficacious and affordable oral cholera vaccines should be used in conjunction with other prevention and control strategies in endemic areas with a potential role in outbreak situations [11] . As final analysis of the five year efficacy results were underway , questions on how best to deploy this vaccine were being raised by national and global policy makers . Based on the currently available efficacy data , Shanchol provides five years of clinical protection to adults in an endemic region [7] . Though no official recommendations on booster regimens are in place , redosing of the related Dukoral ( killed whole cell OCV with recombinant B subunit cholera toxin ) is recommended every 2 years in adults and every 6 months in young children . The aim of this trial was to measure the immunogenicity and safety of a boosting regimen five years after initial dosing in adults and children .
This is a nested , open-label controlled trial of Shanchol , an oral cholera vaccine , conducted among healthy non-pregnant subjects aged 6–14 years and ≥15 years who were initially immunized with two doses of vaccine or placebo five years before as part of a large randomized controlled trial ( RCT ) of an oral cholera vaccine . Because participants of the original RCT were aged ≥1 year of age , the minimal age of participants in this trial five years later was 6 years of age . All participants who were in the placebo arm of the phase III efficacy trial were scheduled to receive two doses of vaccine at the end of the trial period . Since unblinding was performed to conduct analysis and identify all individuals who needed to be given the vaccine , an open label trial design was employed in this study . The modified killed bivalent whole cell vaccine contained 1 . 5 x 1011 inactivated V . cholerae O1 and 5 x 1010 V . cholerae O139 bacteria consisting of: 600 ELISA units of of V . cholerae O1 El Tor Inaba; 300 ELISA units of multiple strains of V . cholerae O1 classical Ogawa , and 600 ELISA units of V . cholerae O139 . Two doses of vaccine were given 2 weeks apart from May 9 to June 11 , 2012 . Additional details on the study site , study agents , and trial conduct for the randomized controlled trial ( RCT ) in Kolkata have been reported previously [12 , 13] . Participants were enrolled from this cohort and all follow up study activities took place at one of nine area health centers within the census area . Of the preselected trial cohort , exclusion criteria included those < six years of age , pregnant women ( identified by verbal screening of married women ) , and individuals too weak to get out of bed , and anyone who had received vaccine following its licensure in 2009 . Endpoints were compared between two intervention groups: a boosted population ( individuals who received vaccine five years prior and were redosed ) and a primary series population ( participants who were placebo recipients in the original RCT and were receiving vaccine for the first time . Both of these groups received vaccine at days 0 and 14 and blood were drawn for measurement of vibriocidal titers . A third blood sample was also drawn on day 28 to compare baseline with titers 14 days following doses one and two ( Fig 1 ) . A documented follow up with a health care provider was performed on day 42 ( 28 days after the second dose in both intervention groups to monitor and document any adverse or serious adverse events . A small non-intervention arm , who did not receive vaccine , was added to ensure that boosting was not due to natural exposure . All of these individuals were given two doses of the vaccine following the final bleed of this study . Though the primary objective was to determine if a two dose OCV booster dose regimen administered to a previously immunized cohort elicits a similar immune response to those achieved by a primary immunization series , we also measured responses following a one dose booster . Seroconversion was defined as ≥ four-fold rise in serum vibriocidal titers measured at baseline ( day 0 ) and 14 days following each dosing schedule . The previously described microtiter technique was used to detect serum vibriocidal antibodies to V . cholerae O1 El Tor Inaba , O1 Ogawa , and O139 [14] . The trial protocol was approved by the Scientific Advisory Committee and Institutional Ethics committee of the National Institute of Cholera and Enteric Diseases ( NICED ) and the International Vaccine Institute ( IVI ) . Written informed consent was obtained from residents aged 18 years and older and from the parents/guardians of residents aged 1–17 years . Written assent was also obtained from residents aged 12–17 years . The trial was registered with Clinical Trial Registry-India ( CTRI/2012/04/002574 ) and ClinicalTrials . gov ( NCT01579448 ) . The funding agencies of the study had no role in study design , data collection , data analysis , data interpretation , or writing of the report Potential enrollees in each arm were identified before-hand by the IVI biostatistics team . All eligible individuals consist of those who received two doses of study agent during the RCT in September 2006 . Furthermore , eligible subjects were verified to be present in the census conducted in the fifth year of the efficacy trial . Any participant diagnosed with cholera during the five year surveillance period was not included . Participants were stratified according to the size of each of the nine health center catchment areas . Randomly generated lists for each stratum were made , from which community health workers contacted and approached potential participants during times of community sensitization of this new project . Sample size calculations were based on the non-inferiority of the primary intervention ( re-immunization with 2 booster doses in previously immunized ) compared to two doses of vaccine in an unimmunized population . Using 80% power , 65% immune response , a non-inferiority limit of 20% , and 10% dropout rate , we would require 194 participants ( 97 children ages 6–14 and 97 adults ages ≥15 ) in each of the two arms . To ensure that natural infection with V . cholerae did not affect serum antibody responses , a smaller non-intervention arm was recruited . Assuming an 80% power , a 5% immune response , and 10% drop out , to accept >20% immune response in intervention arm , we would require 38 participants ( 18 children ages 6–14 and 17 adults ages ≥15 ) in the non-intervention arm . Demonstration of a fourfold or greater rise in serum anti-O1 vibriocidal antibody titer following the second dose was the primary endpoint of immunogenicity . Comparison of the primary endpoint , vibriocidal seroconversion in boosting and primary series arms were evaluated with one-tailed 97 . 5% confidence interval using the Wilson Score method [15] . The dichotomous variables were compared using the chi-square test or by the Fisher’s exact test if a predicted cell count is less than five . For dimensional variables such as vibriocidal titers , Student’s t-test or Satterthwaite method depending on the heterogeneity of variance were used . Vibriocidal titers and fold rises were logarithmically transformed prior to statistical analyses . Statistical significance threshold of all comparisons was set at p<0 . 05 and two-tailed . All statistical analyses were done with SAS version 9 . 3 .
Recruitment and follow up was conducted in May-June 2012 and participant flow is illustrated in Fig 2 . Among eligible participants in the intervention groups , 184/197 ( 97% ) enrollees of the booster arm and 186/196 ( 95% ) of the primary immunization arm took both doses and provided all three blood samples . A total of 27 participants ( 6% ) were lost to follow up or considered ineligible following screening . There were no major differences between groups with regards to key demographic indicators ( Table 1 ) . A per protocol analysis was conducted for all immunogenicity data . Baseline geometric mean titers against V . cholerae O1 Inaba ( GMT , 95% CI ) ranged from 103 to 183 in the previously immunized group compared to a range 70 to 125 in those receiving OCV as a primary series . Baseline titers to O1 Inaba were higher in the previously immunized groups but the difference was not significant ( Table 2 ) . Geometric fold rise ( GFR ) following two doses ranged from 6 to 10 in both intervention groups . Seroconversion rates to O1 Inaba were 66% [95% CI 59%-73%] and 69% [95% CI 62%-76%] following two dose regimens in the boosting and primary immunization arms ( p = 0 . 53 ) , indicating that the boosting arm was non-inferior to the primary immunization arm . Seroconversion of 71% following a one dose regimen in the boosting arm was non-inferior to 69% after a two dose regimen in the primary immunization arm . When comparing the immune responses to O1 Inaba , there was no significant difference in the geometric mean fold rise or the percentage who seroconverted in the boosting and primary arms following two doses of OCV ( 82% [95% CI 73%-88%] vs 79% [95% CI 69%-86%] in the 6–14 year age group and 51% [95% CI 40%-61%] vs 60% [95% CI 50%-70%] in ≥15 years age group ) . While the GMT of the boosting arm in children was double that of the primary series after the first dose , the GMFr was also found to be high in both arms . In contrast , adults demonstrated similar GMT in both arms . Because of the vaccine’s lower efficacy in the youngest group , analysis was further subdivided into a 6–10 year group ( 1–5 year age cohort in the previous efficacy trial ) [S1 Table] . Similar to the 6–14 year age group , the 6–10 year group demonstrated similar baseline GMT in boosting and primary groups , with significantly higher GMTs following the first ( 2642 vs 931 , p = 0 . 01 ) and second dose ( 1201 vs 534 , p = 0 . 02 ) against V . cholerae O1 Inaba . Likewise , both groups also showed no significant differences in geometric fold rises or percent serconversion following the first or second dose . Furthermore , seroconversion rates to V . cholerae O1 Inaba following a single booster dose ( 85% in 6–14 years , 57% in ≥15 years ) were comparable to those following a two dose primary series ( 79% in 6–14 years , 60% in ≥15 years ) . While seroconversion rates following one and two booster doses ( 57% and 51% ) are not significantly different from one and two doses of a primary series ( 70% and 60% ) in volunteers >15 years of age , the results regarding non-inferiority are inconclusive since the 95% CI of the proportion difference includes the clinical margin ( -20% ) and zero . There was no significant rise in seroconversion rates in the non-intervention arm ( 11% in 6–14 years , 0% in ≥15 years ) , supporting the observation that environmental exposure was not a major determinant in seroconversion , as compared to vaccine in accounting for the rises in seroconversion rates in the intervention arms . Vibriocidal responses to O1 Ogawa demonstrated no significant difference in the percentage seroconversion in the boosting or primary arms following two doses of OCV ( 66% [95% CI 55%-74%] vs 72% [95% CI 62%-80%] in the 6–14 year age group and 41% [95% CI 31%-51%] vs 53% [95% CI 43%-63%] in ≥15 years age group ) . Though seroconversion rates for V . cholerae O1 Ogawa were significantly higher in the primary series after one dose ( 73% v 63% p 0 . 03 ) , this could be explained by the significantly higher baseline GMTs observed between the two intervention groups . No significant difference was noted when comparing the boosting regimen to the two dose primary ( 53% vs 62% p = 0 . 08 ) ( Table 3 ) . Because each study population arm differed in their previous antigenic exposure , we also analyzed vibriocidal response based upon median baseline geometric mean titer for each age group . Geometric mean fold rise and seroconversion rates following two doses did not significantly vary between boosting and primary arms measuring response against V . cholerae O1 Inaba and O1 Ogawa ( Table 4 ) . Our analysis notes the recurrent observation of lower immune responses to O139 ( S2 Table ) , likely representing that Shanchol does not provide meaningful sero-responses against O139 , at least as measured by vibriocidal responses . A total of six adverse events were recorded in the boosting arm ( fever , diarrhea , abdominal pain , vertigo ) and seven adverse events in the primary series arm within three days of either dose ( fever , diarrhea , vertigo , body ache ) . All were mild and resolved with symptomatic treatment . No serious adverse events were reported within 28 days of dosing .
The data suggests that repeating the immunization series to an endemic population previously immunized five years prior can induce a strong vibriocidal response , meeting those of similar individuals receiving the vaccine series for the first time . Additionally , a single booster dose achieves levels as high as a two dose OCV primary series . Before summarizing the findings of this study , study limitations should be considered . We did not measure mucosal antibody responses . These are important indicators of immunity that would provide a broader understanding of vaccine response . Still , vibriocidal antibodies are thought to provide a surrogate of protection and add to our understanding of vaccine-induced immunity . Our data provides information on immune responses and adverse events following immunization . These results do not provide information on vaccine efficacy following a boost vs . primary vaccination . Observational studies evaluating one or two dose regimens at five year would , however , provide estimates of booster effectiveness . Finally , our study was conducted in an area were V . cholerae exposure is frequent . Our results may not correlate with immune responses in areas where cholera incidence is lower . It is important to consider unique aspects of local immune responses from the gut subsequent to oral immunization . Mucosally induced antibody secreting cells ( ASCs ) have been shown to migrate into the peripheral circulation , where responses in naturally primed individuals are appreciably quicker than non-primed subjects [16] . Investigators in Sweden have reported substantial ASC responses after a single booster dose over 10 years later , which were highest at 4–5 days and followed by a rapid decline [17] . Much like vibriocidal antibody responses , a second dose of OCV given on day 14 did not boost ASC in Bangladesh [18] , whereas OCV given at the same schedule in a non-endemic areas [19] . These findings support the presence of a long term memory response that may support an extended gap when considering recommended intervals for a boosting regimen . Furthermore , boosting strategies likely differ between areas where V . cholerae infection is historically endemic and those where it is not , suggesting that one dose of OCV may offer some degree of immediate protection in primed populations living in cholera endemic areas . Though vibriocidal antibody response reflects an indirect correlate of protection , the public health implications of these findings could provide a basis to improve implementation of delivering OCV in resource constrained settings . When factoring these results in combination with recent studies expressing longer duration of efficacy [7] and the possibility of a flexible dosing regimen[20] , this boosting data may extend the true ‘benefit horizon’ of this affordable and feasible vaccine . These findings have particular relevance for endemic countries , for which longstanding protection is aided by natural boosting via regular environmental exposure . In light of limited oral cholera vaccine supply , the possibility of a one dose booster regimen would increase the number of individuals for which vaccine is available in endemic areas . If an OCV can provide long lasting clinical protection , capitalizing on the ease of delivery and immunological advantages of using the mucosal port of entry , a boosting regimen at five years in endemic populations could serve to trigger an immune response . Because baseline vibriocidal antibodies and memory to the cholera pathogen already exist in endemic populations , a boosting regimen could raise antibody production , potentially offering protection to a naturally primed population . Pre-existing immunity plays a critical role in understanding the host defense of each unique host population . Investigations on memory B cell and cell-mediated immune responses are lacking in children , and such studies would be interesting to offer important insights into differences in protection offered by natural infection versus current vaccine options [21] . However , our understanding of this phenomenon is limited in less endemic populations , and should be one of the priorities for future OCV field evaluations . As demonstrated in this study and previous trials , the antibody response after the first of two doses are higher than after the second dose [22] , implying that an immune response may begin even before the second dose is administered . Though vibriocidal antibodies have been shown to wane one year following dosing , clinical protection has been maintained for five years in an endemic setting [6 , 7] . Part of this vaccine’s success may have been attributable to natural boosting in a highly endemic context , as in the urban slum populations of Kolkata . Recurring cholera exposures can lead to a progressive age related acquisition of natural immunity due to environmental boosting and memory B cell mediated anamnestic responses [21] . This trial demonstrates that a two dose boosting OCV boosting regimen results in a robust immune response . Because it stimulates vibriocidal titers comparable to those achieved in residents receiving a full primary series of OCV , this data could serve as the base for future investigations examining clinical protection offered by an OCV boosting regimen . A shorter interval may likely be considered in children under five years of age due the lower cumulative protective efficacy ( 42% ) in this age group , Complementary efforts to strengthen effective surveillance are vital in order to accurately assess the impact of any new dosing strategy . With proper disease detection programs in place and supporting epidemiologic data , this evidence could support the initiation of a boosting regimen policy recommendation . Large cholera outbreaks continue to threaten marginalized populations affected by natural disasters or those displaced by war , where there is inadequate sewage disposal and contaminated water . It remains a major international public health priority and is a risk to most developing countries . Following introduction of V . cholerae to previously non-endemic low income countries with weak water and sanitation infrastructure , the delineation of endemic and epidemic is becoming less defined . With the recent support to build a 20 million dose OCV stockpile by 2018 [23] , appropriate boosting strategies need to be considered by policy makers now to successfully prevent future recurrences .
|
The five year efficacy results of the bivalent , killed whole cell oral cholera vaccine ( WC OCV ) was shown to offer 65% protection in cholera endemic Kolkata . Further search strategies focused on natural boosting of immunity , since this trial assessed protection in a population that has endemic cholera at high rates every year . The efficacy demonstrated in this project reflected both vaccine and naturally induced immunity . Though efficacy is maintained for five years , no formal recommendations on a boosting regimen exist . This study suggests that a boosting regimen of killed OCV can elicit vibriocidal titers similar to those levels produced by a primary series in adults and children residing in endemic areas . A boosting recommendation could help to ease logistical challenges faced in maintaining protection in cholera endemic areas . These immunogenicity findings provide initial evidence to support the use of an OCV boosting regimen five years following the primary series , with consideration of a shorter interval for children under the age of 5 years due to a lower observed efficacy in field trials .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
An Open Label Non-inferiority Trial Assessing Vibriocidal Response of a Killed Bivalent Oral Cholera Vaccine Regimen following a Five Year Interval in Kolkata, India
|
Copy number variation ( CNV ) is rife in eukaryotic genomes and has been implicated in many human disorders , particularly cancer , in which CNV promotes both tumorigenesis and chemotherapy resistance . CNVs are considered random mutations but often arise through replication defects; transcription can interfere with replication fork progression and stability , leading to increased mutation rates at highly transcribed loci . Here we investigate whether inducible promoters can stimulate CNV to yield reproducible , environment-specific genetic changes . We propose a general mechanism for environmentally-stimulated CNV and validate this mechanism for the emergence of copper resistance in budding yeast . By analysing a large cohort of individual cells , we directly demonstrate that CNV of the copper-resistance gene CUP1 is stimulated by environmental copper . CNV stimulation accelerates the formation of novel alleles conferring enhanced copper resistance , such that copper exposure actively drives adaptation to copper-rich environments . Furthermore , quantification of CNV in individual cells reveals remarkable allele selectivity in the rate at which specific environments stimulate CNV . We define the key mechanistic elements underlying this selectivity , demonstrating that CNV is regulated by both promoter activity and acetylation of histone H3 lysine 56 ( H3K56ac ) and that H3K56ac is required for CUP1 CNV and efficient copper adaptation . Stimulated CNV is not limited to high-copy CUP1 repeat arrays , as we find that H3K56ac also regulates CNV in 3 copy arrays of CUP1 or SFA1 genes . The impact of transcription on DNA damage is well understood , but our research reveals that this apparently problematic association forms a pathway by which mutations can be directed to particular loci in particular environments and furthermore that this mutagenic process can be regulated through histone acetylation . Stimulated CNV therefore represents an unanticipated and remarkably controllable pathway facilitating organismal adaptation to new environments .
Copy number variation ( CNV ) is widespread in human populations , with 5%–10% of the human reference genome showing CNV between normal individuals [1–3] . CNV of protein-coding genes contributes to multiple disorders , and specific genetic syndromes have been directly attributed to CNV [4–6] . The pathological effects of CNV imply that gene copy number impacts gene expression , and we have recently shown that changing copy number can directly influence RNA processing [7] . However , CNV of protein coding genes is not always detrimental and can enhance cell growth , particularly in challenging environments . Evolution experiments in yeast give rise to novel CNVs that enhance growth under nutrient starvation , bestow drug resistance , and complement genetic defects [8–12] . CNVs in tumour cells also enhance proliferation , albeit at the expense of the host; for example , copy number amplification can drive tumour growth ( e . g . , of FGFR2 or CDK4 [13 , 14] ) or mediate drug resistance ( e . g . , of DHFR , KRAS or BRAF [15–17] ) . These yeast and human CNVs are examples of adaptive events in which the emergence of novel heritable alleles increases the reproductive fitness of the cell in the current environment . The emergence of a novel allele in a population requires extensive selection such that the phenotypic observation is removed from original mutation event by many generations , and therefore causal mechanisms remain uncertain for most adaptive mutations including CNV [18] . Neo-Darwinian theory invokes natural selection of randomly occurring mutations to explain adaptation; however , random mutations need not be accidental , as the induction of genome-wide mutation under stress has been well characterised in bacteria and also reported in yeast [19–21] . Furthermore , a handful of loci in eukaryotes undergo localised and controlled genetic changes , including the mammalian immunoglobulin loci ( widely reviewed , for example see [22–24] ) , as well as the budding yeast ribosomal DNA ( rDNA ) for which multiple CNV mechanisms have been described [25–27] . These loci are , however , highly specialised and their genetic changes are performed by locus-specific machinery; equivalent mechanisms acting genome-wide to induce beneficial genetic changes have not been convincingly demonstrated and present substantial theoretical difficulties [28–30] . The budding yeast rDNA has been used extensively as a model system for CNV . The rDNA consists of ~150 tandem copies of a 9 . 1-kb sequence encoding the ribosomal RNAs and undergoes frequent CNV [31] . rDNA recombination is initiated almost exclusively from a replication fork barrier ( RFB ) present in each rDNA copy ( Fig 1a ) . A single protein , Fob1 , defines the replication fork stalling site at the RFB [32 , 33] , and cleavage of these stalled forks is thought to initiate break-induced replication ( BIR ) , a homologous recombination ( HR ) process that mediates replication fork restart using a homologous sequence on the sister chromatid [34] . Because homologous sequences are present in each rDNA copy , nonallelic homologous recombination ( NAHR ) occurs readily during BIR , causing frequent CNV . rDNA amplification is partly controlled through transcription; recombination requires RNA Pol I transcription [35] , and NAHR is enhanced by expression of 2 noncoding RNAs ( ncRNAs ) , IGS1-F and IGS1-R [27] ( Fig 1a ) . IGS1-F is a stable ncRNA , whereas IGS1-R is a cryptic unstable transcript ( CUT ) , a class of noncoding RNA that is degraded instantly after transcription and is transcribed through the RFB [36 , 37] . Therefore , local transcription in the context of stalled replication forks at the RFB has a profound effect on rDNA CNV and is thought to cause the environmentally regulated rDNA amplification observed in cells with insufficient rDNA copy number [34 , 38 , 39] . Although the rDNA recombination machinery is locus specific , replication fork stalling is not unique to the RFB , and CNV often arises from replication defects [40–43] . Collisions between replication and transcription are known to be particularly mutagenic [44] , and highly transcribed loci are prone to mutation in general [44–47] . Furthermore , transcription in bacteria has been directly observed to cause replisome dissociation , and mutation rates are higher for bacterial genes oriented against the direction of replication [48 , 49] . This lead us to hypothesise that environmental nutrients and toxins that invoke strong transcriptional responses may promote mutations such as CNV at induced loci , effectively focusing mutations at genes that are important for growth in the presence of those nutrients or toxins , thereby accelerating the formation of novel alleles that confer increased fitness in the current environment . Here we demonstrate that CNV of high- and low-copy repeated sequences can be directly stimulated at induced loci in response to the environment , giving rise to novel , advantageous alleles at a rate far in excess of the basal mutation rate .
Replication fork stalling ( RFS ) occurs widely in the yeast genome and is generally mutagenic [42 , 50–52] . By analogy to the rDNA system , we hypothesised that CNV may be instigated from RFS sites upstream of inducible promoters when those promoters are induced ( Fig 1a and 1b ) . This would cause reproducible , environment-specific patterns of gene loss and gene amplification . RFS sites are marked by S139-phosphorylated histone H2A ( γH2A ) [42] . We used chromatin immunoprecipitation ( ChIP ) sequencing ( ChIPseq ) for γH2A to generate a high-resolution profile of RFS in the yeast genome , producing a map that is broadly in accord with published ChIP-microarray data [42] . This experiment showed that peaks of >2-fold γH2A enrichment occur within 1 kb upstream of ~7% of Saccharomyces cerevisiae genes , which we classed as γH2A genes ( S1 Table ) . We then performed a meta-analysis of published RNA sequencing ( RNAseq ) data , comparing steady-state mRNA levels of γH2A and non-γH2A genes; this revealed that γH2A genes are expressed at unusually low levels in yeast grown under optimal culture conditions ( 30°C in YPD ) ( Fig 2a and S1a Fig , compare grey and blue lines ) . However , these γH2A genes are significantly induced under more challenging conditions , such as respiratory growth and industrial fermentation ( Fig 2a and S1a Fig , red lines ) . γH2A genes are therefore biased towards those that are expressed primarily during growth in suboptimal conditions . However , these genes are not rapidly induced by osmotic or oxidative stress and are therefore not simply stress-response factors ( S1b Fig ) . If the induction of RFS genes can instigate CNV , these CNV events should be more frequent at genes induced in response to specific environmental conditions . To experimentally validate this prediction , we focused on 1 γH2A gene , CUP1 , a well-studied gene encoding a metallothionein that sequesters excess copper [53 , 54] . CUP1 occurs in a tandem array of 2-kb repeats and has widely varying copy numbers amongst different yeast strains , with higher copy numbers conferring enhanced resistance to copper toxicity [55 , 56] . The haploid strain BY4741 used here has a CUP1 copy number of 13 in our assays , in keeping with previously reported estimates of CUP1 copy number in the parental S288C background ( 10–15 copies ) ; this copy number is high but by no means exceptional compared to wild isolates [57 , 58] . As expected , most strains that we have tested from the BY4741-derived a mating–type deletion collection [59] also have 13 CUP1 copies , while the S288C-derived MEP diploid has 2 CUP1 alleles of 13 and 14 copies ( see “Stimulated CNV accelerates the acquisition of copper resistance” ) . CUP1 is strongly induced by environmental copper , and we performed a northern blot analysis to determine whether the CUP1 promoter is bidirectional like the rDNA ncRNA promoter , as promoter bidirectionality is important for rDNA CNV [27] . Bidirectional promoters are common in the yeast genome , but often the antisense RNA produced is an unstable CUT that is hard to detect in wild-type cells [36 , 60] . We therefore analysed RNA from a wild type and from an rrp6Δ mutant that lacks a key exonuclease activity required for CUT degradation [36] , revealing that the CUP1 promoter is bidirectional , transcribing a CUT through the RFS site in response to copper exposure ( Fig 2b and 2c ) . γH2A peaks upstream of the CUP1 ORFs are readily seen in our γH2A ChIPseq data , and the CUP1 RFS site is unaffected by growth in copper , in contrast to a previous report that ongoing transcription prevents RFS [42] ( Fig 2d ) . This combination of an inducible bidirectional promoter adjacent to an RFS site fits our model derived from the rDNA locus ( Fig 1b ) , making CUP1 an excellent candidate for stimulated CNV , particularly as CuSO4 induces only a handful of other γH2A genes ( S2 Fig ) . Copper exposure leads to the emergence of cells carrying amplified CUP1 alleles [61] , but proving that the environment actually stimulates CUP1 CNV requires the measurement of CNV in the absence of selective pressure . To achieve this , we reengineered the CUP1 repeat sequence , replacing the copper-responsive CUP1 promoter and ORF with a GAL1 promoter and a 3xHA tag ORF while leaving surrounding sequences , including the RFS region , intact ( Fig 2b ) . This modified construct expresses a nonfunctional protein in response to environmental galactose but not glucose , whereas the endogenous promoter is not induced by galactose . We inserted a construct containing 3 copies of this modified PGAL1-HA repeat in place of the CUP1 locus , which fortuitously amplified to 17 copies upon transformation . PGAL1-HA cells were then grown for 10 generations in glucose or galactose and compared to wild-type controls grown under the same conditions . Growth of the PGAL1-HA strain in galactose gave rise to multiple de novo CNV alleles , detected by Southern blot , whereas no change was observed in the wild-type controls ( Fig 2e ) . This demonstrates that promoter induction in the genetic context of CUP1 is sufficient to stimulate CNV . Equivalent experiments , however , on the wild-type CUP1 locus would not be informative because growth in the presence of copper selects for rare amplified alleles whether they arise through random or stimulated CNV . To determine whether copper stimulates CUP1 CNV requires the analysis of many individual cells that are allowed to replicate with or without copper while excluding cells born during the exposure period . Quantification of de novo CNV events within this defined cohort would provide a measure of CNV rate independent of selection . To achieve this , we employed the mother enrichment program ( MEP ) —a system that selectively renders new-born cells inviable in the presence of β-estradiol [62] . A cohort of MEP cells in the presence of β-estradiol can be treated for a given period with or without copper , after which time only cells in the initial cohort are viable ( Fig 3a ) . Cells from copper-treated and control cohorts are then plated in the absence of copper and estradiol , giving rise to colonies derived from single cells that have ( or have not ) been previously exposed to copper; these colonies inherit the CUP1 copy number of the progenitor cell . If copper stimulates heritable CNV at CUP1 , then a greater number of colonies with CUP1 alleles that deviate from the parental copy number should be detected in the copper-treated cohort . We divided 2 populations of β-estradiol–treated MEP diploid cells and grew them for 24 hours in the presence or absence of 1 mM CuSO4 , then assayed 184 of the resulting colonies for CUP1 copy number ( Fig 3b ) . We observed 31 CNV events ( including 6 amplifications ) in 56 copper-treated diploid cells ( 112 CUP1 alleles , 27% CNV events , 5% amplifications ) , compared to 7 CNV events ( including 1 amplification ) in 128 untreated cells ( 256 alleles , 3% CNV events , 0% amplifications ) . The difference in the number of CNV events and amplifications between copper-treated and untreated cells is significant ( p = 1 . 1x10-7 for CNV events and p = 0 . 028 for amplifications ) and represents a 9-fold stimulation of CNV by copper . Furthermore , based on bud scar counting , the untreated cells undergo more divisions than the copper-treated cells in 24 hours ( 12 ± 2 versus 8 ± 3 divisions ) , meaning that 9-fold is an underestimate of the true extent of CNV stimulation . This finding directly demonstrates that environmental copper stimulates CNV at CUP1 . Changes in CUP1 copy number alter copper resistance , and we therefore measured the ability of cells arising in this experiment bearing an amplified ( +3 ) or a contracted ( –7 ) allele to grow at different copper concentrations . As expected , copper resistance was significantly increased in the amplified clone and decreased in the contracted clone ( Fig 3c ) . This demonstrates that stimulated CNV gives rise to de novo alleles with quantifiable phenotypic differences , including increased copper resistance . To ensure that stimulated CNV is a specific result of promoter induction as opposed to a mutagenic effect of copper treatment , we created diploid MEP cells heterozygous for the wild-type CUP1 allele and the engineered galactose-responsive PGAL1-HA allele . As predicted , galactose treatment induced extensive CNV at the PGAL1-HA allele ( 96% of PGAL1-3HA alleles underwent CNV in galactose compared to 9% in glucose ) ( Fig 3d , left ) . In contrast , the copper-responsive wild-type allele in the same cells was unaffected ( 0% of alleles underwent CNV in galactose and only 2% in glucose ) ( Fig 3d , right ) . This confirms that CNV is not stimulated uniformly and is highly selective for a transcriptionally induced allele over a silent locus of similar sequence and copy number . Sir2 family histone deacetylases ( HDACs ) repress rDNA CNV at multiple levels , leading us to question whether HDACs also control CUP1 CNV [26 , 63] . Indeed , we observed extensive CUP1 CNV after treating wild-type cells with the Sir2 family inhibitor nicotinamide ( Fig 4a ) . Analysis of individual deletion mutants revealed that Sir2 itself has little impact on CNV at CUP1 , but loss of the degenerate H3K56 HDACs Hst3 and Hst4 induces extensive CNV , suggesting a critical role for H3K56ac in regulating CUP1 copy number ( S3a Fig ) . Consistent with this , loss of the H3K56 acetyltransferase Rtt109 rendered the CUP1 locus immune to nicotinamide ( Fig 4a ) . To determine the importance of Rtt109 for stimulated CNV , we repeated the MEP-based assay from Fig 3b in an rtt109Δ background . Remarkably , we found that the transcriptional stimulation of CNV in response to copper was completed abrogated by loss of Rtt109 ( compare Fig 4b [showing rtt109Δ cells] to Fig 3b [showing wild-type cells] ) . This shows that stimulated CNV acts by a defined mechanism involving H3K56ac . H3K56ac has been implicated in both CUP1 promoter induction [64] and replication fork stability or restart [65–67] . Nicotinamide may therefore affect the stimulated CNV mechanism in 2 ways: by inducing the CUP1 promoter or by destabilising the stalled replication fork . We observed that nicotinamide stimulates expression of the CUP1 antisense CUT but causes little or no change in the level of the CUP1 sense mRNA ( Fig 4c ) . This suggests that HDAC inhibition by nicotinamide reduces promoter directionality at CUP1 rather than inducing the promoter per se , as has been recently reported for other ( as yet unidentified ) HDACs [68] . This loss of directionality cannot be ascribed to H3K56 acetylation as it is also observed in rtt109Δ cells , and it must depend on another member of the nicotinamide-sensitive Sir2 family . Importantly , however , loss of promoter directionality cannot be solely responsible for CUP1 CNV , as an equivalent increase in CUP1 CUT transcript is observed in rtt109Δ cells in which CNV does not occur ( compare Fig 4a to 4c ) . Therefore , nicotinamide treatment makes the CUP1 promoter transcribe bidirectionally , but this effect is not Rtt109-dependent and is not the sole driver of CNV stimulation . To assess the potential impact of H3K56 acetylation–associated replication fork defects on CNV , we asked whether mutations that destabilise or impair the processing of stalled replication forks phenocopy nicotinamide without affecting CUP1 promoter induction or directionality . Amongst 11 deletion mutants of replication fork–associated proteins that impact rDNA stability , we observed that mrc1Δ and pol32Δ cells undergo striking CUP1 CNV without affecting the CUP1 promoter ( Fig 4d and S3b Fig ) . Mrc1 stabilises stalled replication forks [69] , while Pol32 is required for efficient DNA synthesis following the BIR events that are initiated from broken replication forks [70 , 71] . The high level of CUP1 CNV observed in both mutants is consistent with abnormally frequent or inefficient BIR being a key driver of CNV . Importantly , increased H3K56ac was recently shown to impair DNA synthesis during BIR , causing frequent replication fork stalling and recombination events [72] . Such additional recombination events occurring in a repetitive region should cause extensive CNV , providing a simple explanation for the induction of CUP1 CNV by nicotinamide , which increases H3K56ac globally through inhibition of Hst3 and Hst4 . To our surprise , however , nicotinamide had little effect on the PGAL1-HA allele , showing that a global increase in H3K56 acetylation is not sufficient to drive CNV ( Fig 5a ) . One difference between the wild-type CUP1 allele and the re-engineered PGAL1-HA allele is that the GAL1 promoter is fully repressed in glucose , and therefore nicotinamide treatment does not cause the expression of an antisense transcript ( S4 Fig ) . Given the dual effect of nicotinamide on CUP1 promoter directionality and post-BIR replication , we suspected that both activities might be required for efficient CNV induction . To test this , we grew PGAL1-HA cells with or without nicotinamide in a range of galactose concentrations known to induce the bidirectional GAL1 promoter to various extents [73] . As predicted , the effect of nicotinamide was minimal and not significant without promoter induction ( Fig 5b , lanes 9–12 ) , but with higher promoter induction , nicotinamide significantly stimulated CNV ( Fig 5b , lanes 1–8 ) . This effect was particularly striking for the formation of de novo alleles with 1–3 copies , which presumably arise through multiple sequential CNV events ( S5 Fig ) . This experiment reveals that promoter activity and H3K56ac make additive contributions to CNV . However , we observed that CNV becomes largely independent of nicotinamide at high galactose concentrations ( Fig 5a , lanes 3–4 ) , which is not consistent with the model proposed above whereby H3K56ac acts during BIR . We therefore tested the importance of H3K56ac in CNV induction from the GAL1 promoter by deleting RTT109 . Just as for the wild-type CUP1 locus , this completely abrogated CNV induction , confirming that H3K56ac is critical for stimulated CNV in the PGAL1-HA system ( Fig 5c ) . These data show that stimulated CNV requires transcription and H3K56ac , but for highly induced promoters the normal physiological level of H3K56ac is sufficient to support extensive CNV such that further deregulation of H3K56 HDAC activity has little effect . Direct detection of stimulated CNV is facilitated by the high copy number of the CUP1 locus . However , this raises the question of whether CNV stimulation is restricted to high-copy tandem repeat loci , which are rare amongst protein-coding genes . In contrast , copy numbers of 2–5 are very common in the yeast and human genome sequences [1 , 14 , 74–77] , and we asked whether a 3-copy CUP1 locus would show equivalent behaviour to the high-copy system . Individual CNV events are too rare in this 3-copy system for direct detection but , having defined the effects of modulating H3K56 acetylation on CNV , we reasoned that if stimulated CNV acts at low-copy loci then H3K56 modulation should alter the rate at which CUP1 amplifications emerge under copper selection in a predictable manner . To test this , we replaced the endogenous CUP1 locus with a synthetic construct containing 3 wild-type copies of the CUP1 repeat sequence while maintaining the reading frame of the overlapping RSC30 gene ( Fig 6a ) . Stimulated CNV is replication linked , effectively requiring cells to grow in a sublethal concentration of copper . We observed that 3xCUP1 cells grow slowly in 0 . 3 mM CuSO4 , although fast-growing resistant populations often emerge late in growth , showing that resistant cells are under positive selection ( S6a Fig ) . Importantly , when 3xCUP1 cells were grown for 10 generations in batch culture in the presence of 0 . 3 mM CuSO4 , copy-number-amplified alleles were almost always detected in the population by Southern blot ( S6b Fig ) , forming a quantitative assay for CNV . We first tested whether nicotinamide treatment stimulated CNV in 3xCUP1 cells as in the high-copy system . Nicotinamide largely stimulates contractions in high-copy CUP1 arrays , and in 3xCUP1 cells the only reproducible CNV event observed on nicotinamide treatment was a –1 contraction to 2xCUP1 ( Fig 6b , red arrow ) . In combination with 0 . 3mM CuSO4 , the 2xCUP1 band largely disappeared , as would be expected under copper selection ( Fig 6c , red arrow ) , and the proportion of amplified alleles increased marginally ( Fig 6c , upper quantification panel ) . However , the most noticeable difference was that the proportion of large alleles ( more than 2-fold the progenitor allele size ) increased dramatically with nicotinamide treatment ( Fig 6c , lower quantification panel ) . These results show that copper and nicotinamide both stimulate CNV , and although CNV stimulation causes many copy number contractions , copper and nicotinamide have an additive effect in the 3xCUP1 system that results in the formation of larger alleles that predominate in batch culture . In contrast , since deletion of RTT109 suppressed stimulated CUP1 CNV in the high-copy system , we then asked if the amplifications observed in 3xCUP1 cells are also Rtt109-dependent . Indeed , when 3xCUP1 rtt109Δ cells were grown for 10 generations in 0 . 3 mM copper , amplification was completely suppressed ( Fig 6d ) . This demonstrates that H3K56 acetylation is required for CUP1 amplification in the presence of copper , a very surprising result because previous studies have shown a critical role for Rtt109 in maintaining genome stability rather than promoting genome change [26 , 66 , 78 , 79] . We then asked whether another 3-copy RFS gene would show similar behaviour . We selected to test the SFA1 gene encoding a formaldehyde dehydrogenase , as this has a clear upstream RFS site ( S6c Fig ) , is inducible in response to formaldehyde , and higher SFA1 copy number increases formaldehyde resistance [80] . A tandem array of 3 SFA1 genes with surrounding sequence was inserted at the CUP1 locus along with a single wild-type CUP1 copy ( Fig 6a ) , while the endogenous SFA1 gene was deleted . These 3xSFA1 cells showed a sharp cutoff for growth in formaldehyde , with <0 . 9 mM allowing robust growth and >1 mM completely suppressing growth , while the formaldehyde concentration that gave slow but reproducible growth ( which is required for these assays ) varied from 0 . 9–1 . 0 mM with formaldehyde batch and had to be empirically determined . We therefore refer to the assay concentration as ~1 mM formaldehyde . Growth of 3xSFA1 cells with ~1 mM formaldehyde induced bidirectional transcription from the SFA1 promoter and again gave rise to amplified SFA1 alleles detectable by Southern blot over 17 generations ( Fig 6f and S6d Fig ) . As in the 3xCUP1 system , growth of 3xSFA1 cells in nicotinamide induced copy number contractions that were readily detected in the absence of formaldehyde ( Fig 6e , red arrows ) , although the additive effect between formaldehyde and nicotinamide was not observed . Importantly however , the copy number amplification of SFA1 in 3xSFA1 cells was completely suppressed in an rtt109Δ mutant ( Fig 6f ) , indicating that SFA1 amplification proceeds in the presence of formaldehyde by the same mechanism as CUP1 amplification . Confirming that CNV is transcriptionally stimulated at SFA1 requires a high-copy system , but we were unable to create a direct equivalent of the high-copy PGAL1-HA strain as this amplified fortuitously during transformation . The alternative is to select for an amplified allele using formaldehyde; however , this requires the SFA1 gene and amplification system to be active , which is not the case if the SFA1 promoter is simply replaced with PGAL1 . We suspected that bidirectional transcription into the SFA1 RFS would stimulate CNV irrespective of where the promoter is placed , so we created a construct in which the promoter and ORF of the upstream divergent gene UGX2 were replaced with PGAL1-GFP in each of the 3 SFA1 repeats ( S6e Fig ) . This strain was grown in 0 . 9 mM formaldehyde , stepped up to 2 . 2 mM formaldehyde and then recovered in glucose media , yielding a PGAL1-GFP-SFA1 strain with an unstable and therefore somewhat heterogeneous copy number but primarily containing 11 copies . Growth of this strain in galactose caused the disappearance of this upper band and the emergence of a prominent ladder ( Fig 6g ) , just as in the original PGAL-HA strain ( Fig 2e ) , showing that transcription directly stimulates CNV at the SFA1 RFS site . Together , these data show that low-copy systems undergo CNV through a mechanism consistent with stimulated CNV , and that this mechanism is not restricted to CUP1 . The ability of transcription to stimulate CNV and the reliance of this mechanism on H3K56 acetylation suggest that the emergence of copper adaptation , instead of being an inevitable consequence of random mutation , follows a defined mechanism that is highly sensitive to alterations in this histone mark . To assess this , we initially tested the copper resistance of the 3xCUP1 wild-type and rtt109Δ cells grown with or without 0 . 3 mM CuSO4 shown in Fig 6d . Growth of wild-type cells in copper caused a dramatic rise in copper resistance , with GI50 ( concentration of CuSO4 causing a 50% inhibition of growth ) rising >3-fold from 0 . 5 mM in untreated cells to ~1 . 7 mM in cells pregrown in 0 . 3 mM CuSO4 ( Fig 7a ) . This increase was clearly attributable to CNV , as cells that grew in 1 mM CuSO4 carried large CUP1 amplifications ( S7a Fig ) . In contrast , rtt109Δ cells underwent a far smaller increase , rising <2-fold from 0 . 5 mM to ~0 . 8 mM after pregrowth in 0 . 3 mM CuSO4 ( Fig 7a ) , and of the rtt109Δ cells that did survive at 1 mM CuSO4 , albeit with much reduced growth compared to wild-type cells , only a fraction had undergone CUP1 amplification , suggesting that most had acquired resistance through other , potentially random , mutations ( S7a Fig ) . This shows that acquisition of copper resistance , far from being inevitable , is strongly dependent on an Rtt109-dependent amplification mechanism . We similarly assessed the effect of nicotinamide on copper resistance using the cells grown with or without CuSO4 and with or without nicotinamide shown in Fig 6c . The additive effect of nicotinamide and copper provided a small and not significant increase to the already substantial copper resistance of cells pregrown in 0 . 3 mM CuSO4 , but more surprisingly , pregrowth in nicotinamide caused a 2 . 5-fold increase in GI50 for CuSO4 , from 0 . 3 mM to 0 . 75mM ( Fig 7b ) . This is in contrast to the Southern blotting data , which show that the primary effect of nicotinamide treatment is copy number loss ( Fig 6b and 6c ) and suggest that a substantial amount of amplifications are also generated . In support of this , the CUP1 copy number of nicotinamide pretreated cells that grew in 0 . 75 mM CuSO4 was substantially amplified ( S7b Fig ) , showing that nicotinamide treatment increases copper resistance by promoting CUP1 amplification . These striking effects of H3K56 acetylation on copper resistance suggested that CNV stimulation should provide substantial selective advantages at the population level . As nicotinamide treatment induced constitutive stimulated CNV in the absence of copper , we used this drug to directly test the selective benefit provided by stimulated CNV . Firstly , we examined strong copper selection based on growth in 0 . 75 mM CuSO4 , a concentration fully inhibitory to growth of 3xCUP1 cells . We pretreated four 3xCUP1 cultures for 10 generations with or without 5mM nicotinamide , then obtained 6 growth curves for cells from each culture with or without 0 . 75mM CuSO4 in the absence of nicotinamide ( S7c Fig ) . Maximum growth rates were derived from each growth curve to determine whether any cells had adapted sufficiently to allow as rapid growth in the presence of 0 . 75 mM CuSO4 as in the absence of copper ( Fig 7c ) . Nicotinamide pretreatment had no effect on growth rate in the absence of copper but dramatically increased adaptation to copper: only 1 of the 24 cultures grown in the presence of 0 . 75 mM Cu without nicotinamide pretreatment grew normally , whereas over half ( 13 of 24 ) cultures derived from the nicotinamide pretreated samples reached growth rates equivalent to cells growing in the absence of copper . This was not due to rare events in a few of the precultures because the distribution of growth rates obtained in samples from each pretreated culture was similar ( S7d Fig ) . Therefore , although CNV stimulation primarily causes copy number contraction , it also dramatically enhances the ability of a subpopulation of cells to thrive in otherwise toxic concentrations of copper . Secondly , we asked whether stimulated CNV provides a competitive advantage in low-copper environments in which nonamplified cells are still capable of growth . Under these conditions , the fact that stimulated CNV primarily causes copy number reduction ( and therefore further slows growth in copper-containing environments ) may put a population of cells using stimulated CNV at a disadvantage relative to a population that does not . To test this , we again made use of the nicotinamide to mimic the effect of stimulated CNV prior to growth in copper-containing media , and we directly competed 1:1 mixtures of untreated and nicotinamide pretreated populations in the same cultures , with or without 0 . 3 mM CuSO4 . Treated and untreated populations carried different selectable markers to allow the composition of the mixture to be determined by plating before and after growth ( Fig 7d , left ) . Nicotinamide pretreatment did not alter the competitive fitness of cells in the absence of copper , but the nicotinamide-treated populations efficiently outcompeted the untreated populations in the presence of 0 . 3 mM CuSO4 , increasing their population share by 40% on average over 10 generations ( Fig 7d right ) . These experiments clearly show that although stimulated CNV engenders many more contractions than amplifications , it still provides a major selective advantage in both low- and high-copper environments . Together , our findings demonstrate that bidirectional promoter induction in the CUP1 genetic context can stimulate CNV to form novel adaptive alleles and that the rate of stimulated CNV is responsive to a controllable histone modification system . Stimulated CNV provides a clear selective advantage , and amplified alleles conferring improved resistance arise in low-copy strains by a mechanism consistent with CNV stimulation .
We propose a model for stimulated CNV in which local bidirectional promoter activity destabilises stalled replication forks , increasing the frequency of error-prone BIR events . Replication fork stalling occurs widely though nonrandomly in the yeast genome , but stalled forks are normally resolved through error-free mechanisms that protect genome stability ( Fig 8a , top left ) ( reviewed in [81] ) . However , we suggest that induction of transcription from an adjacent bidirectional promoter increases the likelihood that stalled replication forks collapse ( Fig 8a , bottom left ) ; this may occur either through direct interference of the transcription machinery with the stalled fork or indirectly through increased topological stress . Either way , the collapsed fork must then be restarted by a mechanism such as BIR ( Fig 8a , bottom middle ) , forming a replication fork with reduced processivity that is prone to CNV , particularly when H3K56ac is high ( Fig 8a , right ) . This mechanism is closely related to the fork-stalling and template-switching ( FoSTeS ) process that is suggested to underlie a wide range of CNV events [82] , but with contributions from transcription and H3K56ac . Our data implicating H3K56ac and Pol32 in CUP1 CNV is consistent with restart by BIR , since the replication forks newly formed through BIR are error-prone in the absence of Pol32 or the presence of H3K56ac [70–72] . It is worth noting , however , that other error-prone replication fork restart mechanisms are known , and these may have similar dependencies [83 , 84] . This mechanism would be expected to yield both expansions and contractions , and we suggest that 2 additional factors drive the contraction bias we actually observe . Firstly , a fork restarted by HR within a high-copy sequence has a high chance of template switching through strand invasion of a homologous sequence in a different copy , whereas a unique sequence provides only a single homologous template and so template switching is disfavoured . Copy number amplification requires the fork to template switch backwards and re-replicate multiple copies ( Fig 8b , upper ) , whereas contraction requires a template switch forwards that moves the fork closer to the end of the high-copy sequence ( Fig 8b , lower ) . A fork that has switched backwards will therefore spend longer replicating a high-copy sequence than a fork that has switched forwards and would have a higher chance of template switching again , with further potential to generate a contraction ( Fig 8b , right ) . The scheme in Fig 8b shows that 2 successive template-switching events can result in a 3:1 ratio of contractions to amplifications . This factor would prevail in nicotinamide-treated cells in which H3K56ac is uniformly high , but in normal cells , H3K56ac is primarily on the newly synthesised histones behind the fork and so H3K56ac is also asymmetrical ( Fig 8c ) . This means that a template switch forwards would lead to BIR using a previously unreplicated template with low H3K56ac and would result in a high-processivity replication fork ( Fig 8c , upper ) . In contrast , a template switch backwards would result in the BIR fork using a previously replicated template with high H3K56ac , and this fork would have low processivity and a higher chance of template switching again ( Fig 8c , lower ) . Together , we believe that these 2 factors would yield a major bias towards contraction events but would not prevent amplifications occurring at a lower frequency . We initially hypothesised that CNV at the CUP1 locus would be mechanistically equivalent to CNV at the rDNA . Indeed , the requirement for bidirectional promoter induction and the effect of H3K56 HDACs are similar in the 2 systems . However , we see important differences: a lack of dependence on Sir2 , which is not surprising as Sir2 acts primarily at heterochromatin; and the suppression of CUP1 CNV in rtt109Δ cells , which conversely undergo massive rDNA amplification [25 , 26] . The mechanistic analysis of rDNA amplification in rtt109Δ mutants by the Kobayashi group [26] provides an explanation for this discrepancy: rDNA amplification in rtt109Δ mutants does not proceed via chromosomal BIR; instead , rolling circle amplification of extrachromosomal rDNA circles ( ERCs ) forms large arrays of additional rDNA copies that can reintegrate into the chromosomal rDNA locus [26] . CUP1 circles have been detected but are 3–4 orders of magnitude rarer than ERCs [85] , and we suspect that rolling circle amplification of these happens in too few cells to make a detectable contribution . Stimulated CNV controls the occurrence of a subset of mutations that allow adaptation to challenging environments . It is commonly assumed that adaptive mutations occur at random , and they are largely inevitable as they occur through multiple poorly defined mechanisms . Under this assumption , loss of many genome stability factors would increase the rate of adaptation , but adaptation cannot be suppressed . In contrast , we show that the adaptation of yeast to environmental copper by amplification of CUP1 is largely dependent on a defined pathway that requires Rtt109 , despite the general role of this protein in maintaining genome stability . Therefore , adaptation occurring through apparently random mutation may in fact be stimulated by a specific cellular mechanism . How widespread is stimulated CNV ? The CUP1 and SFA1 model systems we have analysed are multicopy , but although CNV will be most efficient where multiple homologous sequences surround the RFS site , this mechanism should not be restricted to multicopy loci . Recombination events triggered by error-prone replication forks could easily utilise distal homologous sequences or even microhomology as templates , inducing de novo CNVs and chromosomal translocations with limited homology at the breakpoints . Breakpoints in de novo CNVs would therefore be poorly defined because they initiate nearby but not at the RFS site and utilise unpredictable homologous sequences . Interestingly , CUP1 shows exactly this behaviour in different S . cerevisiae isolates , as the multicopy CUP1 repeat has emerged many times with different breakpoints [58] . Replication fork stalling is by no means restricted to budding yeast , and bidirectional promoters are the norm in organisms , including mammals [86 , 87] . Therefore , the basic machinery required for stimulated CNV is likely to be conserved . Furthermore , the histone deacetylases that regulate CNV outcome are conserved in mammals and appear to have similar functions in modulating DNA repair [88 , 89] . Stimulated CNV in somatic cells of metazoans is rarely likely to be a useful organismal outcome and cannot aid heritable adaptation . However , because stimulated CNV emerges from conserved features of the replication and transcription systems , it seems likely that it would be active in mammalian cells , providing a mechanism that could be readily exploited , for example , by tumour cells . The mechanism that we have proposed is also very consistent with recent reports of nonrandom double-strand breaks formed by neurons in genes important for neuronal function [90 , 91] . As such the stimulated CNV pathway provides a new set of targets by which pharmaceuticals may prevent the emergence of undesirable properties such as drug resistance in tumours or modulate natural genetic changes in particular cell types . Indeed , our observation that adaptation of yeast to copper can be effectively suppressed by removal of Rtt109 , a protein for which chemical inhibitors have been described [92 , 93] , provides good evidence that the emergence of resistance is pharmacologically accessible . Evidence for adaptation through genome-wide nonrandom mutation is substantial , particularly in bacteria [18] , but the ability of stimulated CNV to direct mutations to relevant loci must be reconciled with forceful arguments against previously proposed ‘directed mutation’ systems [18 , 30 , 94 , 95] . The primary issue is that any general mechanism that directs mutations to a particular site must ‘know’ in advance the fitness outcome of a particular genetic change , which is not possible except at singular , highly specialised loci such as the rDNA . However , such arguments ignore the wealth of information regarding the function of particular loci in particular environments that is encapsulated in existing gene regulatory systems . In effect , a signalling pathway that strongly induces a gene in response to a particular environmental stimulus marks that gene as being important in that environment relative to a gene that is not expressed . CNV of such a gene is more likely to yield a useful , adaptive result than CNV of a random gene . Of course , it is also more likely to be damaging , and we see exactly this at CUP1: most of the CNV events we observed were contractions that reduce copper resistance ( Fig 3b ) , but , relative to random mutation , the chance of finding an adaptive CNV remains substantial . Stimulated CNV is therefore a high-risk strategy that does not entail foreknowledge of the fitness outcome of genetic change at a particular locus , only the relative importance of that locus in the current environment . However , many highly expressed genes are not environment specific , and such housekeeping genes are likely to be poor candidates for improving adaptation . Simply focusing CNV events at highly expressed genes would likely entail an unacceptable number of deleterious CNV events involving housekeeping genes . This problem is avoided by restricting RFS sites to the promoters of inducible genes . We suggest that the distribution of RFS sites has arisen through natural selection acting on randomly located RFS sites , as any RFS site that always engenders detrimental CNVs would have been rapidly lost . Stimulated CNV is therefore an imperfect but useful cellular mechanism that increases the rate at which adaptive CNV events occur , particularly in suboptimal environments . Importantly , inducible gene expression systems and the placement of RFS sites are products of natural selection acting on random mutations , but these combine to yield a system that accelerates adaptation beyond what is achievable through random mutation .
Yeast strains used in this work are listed in S2 Table . Plasmids are listed in S3 Table , including construction details , and were verified by restriction digest and/or sequencing . Deletion strains were created by standard methods; oligonucleotides are listed in S4 Table . Deletion strains were verified by PCR . To create the PGAL1-HA strain , ADE2 was replaced with MET25 in the S . cerevisiae strain BY4741 ( EuroSCARF ) , and the resulting strain was transformed with pJH252 ( 1 CUP1 repeat ) ; then , the entire repeated region at the chromosomal CUP1 locus [ChrVIII: 212265 . . 216250] was replaced with a LEU2 marker . Plasmid pJH280 , containing 3 copies of the PGAL1-HA construct and an ADE2 marker , was digested with SacI and transformed to replace the LEU2 cassette , which fortuitously amplified on transformation to yield a 17-copy repeat tract ( based on PFGE Southern blot migration ) . The 17xPGAL1-3HA construct was introduced into the MEP background by mating and sporulation and was remated to a MEP wild-type haploid to form the MEP 17xPGAL1-3HA heterozygote strain . For construction of the 3xCUP1 strain and its derivatives , the entire CUP1 locus [ChrVIII: 212265 . . 216250] was deleted in YRH12 , an ade2Δ BY4742-derivative with a single-copy CUP1 plasmid , to form YRH15 . pRH9 , which contains 3 complete CUP1 copies [ChrVIII: 214256 . . 216239] with an ADE2 marker and CUP1 flanking sequences , was digested with SacI and transformed in YRH15 , followed by FOA selection to yield YRH23 . Construction of the 3xSFA1 strain and its derivatives used the same strategy as for 3xCUP1 , except that SFA1 was additionally deleted in YRH15 and then transformed with SacI-digested pJH312 , followed by FOA selection to yield YRH89 . Cells for Fig 2c and 2d were grown in YPD , and other experiments were performed in yeast nitrogen base media supplemented with CSM amino acids and 2% glucose or galactose; all cells were grown at 30°C . YPD contains trace Cu2+ , and yeast nitrogen base media contains 250 nM CuSO4 . All media components were purchased from Formedium . Nicotinamide ( Sigma I17451 ) was added to media at 5 mM . For Fig 6b–6d , cells were grown in SC with or without 0 . 3 mM CuSO4 in 4-ml cultures , diluted 1:1 , 000 from saturated precultures . For Fig 6e and 6f , cells were grown to log phase in SC then diluted 1:8 , 000 into SC with or without 0 . 9–1 mM ( batch-dependent ) formaldehyde that was freshly diluted from 16% or 37% stock solution . Fig 2a represents a meta-analysis of published data; see the Bioinformatic analysis section below for accession numbers and associated culture details . For cell-tracking analyses using the MEP system , cells were inoculated in SD from a plate for 6–8 hours , diluted and grown overnight to OD 0 . 2–0 . 5 . Cultures were diluted to 2x104 cells/ml , 1 μM β-estradiol ( Sigma E2758 ) was added , and cells were grown for 2 hours prior to plating parental culture and splitting the cells for copper or galactose treatment . After 24 hours , 50 μl of each culture was plated on SD agar , and cells were grown for 2 days at 30°C . Individual colonies were then inoculated in 200 μl of SD in a 96-well plate and grown to saturation . For the MEP strain , 2 . 5 μl saturated culture was diluted to 200 μl SC in each well of a 96-well flat-bottomed cell culture plate , with concentrations of CuSO4 up to 3 mM , along with 0 . 5 mM ascorbic acid . Ascorbate increases the cellular uptake of copper [96] , increasing the effective toxicity of copper to allow the measurement of small changes in resistance in cells with high CUP1 copy number . This is helpful since CuSO4 tends to precipitate out of media during culture at concentrations >2 mM . Plates were covered with a gas-permeable membrane and grown at 30°C for 3 days in the dark . Cells were resuspended by pipetting , and OD660 was measured using a BD FLUOstar Omega plate reader . Area-under-curve measurements were calculated for each sample and compared by 1-way ANOVA . For the 3xCUP1 strain , the assay was performed as above but with lower concentrations of CuSO4 ( see Fig 7a and 7b ) , under normal light and without ascorbic acid . For growth curves , saturated precultures were diluted 1:1 , 000 into 200 μl SC per well with or without CuSO4 at the required concentration . Plates were sealed as above and grown at 30°C with shaking in a BD FLUOstar Omega plate reader; OD660 measurements were taken every 15 minutes . Curves were smoothed by averaging across 9 time points , and derivatives were calculated using GraphPad Prism . Six cultures each of 3xCUP1 wild-type and trp1Δ::NatMX6 were grown for 10 generations , 3 untreated and 3 with 5 mM nicotinamide . Cultures were then mixed 1:1 pairwise to give 6 competition cultures , each containing an untreated and a nicotinamide pretreated population of the opposite genotype . The composition of the mixture was determined by plating on–Trp and +Nat plates , and each mixture was inoculated 1:1 , 000 in cultures containing 0 or 0 . 3 mM CuSO4 and outgrown to saturation over 10 generations . Mixture composition of each outgrowth culture was determined by plating . To ensure that the trp1Δ::NatMX6 marker did not affect the result , we performed equal numbers of assays with this strain as the nicotinamide-treated or untreated population . Cells from a 2 ml saturated culture were washed with 50 mM EDTA then spheroplasted with 250 μl 0 . 34U/ml lyticase ( Sigma L4025 ) in 1 . 2 M sorbitol , 50 mM EDTA , and 10 mM DTT at 37°C for 45 minutes . After centrifuging at 1 , 000g , cells were gently resuspended in 400 μl of 0 . 3% SDS , 50 mM EDTA , and 100μg/ml RNase A ( Sigma R4875 ) and incubated at 37°C for 30 minutes . 4 μl of 20 mg/ml proteinase K ( Roche 3115801 ) was added , and samples were mixed by inversion and heated to 65°C for 30 minutes . 160 μl 5M KOAc was added after cooling to room temperature , and samples were mixed by inversion and then chilled on ice for 30 minutes . After 10 minutes of centrifuging at 20 , 000g , the supernatant was poured into a new tube containing 500 μl phenol:chloroform ( pH 8 ) and samples were mixed on a wheel for 30 minutes . Samples were centrifuged for 10 minutes at 10 , 000g , and the upper phase was extracted using cut tips and precipitated with 400 μl isopropanol . Pellets were washed with 70% ethanol , air-dried , and left overnight at 4°C to dissolve in 20 μl TE . After gentle mixing , 10 μl of each sample was digested with 20 U EcoRI-HF ( NEB ) or EcoRV-HF ( NEB ) for 3 hours , phenol:chloroform extracted , ethanol precipitated , and separated on 25-cm 0 . 8% or 1% 1xTBE gels overnight at 120 V for CUP1 analysis or on 1% 0 . 5xTBE gels in a Bio-Rad CHEF DR-III system at 6 V/cm , 15°C , 0 . 5–1 . 5 second switch , and 120° included angle for 16 or 20 hours in 0 . 5xTBE for SFA1 analysis . Gels were washed in 0 . 25 N HCl for 15 minutes , 0 . 5 N NaOH for 45 minutes and twice in 1 . 5M NaCl , 0 . 5M Tris ( pH 7 . 5 ) for 20 minutes before being transferred to HyBond N+ membrane in 6x SSC . Membranes were probed using random primed probes ( S4 Table ) in UltraHyb ( Life Technologies ) at 42°C and washed twice with 0 . 1 x SSC , 0 . 1% SDS at 42°C . Bands were quantified using ImageQuant ( GE ) and data analysed using the GraphPad Prism v6 . 05 to perform 1-way ANOVA analyses comparing the means of all samples ( unless otherwise noted ) with Tukey correction for multiple comparisons . Colonies were analysed in pools of 4 . Cells obtained from 50 μl from each of the 4 saturated cultures were resuspended in 50 μl 50 mM EDTA containing 17 U lyticase ( Sigma L2524 ) and incubated at 37°C for 45 minutes . 1 . 6 μl 10% SDS and 1 μl 20 mg/ml proteinase K were added and samples were incubated at 65°C for 30 minutes . After addition of 32 μl 5 M KOAc and 30 minutes on ice , samples were centrifuged for 10 minutes at 20 , 000g at room temperature , and the supernatant was decanted to a new tube containing 100 μl isopropanol and 1 μl glycogen . Samples were centrifuged for 15 minutes at 20 , 000g at 4°C , and the pellet was washed with 70% ethanol before overnight elution in 20 μl 1x NEB CutSmart buffer with 20 U EcoRI-HF ( NEB ) at 37°C . DNA was quantified using PicoGreen ( Thermo Fisher Scientific ) and separated on PFGE gels using a Bio-Rad CHEF DR-III ( 1% 0 . 5xTBE gel , 6 V/cm , 15°C , 0 . 5–1 . 5 second switch , 120° included angle for 20 hours in 0 . 5xTBE ) , then blotted and probed as above . Copy numbers of individual alleles were plotted using GraphPad Prism v6 . 05 . To calculate p-values , we first estimated the background CNV or amplification mutation rate in the population based on the number of CNV or amplification events observed by PFGE for the unstimulated condition , including the viability of this population after 24 hours of aging . Using this estimate , we then calculated the number of CNV or amplification events in the stimulated condition that would be expected to arise through unstimulated CNV given the viability of this population . We then compared the number of events observed by PFGE to the expected number of CNV or amplification events for the stimulated condition using a goodness of fit χ2 test with 1 degree of freedom . This provides a p-value based on the null hypothesis that all observed CNV or amplification events arose through random mutation . This estimate includes the conservative assumption that any cells that lost viability during the experiment did not undergo CNV . Total RNA was extracted using a mirVANA kit ( Thermo Fisher Scientific ) according to manufacturer’s instructions ( Fig 2 ) or using GTC-phenol as described [7] ( Figs 4 , 5 and 6 ) , and analysed as previously described [7] using probes listed in S4 Table . RNA probes were hybridised at 65°C , DNA probes at 42°C . Indexed mRNAseq libraries were constructed from 500 ng total RNA using the NEBNext Ultra Directional RNA Library Prep Kit ( NEB ) , with poly ( A ) selection using the NEBNext Poly ( A ) mRNA Magnetic Isolation Module ( NEB ) , and sequenced on an Illumina MiSeq . 0 . 5x109 cells grown in YPD , with or without 4 hours of 1 mM CuSO4 treatment , were fixed for 15 minutes in 1% formaldehyde and quenched with 150 mM glycine . Cells were washed 2 times with cold PBS , then resuspended in 600 μl lysis buffer ( 50 mM HEPES [pH 7 . 5] , 140 mM NaCl , 1 mM EDTA , 1% Triton X-100 , 0 . 1% Na-deoxycholate , 0 . 1% SDS , 1x Roche Complete Protease Inhibitors ) , broken with 500 μl 0 . 5mm zirconium beads ( BioSpec ) in an MP Biomedical Fast Prep ( 6 cycles , 30 seconds each ) , then the lysate was separated from the beads and diluted to 1 ml final volume in lysis buffer . Samples were sonicated 19 times , 30 seconds each in a Diagenode Bioruptor on High and cleared by centrifugation at 20 , 000g at 4°C for 5 minutes . 1 μl anti-γH2A ( Millipore 07-745-I ) was added to 100 μl lysate and incubated overnight at 4°C before addition of 15 μl Gammabind beads ( GE ) in 25 μl lysis buffer ( preblocked with 1% BSA ) and incubation for 2 hours at 4°C . Beads were washed 5 minutes each with lysis buffer , 0 . 5 M salt lysis buffer , wash buffer ( 10 mM Tris [pH 8 . 0] , 0 . 25 M LiCl , 0 . 5% NP-40 , 0 . 5% Na-deoxycholate , 1 mM EDTA ) , and TE , then DNA was eluted overnight at 65°C in 200 μl 50 mM Tris [pH 8 . 0] , 10 mM EDTA , 1% SDS . DNA was purified by phenol:chloroform extraction and then ethanol precipitated and eluted in 50 μl TE . Sequencing libraries were prepared from 5 ng of immunoprecipitated material using a NEBNext DNA Ultra kit ( NEB ) and sequenced on an Illumina HiSeq . γH2A ChIP: Reads were mapped to the S . cerevisiae reference genome R64-2-1 using Bowtie 2 v2 . 2 . 5 ( default parameters ) . Peaks were called using MACS2 v2 . 1 . 0 ( -g 12e6—nomodel—extsize 250—keep-dup all ) , artifactual peaks containing a single mismapped read were manually removed , and only peaks with a 2-fold or greater enrichment were considered . Coding sequences ( CDS ) were categorised as upstream-RFS if a γH2A peak was present in 1 kb upstream of the annotated start site using the R script in S1 Text . RNAseq: Read data ( including accessions GSE61783 [97] , GSE74642 , GSE70835 , GSE54831 [98] , GSE54825 [99] , GSE43002 [100] , and GSE41834 [101] deposited at GEO ) were mapped to genome R64-2-1 using HISAT2 v2 . 0 . 5 ( —sp 1000 , 1000 ) . Log2 read counts were performed for each annotated CDS using Seqmonk v0 . 34 . 1 and normalised for CDS length , then the whole data set was normalised for a median expression of 9 ( an arbitrary value that maintained most expression data as positive and required minimal scaling for most data sets ) . Genes were categorised as γ-H2A or non-γH2A using the R script in S1 Text . Cumulative frequency distributions for the data sets were calculated using GraphPad Prism v6 . 05 . Frequency distributions were compared by nested ANOVA , and assumptions for using parametric tests were checked prior to run the analyses . Values for skewness , kurtosis , and variance were consistent with normality and homoscedasticity . Sequencing data has been deposited at GEO ( GSE86283 ) . Source data for all graphs is provided in S1 Data ( cumulative frequency gene expression data ) , S2 Data ( RNAseq data for cells grown with or without Cu ) , S3 Data ( Southern blot image quantification ) , S4 Data ( northern blot image quantification ) , S5 Data ( γH2A ChIPseq data for chromosomes containing CUP1 and SFA1 loci ) , S6 Data ( mother enrichment allele copy numbers and adaptation assay ) , S7 Data ( 3xCUP1 adaptation assays and competition assay ) , and S8 Data ( Raw growth curve data ) . Gel images were processed with ImageJ 1 . 50i—processing involved rotating and cropping , denoising if required ( Despeckle filter ) , and altering window-level settings to improve contrast of relevant bands . Full images of membranes presented in the manuscript are provided in S9 Data—these have been cropped to the borders of the membrane and have undergone minimal window-level adjustments if required to make the bands shown in the presentation figure visible .
|
Evolutionary theory asserts that adaptive mutations , which improve cellular fitness in challenging environments , occur at random and cannot be controlled by the cell . The mutation mechanisms involved are of widespread importance , governing diverse processes from the acquisition of resistance during chemotherapy to the emergence of nonproductive clones during industrial fermentations . Here we ask whether eukaryotic cells are in fact capable of stimulating useful , adaptive mutations at environmentally relevant loci . We show that yeast cells exposed to copper stimulate copy number amplification of the copper resistance gene CUP1 , leading to the rapid emergence of adapted clones , and that this stimulation depends on the highly regulated acetylation of histone H3 lysine 56 . Stimulated copy number variation ( CNV ) operates at sites of preexisting copy number variation , which are common in eukaryotic genomes , and provides cells with a remarkable and unexpected ability to alter their own genome in response to the environment .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"alkaloids",
"chemical",
"compounds",
"formaldehyde",
"molecular",
"probe",
"techniques",
"carbohydrates",
"galactose",
"organic",
"compounds",
"fungi",
"model",
"organisms",
"experimental",
"organism",
"systems",
"copy",
"number",
"variation",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"saccharomyces",
"genome",
"complexity",
"proteins",
"chemistry",
"nicotine",
"molecular",
"biology",
"genetic",
"loci",
"yeast",
"biochemistry",
"organic",
"chemistry",
"post-translational",
"modification",
"acetylation",
"genetics",
"monosaccharides",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"physical",
"sciences",
"genomics",
"saccharomyces",
"cerevisiae",
"chemical",
"reactions",
"southern",
"analysis",
"computational",
"biology",
"organisms"
] |
2017
|
Environmental change drives accelerated adaptation through stimulated copy number variation
|
In eukaryotes , Dom34 upregulates translation by securing levels of activatable ribosomal subunits . We found that in the yeast Saccharomyces cerevisiae and the human fungal pathogen Candida albicans , Dom34 interacts genetically with Pmt1 , a major isoform of protein O-mannosyltransferase . In C . albicans , lack of Dom34 exacerbated defective phenotypes of pmt1 mutants , while they were ameliorated by Dom34 overproduction that enhanced Pmt1 protein but not PMT1 transcript levels . Translational effects of Dom34 required the 5′-UTR of the PMT1 transcript , which bound recombinant Dom34 directly at a CA/AC-rich sequence and regulated in vitro translation . Polysomal profiling revealed that Dom34 stimulates general translation moderately , but that it is especially required for translation of transcripts encoding Pmt isoforms 1 , 4 and 6 . Because defective protein N- or O-glycosylation upregulates transcription of PMT genes , it appears that Dom34-mediated specific translational upregulation of the PMT transcripts optimizes cellular responses to glycostress . Its translational function as an RNA binding protein acting at the 5′-UTR of specific transcripts adds another facet to the known ribosome-releasing functions of Dom34 at the 3′-UTR of transcripts .
In eukaryotes , secretory proteins can get O-mannosylated at serine or threonine residues by protein mannosyltransferases ( Pmt proteins ) . This modification occurs during or shortly after translation , during transit across the secretory pore complex into the ER lumen . O-mannosylation initiates the typical type of fungal O-chains , which mature in the Golgi . In mammalian cells O-mannosylation is a rare but important process , while the bulk of O-chains is formed post-translationally in the Golgi [1 , 2] . Seven and five Pmt isoforms forming Pmt1 , 2 and 4 subfamilies have been described in the yeast Saccharomyces cerevisiae and the human fungal pathogen Candida albicans , respectively [1–3] . Pmt isoforms are largely specific for their protein substrates and the lack of the Pmt2 isoform in C . albicans or at least two isoforms in S . cerevisiae prevents growth [3 , 4] . In C . albicans , each Pmt isoform affects one or another aspect of fungal growth , morphogenesis and virulence [3 , 5] . Impaired O-mannosylation in pmt1 mutants or upon Pmt1 inhibition of a wild-type strain leads to transcriptional upregulation of PMT2 and PMT4 genes , while inhibition of N-glycosylation by tunicamycin upregulates PMT1 expression [6–8] . In both glycostress conditions , the increased levels of underglycosylated proteins also trigger the unfolded protein response ( UPR ) , e . g . by increasing levels of the Kar2 chaperon and matured HAC1 transcript [6 , 7] . UPR induction is known to lower overall translation in yeast cells , although translation of some transcripts is increased [9] . In S . cerevisiae , the Dom34 protein is involved in the “no-go decay” ( NGD ) process , which is one of at least three mechanisms responding to the quality of translated mRNA . NGD relieves stalled translational complexes arising e . g . by secondary structures or rare codons through dissociation of ribosomal subunits and cleavage of mRNA [10 , 11] . Dom34 also rescues ribosomes that accumulate at the 3′-UTR of transcripts [12 , 13] . To promote subunit dissociation and peptidyl-tRNA drop-off from stalled ribosomes , Dom34 co-operates with its associated GTPase Hbs1 [10 , 11] . Very likely , this function of Dom34 is possible because of its high homology to the translational termination factor eRF1 , as well as the structural similarities of the Dom34-Hbs1 and eRF1:eRF3 complexes that can occupy the ribosomal A-site . S . cerevisiae strains carrying single mutations for Dom34 or Hbs1 grow normally , while their combination with mutations impairing components of 40S ( but not 60S ) ribosomal subunits [14] , mutations delaying translation by phosphorylation of eIF2 [15] and yet undefined mutations [16] , leads to severe impairment of growth . The scarcity of 40S subunits due to “stuck” 80S ribosomes and the resultant impaired translational initiation has been suggested as the mechanism causing slow growth of dom34 hbs1 double mutants [14] . The homology of Dom34 to eRF1 only regards its central and C-terminal domains , while its N-terminal domain adopts a Sm-fold that is characteristic of RNA degradation or recognition domains [17 , 18] . However , recent results indicated that Dom34 is not the endonuclease that degrades mRNA in stalled ribosomes [19] . Besides NGD , other mechanisms including separation of free 80S ribosomes [14 , 20] and nonfunctional rRNA decay ( NRD ) [21 , 22] depend on Dom34 to maintain a sufficient supply of ribosomes for translation . Based on the previous discovery of a S . cerevisiae mutant with defective protein O-mannosylation [23] we report here a novel function of Dom34 in translational upregulation of the PMT1 transcript in C . albicans . By promoting the translational initiation of the PMT1 transcript , which under glycostress is strongly increased [6–8] , Dom34 contributes to optimize the overall output of Pmt1 activity that helps to recover from damage to its glycostructures . Its mode of action as an RNA binding protein for specific transcripts differs from the previously described general roles and mechanisms of Dom34 in promoting translation in eukaryotes . By this action , Dom34 functionally links two essential processes in eukaryotic cells , translation and O-mannosylation .
Previously , a S . cerevisiae mutant ( M577 ) defective in O-mannosylation of a heterologous protein ( hIGF-1 ) and some homologous secretory proteins had been identified [23] . The mannosylation defect was recessive and segregated 2:2 in crosses to a wild-type strain suggesting that it was caused by mutation of a single gene . Because M577 did not show easily scorable phenotypes , pmt1 or pmt2 mutations were introduced to explore synthetic phenotypes with mutations affecting O-mannosylation . This approach was led by the finding that pmt1 pmt2 double mutants , but not the single mutants , are resistant to the K1 killer toxin [24] . In agreement , we found that Pmt+- strains YE449 and mutant M577 , as well as the pmt1 and pmt2 single mutants were toxin-sensitive ( blue/dark appearance of colonies on indicator plates ) , while the pmt1 pmt2 double mutant was completely resistant to the toxin ( white appearance of colonies on indicator plates ) ( Fig 1Aa ) . Importantly , similar to the pmt1 pmt2 double mutant , the pmt1 derivative of mutant M577 but not of the parental strain YE449 was completely toxin-resistant . Because the pmt2 derivative of M577 retained sensitivity , the results indicated that the unknown mutation in M577 generates a synthetic protein-O-mannosylation phenotype in combination with a pmt1 but not a pmt2 mutation . To identify the mutation in M577 , its pmt1 derivative was transformed with a genomic bank in vector YEp13 and 80 , 000 transformants were screened for re-appearance of sensitivity to K1 killer toxin . Among 78 initial transformant isolates , 17 carried the complementing activity on a genomic insertion within the bank plasmid . Two genomic loci occurred repeatedly in overlapping inserts ( Fig 1B ) : 5 plasmids contained a region close to the centromere of chromosome XIV and 4 plasmids contained a region close to the centromere of chromosome IX . In both cases , the overlapping clones had a single gene in common: the chromosome XIV clones contained DOM34 , while the chromosome IX-clones contained YIL001w . To clarify if the K1 killer-resistance phenotype in strain M577 pmt1 was caused by mutation of DOM34 , we introduced pmt1 , dom34 and yil001w mutations singly or in combination into the genetic background of the parental strain YE449 . At variance with an initial report [16] but in agreement with a subsequent report [15] we found that dom34 single mutants did not show significant growth defects . Importantly , we detected that the pmt1 dom34 mutant but not the pmt1 yil001w mutant was killer-resistant , resembling the M577 pmt1 strain ( Fig 1Ab ) . Transformation with either PMT1- , DOM34- or YIL001w-overexpressing vectors restored killer-sensitivity of the pmt1 dom34 mutant ( Fig 1Ac ) . These results suggested that in the parental strain M577 , the DOM34 gene is mutated , while YIL001w may represent an extragenic suppressor of the pmt1 and dom34 mutations . In support of this conclusion we found that a diploid constructed from haploids M577 pmt1 and W21 ( YE449 pmt1 dom34 ) was unable to sporulate , as expected for a homozygous dom34 diploid [15] . Furthermore , sequencing of the DOM34 ORF in mutant M577 revealed that it is mutated by insertion of a single T residue following position 366 generating a UAA stop codon leading to a truncated protein of 122 residues . The above results had suggested that in S . cerevisiae , the activities of specific elements involved in translation ( Dom34 ) and protein-O-mannosylation ( Pmt1 ) are functionally linked . We subsequently found that this cooperation not only prevents resistance ( to killer toxin K1 ) but also enhances resistance to hygromycin B ( HygB ) . HygB is an aminoglycoside antibiotic known to block translation [25] , which is particularly active to block growth of glycosylation mutants [26] . dom34 pmt1 double mutants were significantly more sensitive than the pmt1 single mutant , although the dom34 single mutant did not show any sensitivity phenotype ( Fig 1C ) . To generalize Dom34-Pmt1 functional interactions and because S . cerevisiae contains a DOM34 homolog ( YCL001W-B ) with unknown activity , we also studied the single DOM34 gene in the human fungal pathogen C . albicans that contains a family of well-studied Pmt proteins [3] . A homologue of S . cerevisiae DOM34 , ORF19 . 2419 in the C . albicans genome encodes a protein with 36% , 39% and 40% sequence identity to Dom34 proteins of S . cerevisiae , S . pombe and human , respectively . Sequence similarities are observed in the 3 domains of these proteins including a Sm-fold in domain 1 and a sequence in domain 3 , which were both suggested to bind RNA ( S1 Fig [13 , 15] ) . Compared to its homologues , the CaDom34 protein lacks a potential NLS sequence ( position 173–177 in ScDom34p ) . Strains were constructed that lack both alleles of CaDOM34 in the wild-type background ( SK47 ) , in the pmt1 background ( SK24 ) and in the pmt5 background ( JH8-5-11 ) ( disruption scheme in S2 Fig ) . Phenotypes were determined using mutant strains , in which URA3 was reconstituted at its authentic locus [27] . Similar to S . cerevisiae , a homozygous dom34 single mutation did not generate significant growth or morphogenetic defects in C . albicans; furthermore , this mutant was found not to be supersensitive to numerous tested antibiotics or inhibitors including HygB . On the other hand , the HygB-supersensitive phenotype of the pmt1 mutant was significantly increased by an additional dom34 mutation ( Fig 1C ) . Thus , evidence in both yeast species supported a functional link between Dom34 and Pmt1 proteins to generate HygB resistance . However , for S . cerevisiae it cannot be excluded that the DOM34 paralog YCL001W-B contributes to this phenotype . Therefore , and because of its importance as a human pathogen , we focused subsequent analyses on the C . albicans DOM34 gene . The genetic interaction of dom34 and pmt1 mutations in C . albicans prompted experiments to study effects of DOM34 overexpression in this fungus . The DOM34 transcript level was determined by qPCR and showed an equal amount in homozygous pmt4 , pmt5 , pmt6 and heterozygous pmt2/PMT2 mutants as in the wild-type strain CAF2-1 but surprisingly , a 2–3 fold lower level in the pmt1 mutant ( Fig 2A; S3 Fig ) . A transformant of the pmt1 mutant carrying plasmid pSK2 ( MET3p-DOM34FLAG ) that was grown in SD medium contained about sixfold higher DOM34 transcript levels than the untransformed strain and about threefold higher levels than the wild-type strain . Thus , using pSK2 , a moderate overexpression of DOM34 was achieved in C . albicans . Overexpression of DOM34FLAG was able to rescue several known pmt1 mutant phenotypes [3] . The inability of pmt1 mutant colonies to form hyphae was partially suppressed by DOM34 overexpression , which was observed in > 80% of the colonies ( Fig 2B ) . Furthermore , the sensitivity of pmt1 mutants to HygB and to high temperature ( 42°C ) was also partially suppressed ( Fig 2Ca ) . Interestingly , however , suppression was not achieved using a derivative of pSK2 carrying a point mutation in the DOM34 ORF that encodes the E21A variant Dom34 protein ( Fig 2Cb ) . It has been suggested that this residue is important for RNase activity of Dom34 proteins [18] . We asked next if the suppression of pmt1 mutant phenotypes by DOM34 overexpression depended on the Pmt isoforms remaining in this strain . For this purpose , we constructed pmt1 pmt5 and pmt1 pmt6 double mutants , which both showed the supersensitive pmt1 mutant phenotype ( Fig 2Cc and 2Cd ) , while single pmt5 or pmt6 mutants are not supersensitive to HygB [3] . Interestingly , double mutant transformants carrying the DOM34 overexpression plasmid pSK2 did not show any recovery of pmt mutant phenotypes ( Fig 2Cc and 2Cd ) . This result indicates that Pmt5 and Pmt6 isoforms are required for the observed rescue by Dom34 overproduction . The relevance of the Pmt2 and Pmt4 isoforms could not be tested in this manner , because PMT2 is an essential gene and pmt1 pmt4 double mutants are not viable [3] . These experiments also revealed that unexpectedly , a C . albicans strain lacking all members of the Pmt1 subfamily ( Pmt1 and Pmt5 ) is fully viable , although in S . cerevisiae heteromeric Pmt1-Pmt2 or Pmt1-Pmt5 complexes have been described to be essential for growth [28] . Conceivably , Dom34 could have suppressed Pmt1 deficiencies by several mechanisms , especially by increasing transcription/transcript levels of several PMT genes including PMT5 and/or PMT6 , as suggested by the above experiments ( Fig 2C ) . To explore this notion , we determined transcript levels in the DOM34 overexpression strain and found that none of the PMT transcript levels was increased ( S4A Fig ) . In addition , the absence of Dom34 did not decrease PMT transcript levels , while PMT2/PMT4 transcript levels were increased in the pmt1 dom34 double mutant , as described for the pmt1 single mutant [6] ( S4B Fig ) . Thus , increases and decreases of Dom34 levels were not related to PMT transcript levels . To explore effects on Pmt protein levels a C . albicans strain was constructed , in which one PMT1 allele was fused to sequences encoding the hemagglutinin ( HA ) -epitope ( Fig 3 ) , which was subsequently transformed with the DOM34 overexpression vector pSK2 or the corresponding empty vector . This strain produced considerably higher Pmt1HA protein levels as compared to a transformant carrying an “empty” control vector ( Fig 3A ) . Scanning of band intensities revealed that DOM34 overexpression increased the mean Pmt1-HA/actin ratio 1 . 78 fold ( p = 0 . 025 ) . The 5′-UTR of the PMT1 transcript encompasses 190 or 218 nt [29 , 30] and contains an intriguing CA/AC-rich sequence , which is ordered into three overlapping 11-mer ACAACCACAAC repeats between nt -157 to -179 ( Fig 3B , top ) . To examine if this sequence is involved in overproduction of the Pmt1 protein by DOM34 overexpression we generated genomic fusions containing two different lengths of the PMT1 upstream region joined to the RLUC reporter gene . One fusion did not contain most 5′-UTR sequences ( pPdC3-HIS ) , while the second contained one full and one half of the 11-mer repeat ( pPdC2-HIS ) ( Fig 3B top ) . These strains were transformed with the DOM34 overexpression plasmid pSK2 or a control plasmid and luciferase activity of the double transformants was determined . The results revealed that the construct containing 5′-UTR sequences including the 11-mer repeat was stimulated significantly by DOM34 overexpression , while no activation occurred for the construct lacking UTR sequences ( Fig 3B bottom ) . As stated above , the C . albicans Dom34 protein lacks a consensus NLS sequence , suggesting that its primary action takes place outside of the nucleus . Furthermore , differential centrifugation of cell extracts identified HA-tagged Dom34HA to a large extent in the soluble fraction ( cytoplasm ) ( S5 Fig ) . Collectively , the results suggest that Dom34 overproduction stimulates translation of Pmt proteins . The site of the Dom34 stimulatory activity appears to lie in a specific sequence within the 5′-UTR of target transcripts , as exemplified by the Dom34-mediated regulation of the PMT1-RLUC fusion . We next carried out polysome analyses to establish the role of Dom34 in translation of PMT1 and the ACT1 housekeeping transcripts . For this purpose , cellular lysates of the control strain CAF2-1 and the dom34 mutant JH47-2 were separated by sucrose gradient centrifugation to establish polysomal profiles ( Fig 4A ) . Profile comparisons of both strains revealed that pre-polysomal fractions containing 40S , 60S and 80S rRNA were more pronounced in the dom34 mutant than in the control strain; furthermore , polysomal peaks were lower in the mutant and decreased at >2n polysomes , whereas in the control strain , the 3n peak was even greater than the 2n peak . These results indicate that in the dom34 mutant , translational efficiency is generally but moderately reduced . To examine the efficiency of PMT1 translation in control and dom34 mutant cells , the fractions of the polysomal gradient were examined for the presence of the PMT1 transcript by RT-qPCR , using a spiked-in control RNA as a reference . The result indicates clearly that the PMT1 transcript in the dom34 mutant is found predominantly in the pre-polysomal fraction , while in the control strain , significantly higher amounts reside in the polysomal fractions ( Fig 4B ) . To establish if lower translational efficiency in the mutant concerns only the PMT1 transcript , we also examined the transcript profile for the PMT4 and PMT6 transcripts , which also were enriched in the polysomal fraction in the presence of Dom34 ( S6A Fig ) indicating that Dom34 positively regulates the translation of several PMT transcripts . In comparison , translation of the transcript for the housekeeping gene ACT1 appeared less affected by Dom34 . These effects were quantitated the Kolmogorov-Smirnov test [31 ( Fig 4C ) . By this algorithm , the calculated Dom34-dependent enrichment of transcripts in the polysomal fraction showed a statistically significant increase for the PMT1 transcript ( D = 0 . 6 , p-value = 0 . 031 ) , the PMT4 transcript ( D = 0 . 5 , p = 0 . 042 ) and for the PMT6 transcript ( D = 0 . 5 , p = 0 . 066 ) ; in contrast , a lower and insignificant enrichment was calculated for the ACT1 transcript ( D = 0 . 3 , p = 0 . 675 ) ( Fig 4C ) . Collectively , the results suggest that Dom34 generally enhances but is not absolutely required for translation in C . albicans . The degree of Dom34-mediated translational enhancement differs between transcripts and may particularly affect specific groups of transcripts including transcripts for different Pmt isoforms . Interestingly , the 5′-UTR of all PMT transcripts ( but not the 5′-UTR of the ACT1 transcript ) contains at least one CAAC motif , which in the above-described ACAACCACAAC repeat region of the PMT1 5′-UTR occurs eight times ( S6B Fig ) . Conceivably , the positive action of Dom34 on translation of PMT transcripts is mediated by this sequence . Recombinant Dom34 ( S7 Fig ) was added to a rabbit reticulocyte in vitro translation system , using RNA carrying the coding region for click beetle green luciferase ( CBGluc ) , either containing or not containing the PMT1 5′-UTR ( Fig 5Aa ) . As expected , protein products of identical molecular masses of 60 kDa were obtained for both proteins ( Fig 5Ab ) . Additional experiments indicated that the presence of the 5′-UTR augmented CBGluc biosynthesis , while added Dom34 reduced production , if the 5′-UTR was present , but not in its absence ( Fig 5Ba ) . These effects were quantitated by measuring CBGluc luminescence of the samples , which demonstrated that in the presence of Dom34 , for RNA containing the 5′-UTR , enzyme activity was decreased by about 40% in three independent measurements ( Fig 5Bb and 5Bc ) . Next , using translational assays containing varying amounts of Dom34 protein , it was shown that already at 0 . 1 μM , Dom34 reduces translation efficiency significantly , while it fully inhibits translation at 0 . 25 μM . At higher concentrations of Dom34 , CBGluc production increases and again to diminishes at 2 . 5 μM . These results demonstrate clearly that Dom34 is able to strongly influence translational activity via the PMT1 5′-UTR sequence . Use of a heterologous in vitro system may explain the concentration-dependent , negative rather than the expected positive action of Dom34 on translation . To clarify the mechanisms how Dom34 influences translation we carried out electrophoretic mobility shift assays ( EMSA ) with recombinant Dom34 of the complete PMT1 5’-UTR sequence . The 5′-UTR was obtained by run off in vitro transcription using T7 polymerase using pRG01 as the template , which was cut by BglII at the 3′-end of the UTR . The 3′-[32PpCp] end-labelled RNA was incubated with increasing amounts of Dom34 , separated by native PAGE ( 5% acrylamide ) and examined by autoradiography . The 5′-UTR RNA migrated very slowly compared to the control 6S RNA from E . coli which is not much shorter and has a well-known compact secondary structure [32] ( Fig 6A ) . This suggests a more bulky secondary structure for the PMT1 5’-UTR . At concentrations above 0 . 3 μM , the binding of Dom34 to the UTR became visible resulting in two retarded complexes ( I and II ) . The second complex appeared at high Dom34 concentrations suggesting that multiple proteins are bound to one RNA molecule . At 2 . 5 μM of Dom34 , no free RNA remained suggesting that all of the UTR was bound or degraded; in contrast , little binding of Dom34 ( or Dom34E21A ) to the 6S control RNA was observed ( S8A and S8B Fig ) . Interestingly , starting already at very low concentrations of Dom34 ( 150 nM ) some smaller-size degradation fragments of the UTR were observed ( asterisks ) . Because the degradation products did not appear as sharp bands during native gel electrophoresis , we analyzed the same samples by denaturing gel electrophoresis ( Fig 6B , right panel ) . Autoradiography clearly showed that already at 0 . 1 μM Dom34 the 5′-UTR is partially degraded and that the degradation products have distinct lengths of 1 to 40 nt and around 100 to 120 nt suggesting endonucleolytic cleavage rather than 3’ or 5’ exonucleolytic degradation . Exonuclease-mediated degradation from the 3′- end would have removed the 3′-labeling , which would have decreased amounts of full-length UTR RNA , while degradation from its 5′-end would have generated a smear of cleavage products . However , the results of denaturing gel electrophoresis indicate that the slight decrease of full-length UTR occurring at Dom34 concentrations > 0 . 6 μM is solely due to endonucleolytic fragmentation at specific sites . Under denaturing separation conditions protein-UTR interactions are disturbed , revealing the presence of full-length UTR that is not seen under native conditions . This result also provides an indirect proof for UTR binding by Dom34 at higher concentrations . The 5′-UTR structure predicted by the RNAfold program ( http://rna . tbi . univie . ac . at/cgi-bin/RNAfold . cgi ) indicated that it contains several single-stranded regions including the presumed Dom34-binding CA/AC-rich sequence , as well as two double-stranded regions that compact the structure ( S9A and S9B Fig ) . In the absence of Dom34 , cleavage with single strand-specific RNase U2 confirmed that the CA/AC-region is unpaired , while in the presence of Dom34 , this region was protected from RNase digestion , consistent with Dom34 binding in this region ( S9C Fig ) . Furthermore , besides cutting the 5′-UTR at position 80–100 from 5′-end , binding of Dom34 made paired regions more accessible to RNase attack , thus indicating that significant rearrangement of the 5′-UTR had occurred during Dom34 binding . Specific sequences in Dom34 homologs have been suggested to be important for the translational functions of Dom34 [17–20 , 33] . A conserved glutamic acid residue in domain 1 ( E21 in CaDom34 ) has been suggested to be important for RNase activity [18] and the importance of E21 for CaDom34 function was already shown above by the demonstration that the E21A variant is unable to suppress pmt1 phenotypes ( Fig 2C ) . To test effects of the E21A mutation on binding or cleavage of the PMT1 UTR the Dom34-E21A variant was recombinantly produced in E . coli . Binding of Dom34-E21A to purified 5’-UTR was observed at very low protein concentrations ( > 0 . 1 μM Dom34 ) ; estimation of an apparent KD of about 100 nM deduced from these results demonstrates the very high affinity of Dom34-E21A to the UTR RNA , which is greater than the affinity observed for the wild-type Dom34 protein ( Fig 6C ) . Importantly , minimal UTR cleavage is seen at concentrations > 0 . 6 μM of Dom34-E21A . The denaturing gel confirms this result , because only at 1 μM Dom34-E21A minimal degradation products become visible , while the amount of full-length RNA is not changed significantly ( Fig 6B , left panel ) . We conclude that Dom34 is an RNA binding protein , which favours certain RNA targets including the PMT1 5′-UTR . Furthermore , Dom34 has the capacity to endonucleolytically cleave bound target RNA and requires its E21 residue for this function . The above experiments had indicated that Dom34 stimulates translation ( Fig 4 ) and that a specific sequence containing three overlapping 11-mer repeats within the 5′-UTR of the PMT1 is involved in this function ( Fig 3 ) . To explore if Dom34 directly interacts with this sequence we tested interaction of a corresponding RNA oligomer containing the repeat sequence with Dom34 protein in EMSA . These experiments showed a single retarded complex ( complex I ) with wild-type Dom34 , which in the presence of lauryl sarcosinate ( asterisks ) split into two retarded complexes ( I , II ) ( Fig 7 ) . In contrast , BSA as a control protein did not bind the RNA oligonucleotide and the labelled RNA oligomer could be competed out using a 100-fold excess of unlabeled oligomer . The E21A Dom34 variant showed even enhanced binding compared to native Dom34 , supporting its binding behavior to the full-length 5′-UTR ( Fig 6 ) . Furthermore , the Dom34N317A variant containing a mutation of a key residue in a sequence with high homology to RNA binding proteins [15] also bound to the 5′-UTR RNA oligomer as native Dom34 . All Dom34 versions also led to a slight partial degradation of the labelled RNA oligonucleotide ( smear emanating from free oligonucleotide ) ; however , because a similar pattern was observed using the BSA control protein , it appears that no additional RNase activity is associated with Dom34 .
By NGD , NRD and 80S release mechanisms , Dom34 and co-regulatory molecules including Hbs1 maintain sufficient numbers of ribosomes and thereby assure efficient translation in eukaryotes [10 , 14] . However , recent in vivo results have indicated that Dom34 affects ribosome occupancy at only 11% of all genes and it is yet unknown , why stalled ribosomes on some transcripts are resolved by Dom34 , while other transcripts with similar structural impediments are not affected [13] . A restricted rather than a general function of Dom34 in translation was also suggested by the finding that S . cerevisiae dom34 mutants do not show a general growth phenotype in all genetic backgrounds [15 , 16] . Thus , the target specificity of Dom34 for specific transcripts remains to be clarified . Results presented here indicate for the first time that Dom34 can serve as an RNA binding protein that could enhance translation of specific transcripts . These conclusions were obtained using the yeast C . albicans as experimental organism , which contains a single allele for a Dom34 protein with high similarity to orthologs in other organisms , while S . cerevisiae , because of its ancient genome duplication [34] , contains in addition to DOM34 a paralog ( YCL001W-B ) of unknown function . Our results indicate on the one hand a general function of Dom34 on translation , since polysome gradients in the dom34 mutant showed an increase of monosomes and decrease of polysomes , as compared to a wild-type strain . Furthermore , the abundance of the housekeeping ACT1 transcript encoding actin was shifted slightly to the monosomal fraction in the dom34 mutant suggesting reduced translation . However , the general translational effects of Dom34 appear to be moderate , since growth or morphogenesis of C . albicans was not affected in unstressed conditions . In contrast , in pmt1 mutants with defective O-mannosylation that lack a major isoform of Pmt proteins [3] , the contribution of Dom34 to growth phenotypes was clearly apparent . Protein-O-mannosylation is essential for fungal growth and its absence triggers the UPR response , because of the accumulation of underglycosylated , wrongly folded proteins in the ER lumen [6 , 7] . The HAC1 transcript has recently been identified as a specific target of Dom34 , which releases ribosomes stuck at the 3′-UTR [13] . One possible scenario explaining the genetic interaction of mutations in DOM34 and PMT1 is that Dom34 , by its ribosome releasing function , assures efficient translation of the mature HAC1 transcript . By this action , UPR responses could stimulate growth of pmt1 mutant cells . Aside from this general activity for maintenance of ribosome levels , our results suggest a specific stimulatory function of Dom34 on the translation of certain transcripts . As shown by the polysome profiling experiments , the translation of the PMT1 , PMT4 and PMT6 transcripts were more strongly affected by the presence of Dom34 than that of the ACT1 transcript . Furthermore , Dom34 binding to the 5′-UTR of the PMT1 transcript and activation of a reporter gene by this sequence supported a direct positive role of Dom34 on the translational initiation of the PMT1 transcript . This action is consistent with lowered Pmt1-mediated O-mannosylation of a heterologous protein ( hIGF-1 ) during its massive overproduction in a dom34 single mutant of S . cerevisiae [23] . In this overproduction condition , Dom34-mediated improvement of Pmt1 activity appears to be needed to obtain full O-mannosylation of target proteins . On the other hand , DOM34 overexpression only rescued pmt1 mutant phenotypes , if PMT5 and PMT6 genes were present , suggesting that the posttranscriptional stimulatory action of Dom34 is not exclusive for the PMT1 transcript but applies to transcripts of several PMT genes , which was indeed confirmed for the PMT4 and PMT6 transcripts . Dom34 activity may be especially needed for translation of PMT2 and PMT4 transcripts , which in a compensatory response are upregulated in pmt1 mutants [7] . On the other hand , stalled ribosomes were not found previously in a dom34 mutant at any of the seven S . cerevisiae PMT transcripts [13] , indicating that ribosome release by Dom34 is not a specific translational activation mechanism for PMT transcripts . The function of Dom34 as a relatively specific translational enhancer complements other posttranscriptional mechanisms in C . albicans that recently have been discovered to be essential for the biology and virulence of this fungus [35 , 36] . The ability of Dom34 to act as a RNA binding protein adds a new facet to the mode of action of Dom34 in eukaryotes . RNA binding was specific , because it did not occur with a control RNA , it was outcompeted efficiently and it occurred at low Dom34 concentrations . A stretch of residues in domain 3 of Dom34 had previously been shown to share high homology to RNA binding proteins [15 , 33] . However , mutation of a central residue in this sequence , N317 to N317A , did not affect binding of Dom34N317A to the 5′-UTR oligonucleotide suggesting that this Dom34 sequence may be important for binding to the 3′-UTR but not to the 5′-UTR of transcripts . The predicted structure of the 5′-UTR , which was supported by limited RNase digestion , was found to contain several single-stranded regions , one of which comprised a CA/AC-rich region with three ACAACCACAAC repeats . Binding of Dom34 protected the 5′-UTR at this region from RNase digestion and Dom34 bound to a corresponding oligonucleotide , thus identifying the CA/AC-rich region as the Dom34 binding site . Interestingly , 5′-UTR regions of all five transcripts for C . albicans Pmt isoform ( but not the 5′-UTR of the ACT1 control transcript ) contained at least one CAAC repeat ( eight in the PMT1 5′-UTR ) , which may constitute the minimum requirement for Dom34 binding . Dom34 did not show a major general RNase activity in our experiments , in agreement with Passos et al . [19] , although distinct levels of specific degradation products of the bound 5′-UTR were detected . A Dom34 mutant , in which the conserved glutamic acid residue ( E21 in CaDom34 ) important for in vitro RNase activity [18] was altered ( E21A ) , strongly bound to the 5′-UTR but did not cause its endonucleolytic degradation , suggesting that E21 is relevant for its RNase activity . In agreement , rescue of pmt1 phenotypes did not occur with overexpression of the DOM34E21A allele , thus confirming the importance of the E21 residue for the function of Dom34 . Dom34 binding to the CA/AC-rich sequence caused cleavage of the 5′-UTR by its RNase activity at a distant site , within predicted double-stranded regions causing a major structural alteration of the 5′-UTR . The mechanism , by which Dom34-mediated binding and cleavage of 5′-UTR sequences is able to stimulate translation of specific transcripts remains to be established . Possibly , the actions of Dom34 on the 5′-UTR structure could improve the accessibility of the AUG codon to ribosomal factors and subunits . Whatever the mechanism may be , the results suggest that in addition to its general action at the 3′-UTR to release stalled ribosomes , Dom34 may act at the 5′-UTR of specific transcripts including PMT transcripts to stimulate translational initiation in glycostress conditions .
S . cerevisiae strains are listed in S1 Table . To disrupt PMT genes in strain YE449 , a 3 . 3 kb XbaI-SacI fragment of pDIS2 , carrying pmt1Δ::URA3 , or/and a 2 . 4 kb HindIII-BamHI fragment of pBDis [24] , carrying pmt2Δ::LEU2 , were used for transformation , selecting prototrophs [37] . To disrupt DOM34 in strain YE449 , a fragment generated by PCR on plasmid pUG6 [38] was used for transformation , selecting G418-resistance ( PCR primers ScDOM34 disrupt for /rev ) Oligonucleotides are listed in S2 Table . Likewise , to disrupt YIL001w , a PCR fragment generated by primers ScYIL001w disrupt for/rev was used . Correct integration of the disruption cassettes was verified by Southern blottings . C . albicans strains are listed in S1 Table . For disruption of CaDOM34 its 5′- and 3′-regions flanking the ORF were inserted into pSFU1 to frame the SAP2p-FLP and URA3 markers [39] . The DOM34 5′-region was amplified by genomic PCR on DNA of strain SC5314 using primers FPD34/RPD34 and the 3′-region was amplified using primers FDD34/RDD34 . The resulting DOM34 5′-flanking region , as a SacII-NotI fragment , and the 3′-flanking region , as a XhoI-ApaI fragment , were inserted into the respective sites of pSFU1 to generate pJB28 . JB28 was cut with ApaI and SacII and the large fragment was used for transformation of strain CAI4 or of pmt1 mutant CAP1-3121 [40] . Correct integration in the resulting strains SK47 and SK24 , respectively , was verified by diagnostic colony PCR and confirmed by Southern blottings , using the 5′-region flanking the DOM34 ORF as the probe ( S2 Fig ) . Removal of the URA3 cassette and disruption of the second allele was carried out as described [41] . Following disruption of both CaDOM34 alleles and eviction of the disruption cassette , URA3 was reconstituted at its authentic locus by transformation with a genomic fragment [3] to generate strains JH47-1/2 and JH24-4/5 . The PMT5 gene was disrupted in the dom34 background ( strain SK47 ) or in the pmt1 background ( strain CAP1-3121 ) as previously described [3] to generate strains JH5-3-1 and P15-274 , respectively . To generate a PMT1 gene encoding a C-terminally hemagglutinin ( HA ) epitope-marked Pmt1 protein PCR fragments were used containing the sat1 selectable marker and flanked by regions of homology to PMT1 . For tagging pSAT1-3HA was used as template , which was constructed by replacing the URA3 gene in p3HA-URA [3] situated between PstI and BglII sites with the ACT1p-sat1-ACT1t-cassette of pFC1 [42] on a PstI to BamHI fragment . PCR was done using primers CaPMT1del-for/-rev to generate a tagging fragment for Pmt1 , which was chromosomally integrated by transformation of strain CAI4 selecting for nourseothricin resistance ( sat1 ) ; resulting strain CIS23 . For C-terminal HA-tagging of Dom34 , an insertion fragment was generated by PCR using p3HA-URA as template and primers Dom34-HA-for/rev . The insertion fragment was transformed in strain CAI4 selecting for uridine prototrophy; resulting strains JHCa1-1 ( -2 ) . Correct integration of tagging cassettes was verified by diagnostic PCR of transformants . S . cerevisiae and C . albicans strains were grown in complex YPD and synthetic SD media [37] . For hyphal induction of C . albicans the strains were grown for 3–4 days at 37°C on Spider-medium [43] or on 2% agar containing 5% horse serum . To compare killer sensitivities of several S . cerevisiae strains , YPD agar containing methylene blue and buffered to pH 4 . 5 was autoclaved , cooled to 50°C and 17 μl of a saturated culture of the killer K1- secreting strain RC130 was added before pouring plates [44] . Strains to be tested were pre-grown on YPD medium and replica printed onto these plates and grown at 18°C for 4–7 days . Sensitive strains appeared blue at this time , while resistant strains remained white . To complement killer K1-resistance of strain M577 pmt1 by genomic clones , we transformed it with a genomic S . cerevisiae bank in YEp13 [45] and obtained 80 000 transformants on SD minimal medium . Colonies of transformants were replica-printed onto SD/methylene blue/pH 4 . 5-medium containing killer K1 strain RC180 . Blue colonies were picked , their plasmid was isolated and retransformed into M577 pmt1 . Among 78 initial transformants , 17 transformants were identified , whose plasmids restored Killer-sensitivity upon retransformation . Insert ends in these plasmids were sequenced using primers YEp13-Bamflank-A/-B , flanking the BamHI insertion site of YEp13 . The 7 kb BamHI-XhoI fragment of pDM3 carrying a genomic ScPMT1 fragment [46] was subcloned into YCplac111 , to generate pSW20 . Derivatives of YEp13 containing a 5 . 59 kb genomic insert carrying ScDOM34 ( pSW577/20 ) or containing a 4 . 55 kb genomic insert carrying YIL001w were used in some complementation experiments . Expression vectors encoding HA-tagged proteins were constructed by PCR amplification of ORFs and introducing them into YCpIF17 [47] . DOM34 was amplified using primers ScDom34N/C , the 1 . 2 kb product was digested with EcoRI and PstI and introduced into YCpIF17 , resulting in plasmid pSW22 . The GAL1p-HA-DOM34 fragment of pSW22 was excised with XhoI and XbaI and inserted into YCplac111 ( SalI , XbaI ) to generate pSW25 . Likewise , YIL001w was amplified using primers YIL001wN/C , inserted into YCpIF17 to generate pSW21 and transferred into YCplac111 to generate pSW24 . To construct a CaDOM34 expression vector its coding region was PCR amplified on gDNA using primers p1-DOM-FLAG and p2-DOM-FLAG . We then inserted the resulting PstI-SphI fragment downstream of the MET3 promoter into a derivative of pFLAG-Met3 [48] , which had been modified by adding the CaARS2 replicator on an AatII fragment [49] . The resulting plasmid pSK2 was modified further by oligonucleotide-directed mutagenesis to encode the Dom34-E21A variant; for this purpose , pSK2 was used as template with primers Dom34Mut21for/rev in the QuikChange protocol ( Stratagene ) to generate pSK2mut . Using anti-FLAG antibody the FLAG-tagged Dom34 protein could be identified as a 45 kDa protein in immunoblottings of cellular extracts , although several cross-reacting proteins prevented immunocytological analyses . The C . albicans DOM34 ORF was inserted downstream of the T7 promoter , between the NdeI and XhoI restriction sites of expression vector pET22b ( Invitrogen ) . The resulting vector pET22b-Dom34 encoded a Dom34 protein containing six histidine residues at its C-terminal end . The single CTG sequence encoding non-standard S288 in C . albicans was then altered to standard serine-encoding TCG by oligonucleotide-directed mutagenesis , using primers Dom34-Leu ( mut ) for/rev according to the QuikChange protocol ( Agilent ) . The resulting vector pET22b-Dom34+ was modified further similarly to encode variants potentially important for the function of Dom34 . For the E21A variant and the N317A variant , oligonucleotides Dom34Mut21fw/rev and Dom34Mut317fw/rev were used for mutagenesis . All expression vectors were verified by sequencing and transformed in E . coli strain Rosetta ( Novagen ) . Transformant cultures were grown to OD600 = 0 . 6 in LB medium containing 100 μg/ml ampicillin and 50 μg/ml chloramphenicol , 100 mM IPTG was added to induce the T7 promoter and cultures were incubated further at room temperature . Cell pellets were resuspended in buffer ( 50 mM TrisHCl/pH 7 . 9; 500 mM NaCl; 10% glycerol ) containing protease inhibitors and disrupted by ultrasonication , followed by centrifugation ( 10 , 000 x g for 20 min ) . To assure the solubility of recombinant proteins , 0 . 5% lauryl sarcosinate was added to the buffer in some experiments . Soluble Dom34-His6 proteins in the supernatant were purified by affinity chromatography using Histrap Crude-Agarose columns ( GE ) using 5 mM , 50 mM and 250 mM imidazole for elution . Elution fractions were collected using a ÄKTAprime collector ( GE Healthcare ) and analyzed by SDS-PAGE ( 4–20% acrylamide ) followed by Coomassie Blue staining or by immunoblotting using an HRP-coupled anti-His tag antibody ( Qiagen ) ( S7 Fig ) . About 2 mg of Dom34 variants with at least 95% purity were obtained . Relative transcript levels ( RTL ) of specific C . albicans genes were determined by quantitative reverse transcription PCR ( qPCR ) using the ACT1 transcript as reference , as described [7 , 8] . Oligonucleotides designated RT ( S2 Table ) were used for this purpose . For the in vitro generation of transcripts , plasmid pUC18-T7CBG was constructed , which contains the promoter of bacteriophage T7 upstream of the coding region for click beetle green luciferase ( CBGluc ) . For this purpose , primers CBG-Stu/Bam for/rev were used to amplify the coding region for CBGluc from plasmid CBGluc-pMK-RQ [50] , which was inserted into pUC18T7 [51] . The resulting plasmid was modified further by insertion of the PMT1 5′-UTR sequence ( -218 to -1 [29] ) that was amplified using primers PMT1_5′UTR/-rev BglII_long from gDNA ( strain CAF2-1 ) ; the resulting plasmid was named pRG01 . Plasmids were linearized by BamHI ( pUC18-T7-CBG ) or BglII ( pRG01 ) downstream of the CBGluc coding region and were used as template for in vitro run-off transcription using T7 RNA polymerase ( Ambion SP6 in vitro transcription kit ) according to Gildehaus et al . [51] ) . The resulting transcripts were used for in vitro translation in a rabbit reticulocyte lysate kit ( Promega L4960 ) that incorporates biotinylated lysine in the protein products ( Promega Transcend Detection System ) . Divergent from the manufacturer's protocol the reaction was reduced to a total volume of 20 μl with 0 . 75 μl Transcend™ Biotin-Lysyl-tRNA . For each reaction 4 . 9 fmol RNA ( 244 nM ) and variable amounts of Dom34 were preincubated for 10 min at RT and the reaction was started by adding a premix of reticulocyte-lysate , amino acids and RNasin . After 100 min at 30°C , 1 μl was separated by SDS-PAGE ( 10% acrylamide ) , followed by reaction with horseradish peroxidase-coupled streptavidin , as described by the manufacturer . Furthermore , CBGluc luminescence was measured by adding 45 μl of water to 5 μl of the translation mix , followed by addition of 50 μl Chroma-Glo substrate ( Promega ) . Measurements were done in wells of 96-microtiter plates using TriStar LB 941 luminometer ( Berthold Technologies ) for 1 sec , as described [50] . Labelled RNA encompassing the 5′-UTR was obtained by multiple rounds of run-off transcription of pRG01 linearized by BglII and T7 RNA polymerase [48] . pUC18-T7-6S linearized by StuI served to generate 6S control RNA [51] . After purification on agarose-gels followed by glass wool-elution , the radioactive labeling at the 3′-end was performed by T4-RNA ligase-catalyzed addition of 32P-pCp , as described [52] . If necessary , the labeled RNA was purified in a second step by electrophoresis on a 5% denaturing polyacrylamide gel . RNA-Dom34 complex formation was assayed by incubating 50–100 cps 32P-labeled RNA together with variable amounts of Dom34 for 10 min at 30°C , in 10 mM Tris-HCl/pH 8 . 0 , 100 mM NaCl . Complexes were then challenged by adding of heparin ( final concentration 50 ng/μl ) for 10 min and separated on native 5% polyacrylamide gels or denaturing 10% polyacrylamide gels . For EMSA of the PMT1-5′UTR RNA oligonucleotide , it was biotinylated using the Pierce RNA 3′ End Biotinylation Kit ( Thermo Scientific ) according to the instructions of the manufacturer . Protein-RNA binding assays were done using the LightShift Chemoluminescent RNA EMSA Optimization and Control Kit ( Thermo Scientific ) . In a total volume of 20 μl , binding assays contained 20 nM biotinylated RNA oligonucleotide and 400 nM purified Dom34-His6 ( or its E21A or N317A variants ) , which were supplemented in part by 2 μM unlabelled oligonucleotide . Following incubation at room temperature for 20 min , 5 μl of 5x REMSA loading buffer was added and the assay components were separated by native PAGE ( 6% acrylamide ) and electroblotted onto a nylon membrane; the biotinylated RNA was fixed on the membrane by a short UV treatment and was detected using the Chemiluminescent Nucleic Acid Detection Module Kit ( Thermo Scientific ) . Cells of wild-type strain CAF2-1 and of dom34Δ mutant JH47-2 cells were grown exponentially in YPD media to OD600 0 . 4–0 . 6 . Preparation of cells and polysome gradients were performed as described by Garre et al . [53] with some modifications . A portion of the culture ( 80 ml ) was recovered and chilled for 5 min on ice in the presence of 0 . 1 mg/ml cycloheximide ( CHX ) . Cells were harvested by centrifugation at 6000 x g for 4 min at 4°C and resuspended in lysis buffer ( 20 mM Tris-HCl , pH 8 , 140 mM KCl , 5 mM MgCl2 , 0 . 5 mM dithiothreitol , 1% Triton X-100 , 0 . 1 mg/ml CHX , and 0 . 5 mg/ml heparin ) . After washing , cells were resuspended in 700 μl of lysis buffer , a 0 . 3 ml volume of glass beads was added , and cells were disrupted by shaking on a Vortex Genie 2 ( setting 8 ) using 6 cycles for 40 s at 6 . 5 ms-1 . Between cycles cells were placed on ice for 5 min . Lysates were cleared by centrifuging twice for 5 min , first at 5 , 000 rpm , and then the supernatant was recovered and was centrifuged at 8 , 000 rpm . Finally , glycerol was added to the supernatant at a final concentration of 5% , before storing extracts at -70°C . Samples of 10–20 A260 units were loaded onto 10–50% sucrose gradients and were separated by ultracentrifugation for 2 h and 40 min at 35 , 000 rpm in a Beckman SW41 rotor at 4°C . Then , gradients were fractionated using isotonic pumping of 60% sucrose from the bottom , followed by a recording of the polysomal profiles by online UV detection at 260 nm ( Density Gradient Fractionation System , Teledyne Isco , Lincoln , NE ) . To analyze the RNA of the polysomal fractions , RNA from 200 μl of each fraction was extracted using GeneJet RNA extraction kit ( STREK , Biotools ) . To each sample 1 μg of in vitro transcribed RNA ( HiScribe™ T7 High Yield RNA Synthesis Kit , NEB ) was added and used as spiked-in mRNA for normalization of the transcripts . After reverse transcription of the purified RNA ( Maxima First Strand cDNA synthesis kit , Thermo Scientific ) quantitative PCR ( RT-qPCR ) was performed using gene specific primer pairs to quantify mRNAs of PMT1 , PMT4 , PMT6 and ACT1 . For each fraction three technical replicates were assayed on a Mx3000P LightCycler ( Stratagene ) , with 10 μl of cDNA , 4 μl EvaGreen QPCR-mix II ( Bio-Budget ) and 3 μl each of forward and reverse oligonucleotide primers ( 400 pmol/μl ) in each reaction . The polymerase was activated at 95°C for 10 min , annealing was performed at 60°C for 15 s , extension at 72°C for 30 s and the denaturation step was performed at 95°C for 30 s in a total of 50 cycles . For the statistical assessment of the difference between the transcript distributions in the reference strain CAF2-1 and the dom34 mutant strain , the Kolmogorov-Smirnov test ( KS-test ) was performed [31] . The KS-test was computed using the ks . test CRAN package in the R statistical software environment .
|
Fungi respond to damages of their glycostructures in their cell wall by transcriptional upregulation of genes that specify compensatory activities . Upon block of protein N-glycosylation , the human fungal pathogen Candida albicans increases transcription of PMT1 encoding a major isoform of protein O-mannosyltransferase . Here we demonstrate that the Dom34 protein aids in glycostress responses by upregulating the translation of several PMT isoform transcripts . Dom34 has previously been implicated in mechanisms to secure high levels of ribosomal subunits that promote translation in general , e . g . by no-go decay at the 3′-UTR of transcripts . By binding to the 5′-UTR and activating translational initiation of PMT transcripts we add a novel mode of action and suggest a preferred class of targets for the translational activities of the Dom34 protein . The combination of transcriptional and Dom34-mediated translational upregulation of PMT genes optimizes effective recovery and survival of fungal cells upon glycostress .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"and",
"Materials"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"nucleases",
"enzymes",
"pathogens",
"messenger",
"rna",
"microbiology",
"dna-binding",
"proteins",
"enzymology",
"nucleotides",
"fungi",
"model",
"organisms",
"untranslated",
"regions",
"molecular",
"biology",
"techniques",
"fungal",
"pathogens",
"research",
"and",
"analysis",
"methods",
"saccharomyces",
"mycology",
"proteins",
"medical",
"microbiology",
"ribonucleases",
"microbial",
"pathogens",
"denaturation",
"gene",
"expression",
"molecular",
"biology",
"yeast",
"biochemistry",
"rna",
"candida",
"hydrolases",
"nucleic",
"acids",
"phenotypes",
"oligonucleotides",
"protein",
"translation",
"genetics",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"saccharomyces",
"cerevisiae",
"rna",
"denaturation",
"organisms",
"candida",
"albicans"
] |
2016
|
Dom34 Links Translation to Protein O-mannosylation
|
MnmE , which is involved in the modification of the wobble position of certain tRNAs , belongs to the expanding class of G proteins activated by nucleotide-dependent dimerization ( GADs ) . Previous models suggested the protein to be a multidomain protein whose G domains contact each other in a nucleotide dependent manner . Here we employ a combined approach of X-ray crystallography and pulse electron paramagnetic resonance ( EPR ) spectroscopy to show that large domain movements are coupled to the G protein cycle of MnmE . The X-ray structures show MnmE to be a constitutive homodimer where the highly mobile G domains face each other in various orientations but are not in close contact as suggested by the GDP-AlFx structure of the isolated domains . Distance measurements by pulse double electron-electron resonance ( DEER ) spectroscopy show that the G domains adopt an open conformation in the nucleotide free/GDP-bound and an open/closed two-state equilibrium in the GTP-bound state , with maximal distance variations of 18 Å . With GDP and AlFx , which mimic the transition state of the phosphoryl transfer reaction , only the closed conformation is observed . Dimerization of the active sites with GDP-AlFx requires the presence of specific monovalent cations , thus reflecting the requirements for the GTPase reaction of MnmE . Our results directly demonstrate the nature of the conformational changes MnmE was previously suggested to undergo during its GTPase cycle . They show the nucleotide-dependent dynamic movements of the G domains around two swivel positions relative to the rest of the protein , and they are of crucial importance for understanding the mechanistic principles of this GAD .
Cells devote substantial biosynthetic effort and resources to posttranscriptional modification of tRNAs [1] . A frequent feature of tRNAs in all domains of life are modified nucleosides in the anticodon region and especially at the wobble position ( position 34 ) [2] , which prestructure the anticodon domain to insure correct codon binding during translation [3] . MnmE is an evolutionary conserved G protein found in bacteria , fungi , and humans , which together with the protein GidA catalyzes the formation of a carboxymethylaminomethyl-group ( cmnm ) at the 5 position of the wobble uridine ( U34 ) of tRNAs reading 2-fold degenerated codons ending with A or G , i . e . , tRNAArg ( UCU ) , tRNAGln ( UUG ) , tRNAGlu ( UUC ) , tRNALeu ( UAA ) , and tRNALys ( UUU ) [4]–[6] . This modification ( cmnm5U34 ) together with a thiolation at the 2 position favours the interaction with A and G , but suppresses base-pairing with C and U [3] , [7]–[10] . By controlling rare codon recognition and reading frame maintenance , hypermodified U34 moreover plays a regulatory role in gene expression [11] . Eucaryotic homologues of MnmE and GidA ( termed MSS1 and Mto1 , respectively , in yeast ) are targeted to mitochondria [12] , [13] , and the human homologues ( termed hGTPBP3 and Mto1 , respectively ) have been implicated in the development of severe mitochondrial myopathies such as MERRF ( myoclenic epilepsy ragged red fibres ) , MELAS ( mitochondrial encephalomyopathy lactic acidosis stroke ) , and nonsyndromic deafness [14]–[18] . The crystal structure of MnmE from Thermotoga maritima reveals a three-domain protein consisting of an N-terminal tetrahydrofolate-binding domain , a central helical domain , and a canonical Ras-like G domain inserted into the helical domain [19] . The asymmetric unit of these crystals contained one MnmE molecule and the N-terminal domain of a second proteolysed MnmE chain interacting with the N-terminal domain of the first molecule , suggesting that MnmE is a dimer in solution ( Figure 1A ) [19] . By superposition of the first MnmE chain on the second N-terminal domain a model for the full-length homodimer was generated in which the two G domains face each other with a distance of almost 50 Å between the two P-loops ( Figure 1A ) [19] . In contrast to Ras-like small G proteins that require a guanine nucleotide exchange factor ( GEF ) protein to drive the nucleotide exchange and a GTPase activating protein ( GAP ) to stimulate hydrolysis [20] , [21] , MnmE displays lower affinities towards nucleotides and a higher intrinsic K+-stimulated GTP hydrolysis [19] , [22]–[24] . A G domain dimerization across the nucleotide binding site has been proposed on the basis of biochemical data and the crystal structure of the isolated MnmE G domains in complex with GDP-aluminium tri- or tetrafluoride ( AlFx ) ( a mimic of the transition state of GTP hydrolysis [25] ) [22] . The G domains dimerize via their switch regions to position an invariant Glu-residue ( E282 ) for optimal orientation of a water molecule for the nucleophilic attack of the γ-phosphate group [22] . Dimerization stabilizes a highly conserved loop in switch I , the so-called K-loop , to coordinate K+ in a position analogous to the positive charge of the arginine finger in the Ras-RasGAP system . This explains why K+ is required both for the GTPase stimulation and for G domain dimerization [22] . On the basis of the common feature that the G domain cycle is regulated by homodimerization , MnmE has been categorized as G protein activated by nucleotide-dependent dimerization ( GAD ) [26] , together with the signal recognition particle ( SRP ) and its receptor ( SR ) [27] , [28] , the regulator of Ni insertion into hydrogenases HypB [29] , the dynamins [30] , the human guanylate binding protein hGBP1 [31] , the chloroplast import receptors Toc33/34 [32] , [33] , the septins [34] , and the Roc-COR tandem found to be mutated in Parkinson disease [35] . It has been postulated that nucleotide-dependent G domain dimerization activates the GTPase and the distinct biological functions of these proteins , although the mechanisms of coupling G domain dimerization to biological function within this class are diverse and incompletely understood [26] . So far , neither the structural model of the full-length MnmE dimer nor dimerization of the G domains in the context of the full-length dimer have been proven directly . With the architecture of the proposed dimer model , dimerization of the G domains would require large domain movements suggesting that large conformational rearrangements of the protein are coupled to its GTPase cycle [22] . Here we study these GTPase-coupled rearrangements by trapping the protein in various steps of its GTPase cycle by X-ray crystallography and pulse double electron-electron resonance ( DEER ) spectroscopy in combination with site-directed spin labeling [36]–[38] . The distance distributions obtained for spin labeled sites in the G domains of MnmE allow us to characterize the G domain movements during the GTPase cycle of MnmE .
Various MnmE homologous have been screened for crystallization conditions in the presence of GDP , GDP-AlFx and guanosine-5′- ( β , γ-methylene ) triphosphate ( GppCp ) , and K+ and were found to crystallize readily in diverse conditions , but only in three cases—Chlorobium tepidum MnmE ( CtMnmE ) in the presence of K+ , GDP , or GDP-AlFx; Nostoc MnmE ( NoMnmE ) in the presence of K+ , GDP , or GDP-AlFx; and CtMnmE in the presence of K+ and GppCp-crystals with sufficient diffraction quality were obtained . In the case of CtMnmE , a polyethylene glycol ( PEG ) 6000/NaCl-condition produced diffraction quality crystals in the presence of GPD and GDP-AlFx . Crystals had the same unit cell parameters and the same space group and are thus isomorphous . NoMnmE crystals with sufficient diffraction were obtained in a PEG 550 monomethyl ether ( MME ) condition . As with CtMnmE , crystals obtained in the presence of GDP-AlFx or GDP were isomorphous . Structure determination showed in both cases that the crystals contained the GDP-bound form of MnmE , despite the presence of AlFx . Quality of crystals grown in the presence of GDP-AlFx were somewhat better , hence their datasets were used for structure determination . CtMnmE·GDP and NoMnmE·GDP ( grown in presence of AlFx ) crystallized in the space groups I4 ( 1 ) 22 and P4 ( 3 ) 2 ( 1 ) 2 , respectively , each with one full length protomer in the asymmetric unit . In both cases homodimers are formed via crystallographic symmetry by means of the N-terminal domains ( Figure 1B and 1C ) . Apart from the location of G domains , the structure is very similar to the dimer model proposed for nucleotide-free MnmE ( Figure 1A ) [19] . Strikingly , two molecules of 5-formyl-tetrahydrofolate ( 5-F-THF ) were identified in the structure of NoMnmE·GDP , which were apparently copurified from the bacterial expression system ( Figure S1A ) . This suggests a high affinity for 5-F-THF and supports the recently proposed enzymatic mechanism whereby the C1 group of the cmnm modification is donated by THF [19] , [39] . The cofactor is bound as previously described for the complex prepared in situ [19] , with two folate binding sites within the dimer interface of the N-terminal domains . CtMnmE·GDP crystals were incubated with a 5-F-THF-containing cryoprotectant prior to data collection and in the crystal structure 5-F-THF is found in identical positions as in the NoMnmE·GDP-dimer ( Figure S1B ) and in the TmMnmE-dimer . In the case of CtMnmE·GppCp , the crystallographic asymmetric unit contained three protomers ( chains A , B , C ) . Molecules B and C form a dimer within the asymmetric unit , while protomer A forms a dimer with its crystallographic symmetry mate ( shown in Figure 1D ) . No density is found for the G domain of molecule C , but crystals applied on an SDS-page confirmed an intact protein ( unpublished data ) . Thus two dimeric structures of CtMnmE·GppCp were analyzed , i . e . , the dimer generated by protomer A and a symmetry related chain A ( termed “dimer A” ) and the dimer generated by protomer B and a second protomer B docked onto chain C ( termed “dimer B” ) . The overall homodimer architecture found in the three structures resembles the proposed model obtained from a partial dimer ( Figure 1A ) , with the G domains facing each other with their nucleotide binding sites ( Figure 1B–1D ) . However , even though triphosphate analogues such as GppCp or AlFx and GDP were used in the crystallization trials , the G domains were separated from each other by large distances . They do not display any structural contacts between each other nor to the N-terminal or helical domains . In all the structures , nucleotides are far apart from each other , with distances of 38 to 56 Å between the first P-loop glycines' Cα atom ( GxxxxGKS motif ) . In each structure , the G domain adapts the canonical Ras-fold with either both switch regions ( CtMnmE·GDP , CtMnmE·GppCp ) or switch II ( NoMnmE·GDP ) disordered and thus not resolved . Nucleotides are bound in a way typical for Ras-like G domains ( Figure S1C–S1E ) . In CtMnmE·GDP however , no Mg2+ is coordinated to the phosphates , and switch I-contacts to GDP are absent ( Figure S1C ) . In NoMnmE·GDP , two Zn2+ atoms from the crystallisation condition , localized by their anomalous signal , are coordinated to the G domain . One of these is coordinated to helix Gα4 and is involved in crystal contacts ( see below ) , the other occupies the usual Mg2+-binding site at the β-phosphate of GDP ( Figure S1D ) . As Switch I is resolved , but does not contact the bound GDP and since there is no indication for a physiological role of Zn2+ , we consider this to be a crystallographic artefact also observed in the nucleotide binding pockets of other small G proteins [40] . For conventional G proteins regulated by GAPs [20] as well as for G proteins activated by dimerization [26] , AlFx-in the γ-phosphate binding site mimics the transition state of the phosphor transfer reaction and is considered the litmus test for correct assembly of the active site . In the case of MnmE , this is thought to be achieved by dimerization and close juxtaposition of the two G domains across the nucleotide binding site , as observed for the isolated G domains [22] . Although both GDP-bound structures have been obtained using GDP and AlFx in the crystallization trials , no electron density for AlFx could be observed ( Figure S1C and S1D ) . One would thus conclude that close contact between the G domains is not possible in the full-length protein or that the G domains are too mobile for fixation in the crystal and/or that the crystal lattice forces do not allow the close state to occur . Another possibility would be that crystallisation conditions with high concentrations of precipitants inhibit formation of the closed state of the G domains . Indeed we can show by a previously established fluorometric assay , by which an increase of the fluorescence of 2′-/3′-O- ( N′-methylanthraniloyl ) -GDP ( mGDP ) bound to MnmE upon addition of AlFx in the presence of K+ is attributed to G domain dimerization [22] , that in the presence of any of the precipitants used for crystallisation , dimerization of the G domains is severely inhibited in the full length protein ( Figure S2 ) . This explains why despite the presence of AlFx in the crystallisation trials only the GDP-bound conformations are found . In the crystals , the G domains are thus trapped in an open state that does not allow tight binding of AlFx , into the γ-phosphate binding site . Superposition of the five available homodimer structures ( CtMnmE·GDP , NoMnmE·GDP , CtMnmE·GppCp dimers of molecules A and B , T . maritima MnmE dimer model , generated with pdb 1XZP ) reveals that the N-terminal domains align quite well and only minor displacements are present for the helical domains ( Figures 2A and S3; Table 1 ) . Strikingly , the superposition shows large rotational and translational displacements of the G domains ( Figures 2A–2C and S3 ) , which are reflected in their higher root mean square deviation ( RMSD ) values ( Table 1 ) leading to separation of nucleotide binding sites between , for instance , CtMnmE·GDP and NoMnmE·GDP by 18 Å ( Cα-Cα distance of the first P-loop glycines ) ( Figure 2A ) . This becomes clearly visible in the displacement of the G domain β-sheets and of helix Gα6 ( Figure 2B and 2C ) . A video generated from the five homodimer structures makes the drastic displacements of the G domains evident and highlights the dynamic character of the G domains ( Video S1 ) . The different orientations indicate that the G domains are highly flexible with regard to the rest of the protein probably due to the rather loose connections between G and helical domains . A conserved glycine residue is situated between helix Hα5 and the first strand of the G domain β-sheet ( Figure 2C ) , which because of its higher conformational freedom could function as a hinge ( “N-hinge” ) . A second hinge point ( “C-hinge” ) is where a not-well-ordered loop attaches the C-terminal end of the G domain after Gα6 to the helical domain ( Figure 2B ) . The angle by which Gα6 is shifted spans up to 47° . In the crystal structure of CtMnmE·GppCp this loop region is not resolved underlining its high flexibility . Although crystals grew under many more conditions , crystals diffracting to reasonable resolution were only obtained in the cases reported here . This result is most likely due to the fact that in these cases , crystal contacts trap the G domains in defined orientations ( Figure S4 ) , whereas in the weakly diffracting crystals the G domains are only loosely packed causing lattice disorder . The G domains in the CtMnmE·GDP and NoMnmE·GDP structures pack against symmetry mates with contact areas of 376 Å2 and 488 Å2 . In NoMnmE·GDP a Zn2+-ion tightly links the G domain to symmetry mates ( Figure S4A ) , while in CtMnmE·GDP the G domains fix each other by a toothing upside-down arrangement ( Figure S4B ) . Crystal contacts of G domains A and B in the CtMnmE·GppCp structure comprise areas of 845 Å2 and 987 Å2 , respectively . Docking the G domain of molecule B ( or A ) into the asymmetric unit of the CtMnmE·GppCp structure to the position expected for the G domain of molecule C would create a much smaller hypothetical crystal contact area of only 18 Å2 ( or 131 Å2 ) . Thus we would expect that the G domain of molecule C is present in the crystal but , due to its high mobility and absence of sufficient crystal contacts , is not visible in the electron density map . This is similar to the recent structure of the Roco protein , which is also a GAD protein . There , the second G domain of the constitutive dimer is present in the crystal but can not be identified in the electron density map [35] . To test whether the “open” G domain arrangement found in our GDP- and GppCp-bound structures is representative for the conformation in solution and to identify and characterize the putative transition state with closed G domains , which could not be obtained by crystallization , we applied four-pulse DEER spectroscopy [36]–[38] , to measure distances between nitroxide spin labels in the G domains of full-length EcMnmE in different steps of the GTPase cycle . Positions mutated to cysteine for spin labeling with ( 1-oxyl-2 , 2 , 5 , 5-tetramethyl-3-pyrroline-3-methyl ) methanethiosulfonate spin label ( MTSSL ) are Glu287 , close to the top of the G domain in Gα2 , Ser278 in switch II , and Asp366 , located in Gα6 , and , as shown in Figure 3 , result in the introduction of two symmetry-related spin labels in the functional MnmE dimer . As a possible “negative control” we also spin labeled position Ile105 in the N-terminal domain , for which no distance changes are expected . The Cβ-Cβ distances between these sites derived from the structures of the open and the model of closed state are listed in Table 2 . To avoid unwanted side effects of cysteine substitutions , only nonconserved , surface-exposed residues have been selected . Furthermore , mutant proteins were assayed for K+-stimulated GTPase activity with and without attached MTSSL-label . No impairment of GTPase activity in comparison to wild type could be observed by the mutation itself or the introduction of the spin label ( Table S1 ) . Since efficient GTPase activity in the presence of K+ is strictly dependent on correct K+-binding and G domain dimerization [22] , we can conclude that the structural and functional aspects of G domain dimerization and GTPase activity of the mutants are preserved in the proteins used for DEER . Figure 4A illustrates the results of the DEER measurements in the presence of 100 mM KCl , where the left panel shows the background-corrected dipolar evolution data , the centre panel the respective dipolar spectra , and the right panel the corresponding distance distributions ( obtained by Tikhonov regularization; see Methods ) , which are summarized in Table 2 . The DEER analysis of mutant E287R1 ( R1 denotes the MTSSL side chain ) , close to the top of the G domain in Gα2 , indicates one major peak centered at a distance of 55 Å for the apo- and 53 Å for the GDP-bound state . This distances correspond well to the Cβ-Cβ distances in the TmMnmE crystal structure model of 53 Å ( the corresponding residues in the CtMnmE and NoMnmE structures are not resolved ) and is therefore in agreement with an open conformation of the G domains . For D366R1 ( situated at Gα6 ) , a well-defined interspin distance distribution centered at 67 Å in the apo- state and 65 Å in the GDP-bound state could be observed in good agreement with the distances obtained from the TmMnmE dimer model ( 62 Å ) and NoMnmE·GDP ( 63 Å ) , suggesting again an open conformation of the G domains . The corresponding Cβ-Cβ distance in CtMnmE·GDP dimer is somewhat shorter ( 57 Å ) , which is due to the different orientation of G domains in this structure ( Figure 2A ) and to the different tilting of Gα6 ( Figure 2B ) . From the E287R1 and D366R1 data in the apo- and GDP-bound states , we conclude that instead of a continuum of freely moving orientations , the MnmE G domains seem to have defined major orientation reflected by the distance distributions . In contrast , the analysis for S278R1 ( switch II region ) by Tikhonov regularization did not allow discrimination between a continuum of distances ranging from 25 Å to 50 Å with increasing probabilities for larger distances ( shown in dark colours ) or three to four distinct distances corresponding to different protein and/or spin label conformers ( shown in pale colours ) . To clarify this issue , we additionally fitted the GDP data with a Monte Carlo/SIMPLEX algorithm assuming a sum of Gaussian-distributed conformers contributing to the dipolar evolution data ( Figure 4B ) [41] . The experimental data were satisfactorily reproduced by a distance distribution with two Gaussian populations , which are well defined as judged by the χ2 surfaces , summing up to a broad distribution in the range 30–50 Å . Possible explanations for such a continuum in the distance distribution could be ( i ) that the labeled position is located in the switch II region , which is flexible in the free and GDP-bound states , in line with the X-ray results , or ( ii ) that the spin label side chains are not restricted in their conformational space and populate multiple rotamers , or ( iii ) a combination of ( i ) and ( ii ) . A general continuum of G domain orientations can be excluded from the results for positions E287R1 and D366R1 . Control measurements of K+-stimulated GTPase activity ( Table S1 ) make severe structural perturbations appear unlikely . Instead the deviation from the Cβ-Cβ distance of 22 Å in the TmMnmE dimer model is probably due to a switch II conformation induced by crystal packing forces . It has been observed before , that even in structures of the same G protein-nucleotide complex different switch II conformations were induced by crystal packing forces [42] . Nevertheless , the most pronounced distances between 40–50 Å as well as the minor fractions situated between 30 and 40 Å observed by DEER are in strong agreement with an open state of the G domains as observed in the apo- and GDP-bound crystal structures . In the presence of the nonhydrolizable GTP analogue guanosine 5′-imidotriphosphate ( GppNHp ) the distance distributions comprise two fractions with different interspin distances for all three labeled positions . One larger distance ( E287R1 , 55 Å; D366R1 , 63 Å; and S278R1 , 43 Å ) corresponds to the open state of the G domains as observed for the nucleotide-free and GDP-bound forms , whereas the other distance , contributing about 30% to the distance distribution ( average value calculated from the area under the distance distribution curve ) is characterized by significantly shorter distances ( E287R1 , 37 Å; D366R1 , 47 Å; and S278R1 , 27 Å ) , clearly indicating the presence of a second conformation , where the two G domains are in close proximity . As for the GDP-bound state , the GppNHp data for S278R1 were additionally fitted assuming a sum of Gaussian distributions . Despite differences especially in the distribution width for the two populations , this approach also reveals the presence of the two conformations of the G domains . In the X-ray structure of the AlFx-complexed G domain dimer , the Cβ-Cβ distance of the S278- and E287-pair are 18 Å and 28 Å , respectively and thus somewhat shorter as compared to the GppNHp DEER data ( S278R1 , 27 Å; E287R1 , 37 Å ) . However the MTSSL-side chain itself has an average length of 7 Å between the nitroxyl-radical and the Cβ-atom [43] . This can increase the measured distance up to 14 Å for a pair of MTSSL side chains , depending on their rotamer orientation . The longer distances of the short distance maxima in the GppNHp-distance distributions of S278R1 and E287R1 measured in solution are thus most likely the result of a closed conformation of G domains , where the MTSSL side chains protrude away from the symmetry axis of the G domain dimer . Overall , the GppNHp measurements lead us to conclude that in the presence of GppNHp two conformations are in thermal equilibrium . In the crystal structure of GppCp-bound MnmE the G domains are found in the open state , indicating that this equilibrium is shifted towards the open state under the crystallization conditions . In the presence of the transition state mimic GDP-AlFx , S278R1 and E287R1 show a single population maximum , with defined distances of 28 Å and 36 Å , respectively , in line with a closed conformation ( Figure 4 ) . The observed distances are close to the observed Cβ-Cβ distances in the crystal structure of the GDP-AlFx–bound G domain dimer structure ( S278 , 18 Å; E287 28 Å ) , with deviations due to spin label conformations as discussed above . Compared to the distances characterizing the closed conformation in the presence of GppNHp , the distance distributions for the transition state mimic are sharper and the maxima are slightly shifted . For E287R1 it decreases by about 1–2 Å and for position S278R1 the broad distribution between 20 and 30 Å converts to a more defined but asymmetric distribution with a major distance of 28 Å , which is well reproduced also by the Monte Carlo approach ( Figure 4B ) . For position D366R1 two major fractions with inter spin distances of 58 and 48 Å are visible , presumably due to two different rotamer populations of the spin label side chain . The maximum at 48 Å corresponds nicely to the Cβ-Cβ distance in the GDP-AlFx–bound G domain structure , whereas the 58Å distance likely represents an MTSSL-rotamer population pointing away from each other . As is obvious from the distance distributions for the GppNHp and the GDP-AlFx state , the closed state in the presence of GDP-AlFx slightly differs from that in the presence of GppNHp suggesting that on the reaction pathway from the triphosphate state to the GTPase competent conformation further rearrangements in the active site of the G domains take place . Overall the distance maxima are shifted to shorter distances in the GDP-AlFx state as compared to the apo- , GDP- and GppNHp distances . This shows that the G domains adapt a closed conformation as observed in the GDP-AlFx-complexed G domain structure . To explore whether G domain dimerisation leads to domain rearrangements in the N-terminal dimerization domain , a spin label was introduced at position Ile105 ( Figure 3A ) . A comparison of the distance distributions obtained for the GDP state ( open conformation ) and GDP-AlFx state ( closed conformation ) does not show any significant differences concerning the major population in the distance distribution with an average distance of 29 Å for both nucleotide states ( Figure 4A; Table 2 ) , indicating , that closing of the G domains does not significantly disturb the overall integrity of the N-terminal domains . The deviation to the corresponding Cβ-Cβ distances in the various dimer models ( 36 Å , 37 Å ) are likely due to spin label rotamer conformations . Previous studies have shown K+ ions to activate the MnmE GTPase . This follows from the finding that dimerization of the MnmE G domains and GDP-AlFx complex formation strictly require K+ , which is bound in the dimer interface ( Figure 3B ) , such that its position overlaps with that of an Arg finger required for the GAP-mediated GTP hydrolysis on Ras-like G proteins [20] , [22] . Moreover , GTPase activity and AlFx-induced dimerization are at least partially stimulated by cations with an ionic radius comparable to K+ ( 1 . 38 Å ) such as Rb+ ( 1 . 52 Å ) and , to a lesser extent , NH4+ ( 1 . 44 Å ) , whereas Na+ ( 0 . 99 Å ) and Cs+ ( 1 . 67 Å ) do not show this effect [22] . Consistent with this , Rb+ and NH4+ were also found to be coordinated to the K+ binding site in two MnmE G domain dimer structures GDP complexed with AlFx ( pdb 2GJ9 and 2GJA ) [22] . To analyze the cation dependency of G domain dimerization in full-length protein in solution , we determined distance distributions for the sites S278R1 and E287R1 in the apo , GDP , GDP-AlFx , and GppNHp bound state in the presence of various cations , i . e . , Na+ , K+ , Rb+ , Cs+ , and for S278R1 additionally in the presence of NH4+ for the GDP and GDP-AlFx state ( Figure 5 ) . The distance distributions and dipolar time traces show that in the presence of GDP-AlFx only K+ is capable for shifting the equilibrium completely towards the closed G domain dimer . The ability of the respective cations to stabilize G domain dimerization follows the order K+>Rb+>NH4+>Cs+≈Na+ , clearly correlated with their ionic radii and their ability to stimulate GTP hydrolysis [22] . In the presence of GppNHp , we observe the same order of cations with regard to their capability for shifting the equilibrium towards the closed state . Notably , Cs+ , which is completely unable to stabilize G domain dimerization , seems to have an influence on switch II conformational dynamics and on the overall orientation of the G domains , as seen from the significantly broadened and shifted distance distributions compared to those for the other cations .
Understanding how GADs use the GTPase cycle as the driving force to perform a variety of functions like insertion of signal sequences into the ER translocon by the SRP/SR system [44] , tRNA modification by MnmE [19] , [22] , kinase activation by the Parkinson kinase LRRK2 [45] , or metal ion delivery to hydrogenases [46] is a crucial step for elucidating the diverse mechanism by which these proteins operate . Although within this class of proteins MnmE is one of the structurally and biochemically best characterized and a model for the GTPase cycle dependent G domain dimerization has been proposed [22] , neither the structural model of the full-length MnmE dimer nor dimerization of the G domains in the full-length dimeric protein have been proven directly . Here we have applied a combined approach of X-ray crystallography and pulse electron paramagnetic resonance ( EPR ) spectroscopy to study the behavior of the G domains in full-length MnmE in different steps of the GTPase cycle . We were able to solve the first X-ray structures of full-length MnmE in complex with nucleotides . The structures confirm the previously postulated homodimer model [19] according to which MnmE constitutively dimerizes via its N-terminal domain whereas the G domains , separated by a large distance of approximately 48 Å ( measured from Cα of the first glycine of the P-loop ) , face each other with their nucleotide binding sites . The distance distributions obtained by DEER of MnmE in the apo , GDP , and GppNHp state reveal that the G domains are far apart also in solution excluding that the open conformations in the crystal structures are crystallographic artefacts . Comparison of the different full-length structures reveals that the G domains are present in drastically different orientations suggesting them to be highly mobile elements capable of moving independently with regard to the other domains . As judged from the X-ray structure , they need to overcome a 20–30-Å distance gap on formation of the GDP-AlFx complex [22] . In contrast to the X-ray data , the DEER distance distributions suggest the presence of one defined orientation for the open state in solution , arguing that the different G domain orientations in the X-ray structures result from crystal packing forces . That reasonable diffraction data can only be obtained when the G domains are stabilized by packing interaction is a further indication for their high mobility . Moreover , for CtMnmE we find different orientations between the GDP- and GppCp-bound structures and even between different molecules in one asymmetric unit of CtMnmE·GppCp . Although the crystals for all structures presented here were grown in the presence of K+ and triphosphate or a transition state mimic to induce G domain dimerization , the structures show the G domains in an open state , suggesting that the closed state is not stable under crystallization conditions . We can demonstrate indeed using a fluorometric assay with mant-GDP , that close juxtaposition of G domains with AlFx is inhibited in the presence of crystallization precipitants . The interspin distances between the spin labeled G domains obtained by DEER directly prove for the first time that the G domains contact each other in the presence of triphosphate or transition state analogs . A notable feature of the GppNHp-bound state is the coexistence of an open and closed state , pointing out that a triphosphate analog is not sufficient to fully stabilize the closed state . However , recently a stabilizing effect of GidA on the closed state of the G domains was shown , indicating that regulation of the MnmE G domain cycle is coupled to other components of the tRNA-modification system [6] . Unlike the results from X-ray structures , the EPR data , under low salt and in the absence of PEG , do not show a continuum of conformations but rather particular conformations in the open and closed state not observable in the X-ray experiment . We further show that only the presence of GDP , AlFx and K+ is capable of stabilizing the closed state , and that this effect is specific , since the effect is absent with Na+ and Cs+ and is smaller with similar size cations such as Rb+ and NH4+ . In summary , we were able to directly demonstrate the conformational changes MnmE was suggested to undergo during its GTPase cycle [6] , [19] , [22] , [39] , [47] . Dimerization of the MnmE G domains is accompanied by large domain movements of up to 20 Å from the open to the closed state , which is an apparently unique feature of MnmE with regard to other GADs , suggestive for a functional or regulatory coupling of these domain movements to the tRNA-modification reaction . For the architecturally similar Roc-COR tandem ( see above ) , the G domains in the nucleotide free state are already in close proximity [35] , rendering similar extensive domain rearrangements unlikely . Yet such drastic rearrangements are not untypical for NTPases , as for example Hsp90 , which constitutively dimerizes via its C-terminal domain , undergoes dramatic domain movements during its ATPase cycle involving juxtaposition of its N-domains in the triphosphate state [48] , [49] . MnmE forms a heterotetrameric complex with GidA [50] , which is stabilized in the triphosphate state [6] , [39] , and tRNA modification was suggested to be exerted by this complex rather than by the individual proteins [5] , [50] , which was recently proven by an in vitro modification assay [6] . Furthermore active GTP-turnover rather than simple GTP-binding was shown to be essential for the modification reaction [6] , [47] and in particular , nucleotide dependent G domain dimerization is tightly coupled to the tRNA-modification process both in vitro and in vivo [6] . According to a proposed reaction mechanism , the reaction itself does not require energy , but rather comprises several steps at presumably different , spatially separated active sites , requiring tight regulation [19] , [39] . We thus speculate that G domain dimerization during GTP hydrolysis is required for orchestration of the multistep tRNA-modification reaction [6] . The exact link between G domain dimerization , GTP hydrolysis , conformational changes , and tRNA modification is focus of current investigations .
C . tepidum and Nostoc sp . 7120 MnmE ( CtMnmE , NoMnmE ) were cloned into pET14b ( Novagene ) and expressed as N-terminal His-tagged proteins in Escherichia coli BL21-DE3 . Cells were lysed in 50 mM Tris ( pH 7 . 5 ) , 100 mM NaCl , 5 mM MgCl2 ( = buffer A ) with 20 mM imidazole , 5 mM β-mercaptoethanol , 150 µM PMSF , and the proteins were purified by Ni-NTA , thrombin-cleavage of the His-Tag , and gel filtration on Superdex 200 in buffer A with 5 mM dithioerythritol ( DTE ) . Cloning , expression , and purification of E . coli MnmE and mutants and preparation of nucleotide-free MnmE was carried out as described elsewhere [39] . Crystals were obtained by hanging-drop vapour diffusion . For CtMnmE·GDP crystals , 1 µl each of 50 mg/ml protein in 50 mM Tris ( pH 7 . 5 ) , 100 mM KCl , 5 mM MgCl2 , 5 mM DTE ( buffer B ) plus 5 mM GDP , 5 mM AlCl3 , 50 mM NaF , and precipitant ( 100 mM Tris-HCl [pH 8 . 5] , 2 . 250 M NaCl , 15% [w/v] PEG 6000 ) were mixed . After 3 d the reservoir was changed to 100 mM Tris-HCl ( pH 8 . 5 ) , 2 . 250 M NaCl , 30% PEG 6000 , and equilibrated for 2 more days . Crystals were soaked with precipitant supplemented with 12% glycerol and 5 mM 5-F-THF for 30 min and flash-frozen in liquid nitrogen . For NoMnmE·GDP , 1 µl of 20 mg/ml protein in buffer B with 5 mM GDP , 5 mM AlCl3 , 50 mM NaF , and precipitant ( 100 mM Tris [pH 7 . 5] , 22% [w/v] PEG 550 MME , 10 mM ZnSO4 ) were mixed and grown at 20°C . After 2 d crystals were cryo-dipped into reservoir solution with 28% ( w/v ) PEG 550 MME and flash-frozen into liquid nitrogen . For CtMnmE·GppCp , 40 mg/ml nucleotide free protein in buffer B with 5 mM GppCp was mixed ( 1∶1 ) with 100 mM MES , 46 mM NaOH , 12% PEG 4000 , 40 mM NaCl , and crystals were grown at 20°C . After 2–3 d , crystals were flash-frozen in reservoir containing 20% glycerol . All datasets were collected at 100 K on beamline PX2 ( SLS , Villingen ) at wavelengths of 0 . 98003 , 0 . 9796 , and 1 . 28186 Å ( Zn2+-edge ) for CtMnmE·GDP , CtMnmE·GppCp , and NoMnmE·GDP , respectively . All datasets were processed , indexed , and scaled with XDS [51] . Initial phases were obtained by molecular replacement with the N-terminal and the helical domain of T . maritima MnmE ( pdb 1XZP ) with MOLREP [52] . Coot [53] and REFMAC [54] , [55] were used for model building and translation , libration , screw rotation ( TLS ) -refinement including NCS restraints and NCS-averaged maps in the case of CtMnmE·GppCp . Crystallographic simulated annealing of models was carried out with CNS [56] . Structural representations were prepared with pymol ( www . pymol . org ) . For NoMnmE·GDP , Zn2+ atom positions were located by their anomalous signal . For CtMnmE·GppCp , a positive peak in the FO-FC-map close to the β- and γ-phosphate in the nucleotide binding site of G domain A was assigned to Mg2+ , on the basis of its position at the usual Mg2+-site in G protein structures . Structures were analyzed by PROCHECK [57] revealing for all three structures 100% of torsion angles within the allowed Ramachandran regions . Data collection and refinement statistics are listed in Table 3 . Structures were aligned with coot [53] and Superpose of the CCP4-package [54] . Crystal contact areas were calculated using the PROTORP server [58] . 10 µM of nucleotide-free E . coli MnmE loaded with 0 . 5 µM of mGDP were incubated in 50 mM TriS-HCl ( pH 7 . 5 ) , 100 mM KCl ( or NaCl ) , 5 mM MgCl2 , 10 mM NaF with or without the precipitants 15% PEG 6000 , 2 , 250 mM NaCl , or both or 22% PEG 550 MME at 20°C . The fluorescence of mGDP bound to MnmE , excited at 366 nm and detected at 450 nm , was monitored over time in a Fluoromax 2 spectralfluorimeter ( Spex Industries ) . To initiate AlFx-complex formation , 1 mM AlCl3 was added and the fluorescence was continuously monitored . For analysis , fluorescence amplitudes were normalized to the amplitude before addition of AlCl3 . Purified , nucleotide-free Cys-mutants of E . coli MnmE-C451S ( Table 2 ) were pretreated with DTE ( 4°C ) . After removal of DTE protein solutions were incubated with 1–5 mM MTSSL ( Toronto Research , Alexis ) for 16 h ( 4°C ) . Excess MTSSL was removed by gel filtration . Labeling efficiencies have been determined to be >80% in all cases . GTPase reactions were started by adding 0 . 5 µM of wild type or mutant MTSSL-labeled or nonlabelled MnmE protein to 186 µM of GTP in 50 mM Tris-HCl ( pH 7 . 5 ) , 100 mM KCl , 5 mM MgCl2 , and performed at 20°C . At time points 0 , 1 , 2 , 3 , 5 , 7 , and 10 min aliquots were taken and analyzed for their nucleotide content by HPLC as described elsewhere [22] . For comparison , vapp was determined as the absolute value of the slope of a linear fit of GTP consumption over time , normalized to the total amount of enzyme for a range in which 10% of initial GTP was consumed . Pulse EPR experiments ( DEER ) were accomplished at X-band frequencies ( 9 . 3–9 . 4 GHz ) with a Bruker Elexsys 580 spectrometer equipped with a Bruker Flexline split-ring resonator ER 4118X-MS3 and a continuous flow helium cryostat ( ESR900 , Oxford Instruments ) controlled by an Oxford Intelligent temperature controller ITC 503S . Buffer conditions for the EPR experiments were 200–500 µM protein in 100 mM KCl ( or NaCl , RbCl , CsCl , NH4Cl ) , 50 mM Tris-HCl , 5 mM MgCl2 ( pH 7 . 4 ) with 5% ( v/v ) ethylene glycol ( for H2O buffer ) or 12 . 5% ( v/v ) glycerol-d8 ( for D2O buffer ) , and 1 mM GDP , 1 mM GppNHp or 1 mM GDP , 1 mM AlCl3 , 10 mM NaF , respectively . All measurements were performed using the four-pulse DEER sequence: [59] . A two-step phase cycling ( + 〈x〉 , − 〈x〉 ) was performed on . Time t′ is varied , whereas τ1 and τ2 are kept constant , and the dipolar evolution time is given by . Data were analyzed only for t>0 . The resonator was overcoupled to Q∼100; the pump frequency υpump was set to the center of the resonator dip and coincided with the maximum of the nitroxide EPR spectrum , whereas the observer frequency υobs was 65 MHz higher , coinciding with the low field local maximum of the spectrum . All measurements were performed at a temperature of 50 K with observer pulse lengths of 16 ns for π/2 and 32 ns for π pulses and a pump pulse length of 12 ns . Proton modulation was averaged by adding traces at eight different τ1 values , starting at and incrementing by . For proteins in D2O buffer with deuterated glycerol used for their effect on the phase relaxation , corresponding values were and . Data points were collected in 8-ns time steps or , if the absence of fractions in the distance distribution below an appropriate threshold was checked experimentally , in 16- or 32-ns time steps . The total measurement time for each sample was 4–24 h . Analysis of the data was performed with DeerAnalysis2006 . 1/2008 [60] . Additionally , the data was fitted assuming a sum of Gaussian distributed conformers utilizing the program DEFit 3 . 9 [41] , which employs a Monte Carlo/SIMPLEX algorithm to find a distance distribution to which the corresponding dipolar evolution function represents the best fit to the experimental data . Protein Data Bank ( PDB ) ( http://www . rcsb . org/pdb ) : Coordinates und structure factors have been deposited with accession codes 3GEE ( CtMnmE·GDP ) , 3GEI ( CtMnmE·GppCp ) , and 3GEH ( NoMnmE·GDP ) .
|
MnmE is an evolutionary conserved G protein that is involved in modification of the wobble U position of certain tRNAs to suppress translational wobbling . Despite high homology between its G domain and the small G protein Ras , MnmE displays entirely different regulatory properties to that of many molecular switch-type G proteins of the Ras superfamily , as its GTPase is activated by nucleotide-dependent homodimerization across the nucleotide-binding site . Here we explore the unusual G domain cycle of the MnmE protein by combining X-ray crystallography with pulse electron paramagnetic resonance ( EPR ) spectroscopy , which enables distance determinations between spin markers introduced at specific sites within the G domain . We determined the structures of the full-length MnmE dimer in the diphosphate and triphosphate states , which represent distinct steps of the G domain cycle , and demonstrate that the G domain cycle of MnmE comprises large conformational changes and domain movements of up to 18 Å , in which the G domains of the dimeric protein traverse from a GDP-bound open state through an open/closed equilibrium in the triphosphate state to a closed conformation in the transition state , so as to assemble the catalytic machinery .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biophysics/experimental",
"biophysical",
"methods",
"biochemistry/macromolecular",
"assemblies",
"and",
"machines",
"biochemistry/experimental",
"biophysical",
"methods"
] |
2009
|
Kissing G Domains of MnmE Monitored by X-Ray Crystallography and Pulse Electron Paramagnetic Resonance Spectroscopy
|
Careful monitoring for recrudescence of Wuchereria bancrofti infection is necessary in communities where mass drug administration ( MDA ) for the elimination of lymphatic filariasis ( LF ) as a public health problem has been stopped . During the post-MDA period , transmission assessment surveys ( TAS ) are recommended by the World Health Organization to monitor the presence of the parasite in humans . Molecular xenomonitoring ( MX ) , a method by which parasite infection in the mosquito population is monitored , has also been proposed as a sensitive method to determine whether the parasite is still present in the human population . The aim of this study was to conduct an MX evaluation in two areas of Bangladesh , one previously endemic district that had stopped MDA ( Panchagarh ) , and part of a non-endemic district ( Gaibandha ) that borders the district where transmission was most recently recorded . Mosquitoes were systematically collected from 180 trap sites per district and mosquito pools were tested for W . bancrofti using real-time PCR . A total of 23 , 436 intact mosquitoes , representing 31 species , were collected from the two districts , of which 10 , 344 ( 41% ) were Culex quinquefasciatus , the vector of W . bancrofti in Bangladesh . All of the 594 pools of Cx . quinquefasciatus tested by real-time PCR were negative for the presence of W . bancrofti DNA . This study suggested the absence of W . bancrofti in these districts . MX could be a sensitive tool to confirm interruption of LF transmission in areas considered at higher risk of recrudescence , particularly in countries like Bangladesh where entomological and laboratory capacity to perform MX is available .
Lymphatic filariasis ( LF ) , an important cause of acute and chronic morbidity worldwide , is caused by infection with the thread-like nematodes Wuchereria bancrofti , Brugia malayi and Brugia timori . The Global Programme to Eliminate Lymphatic Filariasis was established in 2000 by the World Health Organization ( WHO ) and has two objectives: ( i ) the interruption of LF transmission through mass drug administration ( MDA ) using the combination of albendazole plus diethylcarbamazine or ivermectin , or all three drugs together in specific contexts as recommended recently by WHO [1] and ( ii ) the alleviation of the suffering of affected populations through morbidity management and disability prevention [2] . Interruption of transmission is thought to require at least five rounds of MDA , after which national LF elimination programs conduct a Transmission Assessment Survey ( TAS ) to determine whether MDA can be stopped [3] . After MDA is ceased , programs must conduct surveillance to identify and respond to the possibility of re-emergence of transmission . Current WHO recommendations for post-MDA surveillance include repeating TAS twice at 2–3 year intervals after stopping MDA , and ongoing surveillance [3] . Detection of parasites in vector mosquitoes is one of the surveillance strategies that countries can consider . Molecular xenomonitoring ( MX ) , the use of PCR to identify parasite DNA in vector mosquitoes , has previously been used for LF surveillance after cessation of MDA [4–6] to identify residual foci of transmission . It has the advantage of being non-invasive to humans and could be useful when willingness of people to be tested is an issue , especially as households ( HH ) that refused MDA may also refuse testing during post-MDA surveillance . However , MX requires entomological expertise and laboratories with molecular capacity . In Bangladesh , 70 million individuals were at risk of LF before the Ministry of Health and Family Welfare ( MoHFW ) started its LF elimination program in 2000 [7–8] . Wuchereria bancrofti is the only species of human filarial worm currently known to be present in Bangladesh and the main vector is Culex quinquefasciatus [9] . Based on initial mapping , 19 of 64 districts were classified as endemic ( baseline microfilaria prevalence between 1% and 16% ) and therefore required MDA [7] , which began in 2001 . By 2016 , all 19 districts had passed the TAS and were eligible to cease MDA activities [7] . An ongoing surveillance project was initiated in April 2014 in Panchagarh ( one of the previously treated endemic districts ) and in Gaibandha ( a non-endemic district that had never conducted MDA ) . The latter district was selected because it borders a district with recent LF transmission and was considered at high risk for re-introduction . The objective of the project was to monitor W . bancrofti transmission trends through the assessment of microfilaremia ( Mf ) , antibodies , and antigenemia among adults in these two districts . Molecular xenomonitoring [10] was implemented as a complementary strategy for identifying areas of active transmission [11] . We sought to use MX to measure if the mosquito infection rate with W . bancrofti in the two districts was less than the cut-off point of 0 . 25% , a threshold that has been suggested for areas where Culex mosquitoes are the vector [10] .
Mosquitoes were collected in two evaluation units , one in Panchagarh district and one in Gaibandha district ( Fig 1 ) . Panchagarh district is part of the Rangpur division and is the most northeasterly district in Bangladesh , with a population of 987 , 644 and an area of 1404 km2 [12] . It is bordered on three sides by India and in the south by three other districts belonging to the Rangpur division , all of them previously endemic for LF but without any positive cases identified during the TAS1 and TAS2 ( 2013 , 2015 ) . The first evaluation unit in the MX study included all five sub-districts of Panchagarh district ( Atwari , Tetulia , Panchagarh Sadar , Debiganj , and Boda ) . The LF mapping carried out in 2001 showed an LF baseline prevalence of 10 . 8% ( Mf ) in Panchagarh . MDA activities started in 2001 . Following 12 rounds of MDA , the first TAS was carried out in April 2013 . None of the children included in the TAS were positive for circulating filarial antigen ( CFA ) , which made the district eligible to stop MDA . A second TAS carried out in 2015 also found no antigen-positive children [7] . Gaibandha district is also part of the Rangpur division with a population of 2 . 4 million and a total area of 2115 km2 [12] . It is bordered by six districts , three of which were previously endemic , two were classified as “low endemic” , and one as non-endemic district . One of the districts to the north-west of Gaibandha , Rangpur district , had a baseline Mf prevalence of 10 . 0% ( 2002 ) . Rangpur did not pass the first TAS in 2013 , but did pass in 2016 . No MDA or TAS has been conducted in Gaibandha district , as it was not considered endemic after mapping . The evaluation unit in Gaibandha district included only three of the seven sub-districts ( Palashbari , Sadullapur , and Sundarganj , population of just under 994 , 138 , area of 785km2 ) bordering the Rangpur district . These three sub-districts were selected due to ( i ) the risk of LF introduction due to the proximity with the Rangpur district and; ( ii ) the size of Gaibandha , which was considered too large for the implementation of MX throughout the entire district . This study used a two-stage cluster sampling design . Probability proportional to estimated size ( PPES ) was used for the selection of villages based on the 2011 Bangladesh Census Data [12] . Thirty villages were selected for each evaluation unit . The sampling interval used to select the villages was calculated by dividing the estimated number of HHs in the evaluation unit by the number of villages ( 30 ) . A number between zero and the sampling interval was randomly chosen to select the first village from the HH list , and then the sample interval was added to that number repeatedly until the 30 villages were chosen . Six HHs in each village were randomly selected ( 180 HHs per evaluation unit ) from a numbered list of HHs provided by the village health assistants; trapping was done at these HH . CDC gravid traps were placed in each site for three consecutive nights unless more than 100 Cx . quinquefasciatus were collected prior to the third night . The Culex quinquefasciatus sample size was based on a positivity threshold of <0 . 25% , a threshold previously suggested for areas where Culex mosquitoes are the vector [10] . When simple random sampling is presumed , a sample size of 6 , 850 Culex mosquitoes was required ( alpha = 0 . 05 , power = 0 . 75 ) . To account for community level clustering , this sample size was multiplied by a design effect of two ( 13 , 700 ) . This sample size was slightly decreased to 13 , 500 mosquitoes per evaluation unit ( overall total 27 , 000 mosquitoes ) so that it was evenly divisible by the cluster and pool size . We expected to collect at least 75 female Cx . quinquefasciatus mosquitoes over three nights at each household to be able to create three pools of 25 mosquito per trap site , which would result in a total of 540 pools of 25 mosquitoes per evaluation unit . No more than four pools were tested per site . Four teams of three entomologists from the MoHFW were tasked with trap deployment and mosquito identification . Each team was generally able to complete trapping in two villages per three-day collection period , resulting in collections occurring in eight villages per three-day period for the four teams . Mosquitoes were collected using CDC gravid traps ( John W . Hock Co . , Gainesville , FL ) , which are commonly used for sampling Cx . quinquefasciatus [13] . The traps were placed within 20 meters of the selected HH . The traps , containing 2–3 day old grass infusion bait , were turned on around 18:00 , and were retrieved the following morning between 08:00 and 11:00 . Gravid traps and nets were labelled using stickers with barcodes . Nets containing captured mosquitoes were returned to the laboratory and placed in the freezer to kill the mosquitoes . The mosquitoes were then identified using appropriate morphological identification keys [14 , 15] . The sex and physiological status ( unfed , fed , semigravid , gravid ) was recorded for each female mosquito . Once all intact mosquitoes from a trap were identified , they were placed in a 25ml Falcon tube with silica gel , which was also labeled with a barcoded sticker . CDC gravid traps can damage some of the collected mosquitoes [13] . Only intact mosquitoes were used for the pooling for two reasons: 1 ) identification of Cx . quinquefasciatus in Bangladesh requires observation of features from the head , thorax , and abdomen , and 2 ) association of partial mosquitoes might result in mismatches which could then lead to an overestimation of infection rates if a single positive mosquito was split between two pools . After the three days of collection unfed , fed , semigravid , and gravid female Cx . quinquefasciatus from a single trap site were pooled ( aiming for 25 mosquitoes per pool ) and desiccated in 1 . 5ml Eppendorf tubes with silica gel . Unfed mosquitoes are defined as nulliparous mosquitoes , which have never fed on humans so cannot be infected , as well as parous mosquitoes , which have laid their eggs , but have not taken a next blood meal . For each trap collection site , district , sub-district , house numbers , GPS coordinates , and trap barcode numbers were entered into a data collection application developed on an Android smart phones and using the LINKS System ( Task Force for Global Health , Decatur , Georgia ) [16] . Data on mosquito species were recorded on paper sheets and then double-entered into Excel ( Microsoft Corporation , Redmond , WA ) by two people . Differences were resolved by referring to the original data sheets . Pooled mosquitoes were tested for presence of W . bancrofti DNA at the molecular laboratory of the Institute of Epidemiology , Disease Control & Research ( IEDCR , MoHFW ) in Dhaka , Bangladesh . Positive and negative controls were also tested . DNA was extracted from pools using DNeasy Blood and Tissue kits ( Qiagen , Hilden , Germany ) following the manufacturer’s instructions . DNA quality was confirmed prior to PCR using a NanoDrop ( Thermo Fisher , Waltham , MA , USA ) . The real-time PCR protocol described by Rao et al . [17] was used to detect W . bancrofti DNA in the pools . All reactions were carried out in an ABI 7500 Fast Dx real-time PCR system ( Thermo Fisher Scientific , Waltham , MA , USA ) using Taqman Universal PCR Master Mix ( Thermo Fisher Scientific , Waltham , MA , USA ) , and all pools were run in duplicate . Data analysis was conducted using Stata 14 ( College Station , TX ) for descriptive statistics . The PoolScreen software ( version 2 . 0 . 3 ) was used to determine the maximum likelihood estimate and 95% confidence intervals of W . bancrofti infection prevalence in mosquitoes [18] . The map ( Fig 1 ) was created using ArcGIS 10 . 2 ( ESRI , Cary , NC , USA ) . This protocol was approved as a program evaluation by the U . S . Centers for Disease Control and Prevention ( #2015–180 ) . The protocol was also approved by the Bangladesh MoHFW . Written consent to place a gravid trap in the courtyard was obtained from household members .
Mosquitoes were collected in Gaibandha district from September 28 to October 11 , 2016 , and in Panchagarh district from October 19 to November 1 , 2016 . In each site , four teams of three entomologists were needed to place the mosquito traps , collect and identify mosquitoes , and create pools . The teams took 14 days in each district to place 6 traps in each of the 30 villages and collect mosquitoes for three consecutive nights at each site . Over the course of the entire collection period 24 , 408 mosquitoes were collected in gravid traps; 23 , 436 ( 96% ) were intact . Most mosquitoes identified to species were female ( 95% ) and the most commonly collected species was Cx . quinquefasciatus ( 47% ) ( Table 1 ) . A total of 10 , 344 female Cx . quinquefasciatus were collected during 1 , 079 trap nights ( 2 evaluation units , 180 HH per evaluation unit , 3 nights of trapping per household ) . For one trap location , only two nights were necessary to collect the number of mosquitoes needed . Of these , 10 , 021 ( 97% ) mosquitoes were sorted into 594 pools , 267 collected in Gaibandha and 327 collected in Panchagarh ( Table 2 ) . The range of mosquitoes per pool was 1–25 ( mean 16 . 9 ) ; and 256 pools ( 43 . 1% ) were composed of 25 mosquitoes . The target of 75 mosquitoes collected per collection site was not met in 324 of 360 sites ( 90% ) ; 222 of the 360 collection sites ( 61 . 7% ) collected fewer than 25 Cx . quinquefasciatus over 3 nights of trapping . As shown in Table 1 , Cx . quinquefasciatus and Cx . hutchinsoni were the only species that were predominantly collected in the gravid stage . Of the intact female Cx . quinquefasciatus collected , 88% were gravid , 10% were unfed , 1% were semigravid , and 1% were fed ( Table 1 ) . All mosquito collection data are presented in S1 Table . None of the 594 pools tested positive for the presence of W . bancrofti DNA by PCR . Using the Clopper-Pearson method [19] in PoolScreen , the 95%CI for the infection prevalence was 0–0 . 00051 for Panchagarh and 0–0 . 00073 for Gaibandha .
This study describes the first MX evaluation carried out in Bangladesh during the post-MDA period to evaluate the presence of mosquitoes infected with W . bancrofti . The results showed that none of the mosquito pools tested were positive for W . bancrofti DNA . This finding correlates with results from TAS surveys carried out in the previously endemic district ( Panchagarh ) , in 2013 and 2015 among children 6−7 years old , which did not identify any children with positive antigenemia . In the same district , the ongoing surveillance system among adults ≥ 18 years in five health facilities identified a circulating filarial antigen prevalence of less than 1% among the participants . TAS was not conducted in the non-endemic district ( Gaibandha ) , but the routine surveillance system among adults in seven health facilities found a circulating filarial antigen prevalence of less than 1% . The sum of our MX data and recent district-level data is consistent with the absence of W . bancrofti transmission in the two districts where the MX evaluations took place . The practical application of xenomonitoring activities is worthy of discussion . A key issue is to ensure that an adequate sample size can be attained . The main limitation encountered during the MX evaluation in Bangladesh was the difficulty in collecting sufficient mosquitoes to reach the targeted sample size of 13 , 500 female mosquitoes per district . Our sample size estimate was based on a on a positivity threshold of <0 . 25% and a design effect of two . Several sample sizes to detect culicine vector infection thresholds ( all larvae stage infection ) have been proposed , and range from 0 . 25% to 0 . 65% . [10 , 20] . MX evaluations are labor and time intensive and a balance is necessary between programmatic feasibility and scientific rigor . In a study conducted in Sri Lanka , Rao et al . [5] evaluated a programmatically scalable xenomonitoring program in a district where they had conducted xenomonitoring previously . They estimated that 75–150 traps placed in 30 clusters to collect 300 pools ( 25 mosquitoes per pool ) would be acceptable for programmatic implementation of xenomonitoring . Our study used approximately the same number of trap sites ( 180 per evaluation unit ) and trap nights ( 3 per trap site ) , and the areas of Panchagarh district ( 1405km2 ) and the three sub-districts of Gaibandha ( 785km2 ) were similar in size to that of Galle district in Sri Lanka ( 1652km2 ) . The fact that in Bangladesh we were unable to obtain the target sample size in each district is a key limitation of our study . However , if the mosquito infection rate were truly close to zero in all of the selected villages , the design effect would have been close to one . In that case , the sample size needed was half the one calculated initially ( 13 , 500 ) , closer to the 10 , 021 tested in this study . The large variation in size amongst clusters/villages also posed logistical challenges . Villages within the two study areas had sizes that ranged from two to 2875 HH according to the 2014 census . If we had calculated the sampling interval by dividing the total number of HH in the evaluation unit by the number of trap sites ( 180 houses per evaluation unit ) , some of the randomly selected villages would have received 0 traps and others 26 . Instead , we calculated the sampling interval by dividing the total number of HH per evaluation unit by 30 ( the total number of villages/clusters to be selected ) . In each of the 30 villages selected per evaluation unit , six trap-sites were systematically selected , regardless of the size of the village . By using this method , the number of traps allocated to each team and the number of villages visited daily could remain constant throughout the study . The interpretation of zero positive pools represented a challenge in this study . LF is a focal disease , and cross-sectional cluster surveys like MX or TAS have an inherent risk of missing residual foci of transmission [21] . MX provides an indication of the potential for ongoing filariasis transmission but its most efficient use as a surveillance tool remains to be determined . For example , MX might be a complementary surveillance tool implemented in parallel with TAS for surveillance throughout an entire district ( as done in this study and others [5] ) , with the risk of missing foci of transmission if the evaluation areas is large . A more efficient way to identify foci of transmission might be to use MX in smaller geographical areas , for example , where positive cases have been identified through a TAS [22] . In most cases , three nights were not enough to collect three pools of 25 mosquitoes per site . We wanted to undertake these collections at the time of year when the highest Cx . quinquefasciatus densities were present . However , baseline data from the study areas regarding seasonal densities of Cx . quinquefasciatus were not available , and there are differing accounts of Cx . quinquefasciatus seasonality in Bangladesh in the published literature . Begum et al . [23] found the highest densities of Cx . quinquefasciatus in December in Dhaka using human landing collections . Ameen & Moizuddin [24] found peaks in November and March , based on mosquito collections from cattle in Dhaka . Aslamkhan & Wolfe [9] found peak numbers in human landing collections in March/April in Dinajpur district , which borders Panchagarh and Gaibandha . Finally , Karim et al . [25] found the highest number of Cx . quinquefasciatus in Dhaka in the months of March and November . These data made it difficult to identify the best time of year to collect Culex quinquefasciatus in Gaibandha and Panchagarh . Because of budget and time constraints , we were unable to extend the collection times to ensure that three pools of 25 mosquitoes were obtained at each site . The MoHFW entomologists conducting the field work were responsible for other entomological activities in the country , ranging from sampling of malaria vectors to surveillance of Aedes aegypti . As such , we had to define a discrete period of collection in order for the program to adequately balance available human and budgetary resources . The composition and maturity of the grass infusion used to bait the traps may have impacted trap yields . While grass infusion has been shown to be an effective attractant for Cx . quinquefasciatus in Tanzania [13] as well as in Dhaka during preliminary trapping , a change in the quality of the infusion was noticed over time . Drums were refilled with water and grass as soon as the previous batch was used , allowing the bacteria remaining in the drums to provide a culture for the subsequent batch . We noticed an upward trend in the number of mosquitoes collected per consecutive trap night , which might be the result of increased attractiveness of the infusion due to bacterial colonization [26] . However , as we were collecting in different villages every three nights , it is not possible to know whether this increase was due to the grass infusion or the sequence of villages where trapping was conducted . Different infusions have been used in previous studies; this likely affects catch size . For example , Rao et al . [4] used an infusion of yeast , milk powder , and dry straw . In India , a bait of hay , yeast , and water was used ( S Subramanian , personal communication ) . It would be worthwhile to conduct an experiment to find the most attractive infusion for Cx . quinquefasciatus to standardize this aspect of xenomonitoring , while realizing that the attractiveness of these infusions may be variable from site to site . A highly attractive infusion could optimize collection efficiency and reduce the number of days of trapping needed to reach the desired sample size . In addition to Cx . quinquefasciatus , 30 other mosquito species were collected , all of which had been previously recorded from Bangladesh [27] . Non-vectors might also be used for xenomonitoring , and testing non-vectors might help increase certainty of elimination of transmission in an area [28 , 29] . Additionally , identification of all mosquito species was possible in this study because of the skill of the entomologists , but it did increase the time needed for processing mosquitoes and this capacity is not present in all locations . While our aim was to collect vectors of W . bancrofti , vectors of other diseases were collected as well . Al-Amin et al . [30] previously found malaria parasites in three of the Anopheles species collected in this study ( Anopheles barbirostris , An . vagus , and An . umbrosus ) . An . annularis [31] and An . philippinensis [32] , have also been identified as malaria vectors in Bangladesh . Of particular interest was the collection of vectors of Japanese encephalitis virus ( JEV ) ( Cx . gelidus , Cx . pseudovishnui , Cx . vishnui , and Cx . tritaeniorhynchus ) . As resting collections near oviposition sites enhance the likelihood of collecting JEV vectors [33] , gravid trapping might be of use for surveillance of these mosquitoes , particularly if attractants resemble their natural oviposition sites . While all species collected had been previously collected in Bangladesh , the collections of Aedeomyia catastica , Armigeres kesseli , Uranotaenia rampae , Ur . campestris , Cx . infula , and Mansonia annulata are especialy noteworthy as records of these species in Bangladesh are relatively rare [27] . Although this was the first MX evaluation carried out in Bangladesh , MX evaluations have been used in other countries to evaluate the impact of MDA on human infection prevalence [4–6 , 10 , 22 , 34 , 35] . After stopping MDA , national LF elimination programs will need to plan for post-elimination validation surveillance activities that could be routinely implemented to detect recrudescence or re-introduction of LF and to confirm interruption of transmission [36] . Though MX evaluations provide a sensitive method to detect residual foci of transmission and have been suggested by WHO as an alternative surveillance method for LF [37] , not all countries have the capacity to include MX as a routine activity in their post-validation surveillance plan . MX evaluations for post-elimination validation surveillance could be recommended in high-risk transmission areas in countries with the appropriate entomological and laboratory capacities . In Bangladesh for example , where the entomological and laboratory capacity to perform MX is available , MX could be used as one of the post-validation surveillance strategies to confirm interruption of transmission in areas at higher risk of transmission identified during TAS [7] . Further operational research and information sharing about how to programmatically simplify and standardize MX evaluations will also make these evaluations more accessible to a larger number of LF endemic countries entering the post-elimination validation phase .
|
To ensure elimination of lymphatic filariasis , efficient surveillance methods are needed . While some available methods rely on the detection of Wuchereria bancrofti microfilaria , antigen or antibody in human blood samples , molecular xenomonitoring can identify parasite DNA in vector mosquitoes . We collected the main vector of lymphatic filariasis in Bangladesh , Culex quinquefasciatus mosquitoes , in two districts in Bangladesh to see if W . bancrofti could be detected . One of the districts never had evidence of widespread transmission but borders another district where transmission was most recently detected . The other district had previously had W . bancrofti transmission , but after 12 rounds of mass drug administration , had been deemed to have little to no ongoing transmission . In each district , traps were set at 180 sites to collect mosquitoes . Over ten thousand Cx . quinquefasciatus mosquitoes were collected , but none of them tested positive for presence of W . bancrofti . The practice of trapping mosquitoes was feasible for the national program to execute , and the absence of infected mosquitoes suggests that parasite rates are nearing zero .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"geographical",
"locations",
"tropical",
"diseases",
"parasitic",
"diseases",
"animals",
"filariasis",
"neglected",
"tropical",
"diseases",
"molecular",
"biology",
"techniques",
"insect",
"vectors",
"zoology",
"lymphatic",
"filariasis",
"bangladesh",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"wuchereria",
"bancrofti",
"artificial",
"gene",
"amplification",
"and",
"extension",
"wuchereria",
"culex",
"quinquefasciatus",
"molecular",
"biology",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"helminth",
"infections",
"mosquitoes",
"eukaryota",
"asia",
"entomology",
"polymerase",
"chain",
"reaction",
"nematoda",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"organisms"
] |
2018
|
Molecular xenomonitoring for Wuchereria bancrofti in Culex quinquefasciatus in two districts in Bangladesh supports transmission assessment survey findings
|
Comparisons of DNA sequences between Neandertals and present-day humans have shown that Neandertals share more genetic variants with non-Africans than with Africans . This could be due to interbreeding between Neandertals and modern humans when the two groups met subsequent to the emergence of modern humans outside Africa . However , it could also be due to population structure that antedates the origin of Neandertal ancestors in Africa . We measure the extent of linkage disequilibrium ( LD ) in the genomes of present-day Europeans and find that the last gene flow from Neandertals ( or their relatives ) into Europeans likely occurred 37 , 000–86 , 000 years before the present ( BP ) , and most likely 47 , 000–65 , 000 years ago . This supports the recent interbreeding hypothesis and suggests that interbreeding may have occurred when modern humans carrying Upper Paleolithic technologies encountered Neandertals as they expanded out of Africa .
A much-debated question in human evolution is the relationship between modern humans and Neandertals . Modern humans appear in the African fossil record about 200 , 000 years ago . Neandertals appear in the European fossil record about 230 , 000 years ago [1] and disappear about 30 , 000 year ago . They lived in Europe and western Asia with a range that extended as far east as Siberia [2] and as far south as the middle East . The overlap of Neandertals and modern humans in space and time suggests the possibility of interbreeding . Evidence , both for [3] and against interbreeding [4] , have been put forth based on the analysis of modern human DNA . Although mitochondrial DNA from multiple Neandertals has shown that Neandertals fall outside the range of modern human variation [5] , [6] , [7] , [8] , [9] , [10] , low-levels of gene flow cannot be excluded [10] , [11] , [12] . Analysis of the draft sequence of the Neandertal genome revealed that the Neandertal genome shares more alleles with non-African than with sub-Saharan African genomes [13] . One hypothesis that could explain this observation is a history of gene flow from Neandertals into modern humans , presumably when they encountered each other in Europe and the Middle East [13] ( Figure 1 ) . An alternative hypothesis is that the findings are explained by ancient population structure in Africa [13] , [14] , [15] , [16] , whereby the population ancestral to Neandertal and modern human ancestors was subdivided . If this substructure persisted until modern humans carrying Upper Paleolithic technologies expanded out of Africa so that the modern human population that migrated was genetically closer to Neandertals , people outside Africa today would share more genetic variants with Neandertals that people in sub-Saharan Africa [13] , [14] , [15] ( Figure 1 ) . Ancient substructure in Africa is a plausible alternative to the hypothesis of recent gene flow . Today , sub-Saharan Africans harbor deep lineages that are consistent with a highly-structured ancestral population [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] , [27] . Evidence for ancient structure in Africa has also been offered based on the substantial diversity in neurocranial geometry amongst early modern humans [28] . Thus , it is important to test formally whether substructure could explain the genetic evidence for Neandertals being more closely related to non-Africans than to Africans . A direct way to distinguish the hypothesis of recent gene flow from the hypothesis of ancient substructure is to infer the date for when the ancestors of Neandertals and a modern non-African population last exchanged genes . In the recent gene flow scenario , the date is not expected to be much older than 100 , 000 years ago , corresponding to the time of the earliest documented modern humans outside of Africa [29] . In the ancient substructure scenario , the date of last common ancestry is expected to be at least 230 , 000 years ago , since Neandertals must have separated from modern humans by that time based on the Neandertal fossil record of Europe [1] . In present-day human populations , the extent of LD between two single nucleotide polymorphisms ( SNPs ) shared with Neandertals can be the result of two phenomena . First , there is “non-admixture LD” [30] whose extent reflects stretches of DNA inherited from the ancestral population of Neandertals and modern humans as well as LD that has arisen due to bottlenecks and genetic drift in modern humans since they separated from Neandertals . Second , if gene flow from Neandertals into modern humans occurred , there is “admixture LD” [30] , which will reflect stretches of genetic material inherited by modern humans through interbreeding with Neandertals . The extent of LD between single nucleotide polymorphisms ( SNPs ) shared with Neandertals will thus reflect , at least in part , the time since Neandertals or their ancestors and modern humans or their ancestors last exchanged genes with each other . The strategy of using LD to estimate dates of gene flow events has been previously been explored by several groups [31] , [32] , [33] , [34] , [35] . Our methodology is conceptually similar to the methodology developed by Moorjani et al . , but is dealing with a more challenging technical problem since the methodology developed by Moorjani et al . is adapted for relatively recent admixtures . In recently admixed populations that have not experienced recent bottlenecks , admixture LD extends over size scales at which non-admixture LD makes a negligible contribution . Thus , one can infer the time of gene flow based on inter-marker spacings that are larger than the scale of non-admixture LD . For older admixtures however ( such as may have occurred in the case of Neandertals ) , non-admixture LD occurs almost at the same size scale as admixture LD . To account for this , we study pairs of markers that are very close to each other , but ascertain them in a way that greatly minimizes the signals of non-admixture LD while enhancing the signals of admixture LD . Thus , unlike in the case of recent admixtures , non-admixture LD could bias an admixture date obtained using our methods; however , we show using simulations of a very wide set of demographic scenarios that our marker ascertainment procedure makes the bias so small that our inferences are qualitatively unaffected . Our methodology is based on the idea that if two alleles , a genetic distance x ( expected number of crossover recombination events per meiosis ) apart , arose on the Neandertal lineage and introgressed into modern humans at time tGF , the probability that these alleles have not been broken up by recombination since gene flow is proportional to . We show that the LD across introgressed pairs of alleles is expected to decay exponentially with genetic distance . The rate of decay is informative of the time of gene flow and is robust to demographic events ( Appendix A , Text S1 ) . In practice , we need to ascertain SNPs that , assuming recent gene flow occurred , are likely to have arisen on the Neandertal lineage and introgressed into modern humans . We choose a particular ascertainment scheme and show , using simulations of a number of demographic models , that the exponential decay of LD across pairs of ascertained SNPs provides accurate estimates of the time of gene flow . A second potential source of bias in estimating ancient dates arises from uncertainties in the genetic map . We develop a correction for this bias and show that this correction yields accurate dates in the presence of uncertainties in the genetic map . Combining these various strategies , we are able to obtain accurate estimates of the date of last exchange of genes between Neandertals and modern humans ( also see Discussion ) . This date shows that recent gene flow between Neandertals and modern humans occurred but does not exclude that ancient substructure in Africa also contributes to the LD observed .
To study how LD decays with the distance in the genome , we computed the average value , , of the measure of linkage disequilibrium D ( the excess rate of occurrence of derived alleles at two SNPs compared with the expectation if they were independent [36] ) between pairs of SNPs binned by genetic distance x ( see Methods ) . Immediately after the time of last gene flow between Neandertal ( or their relatives ) and human ancestors , long range LD is generated , and it is then expected to decay at a constant rate per generation as recombination breaks down the segments shared with Neandertals . Thus , in the absence of new LD-generating events ( discussed further below ) , the statistic across pairs of introgressed alleles is expected to have an exponential decay with genetic distance , and the genetic extent of the decay can thus be interpreted in terms of the time of last shared ancestry between Neandertals ( or their relatives ) and modern humans ( Section S1 and Appendix A in Text S1 ) . To amplify the signal of admixture LD relative to non-admixture LD , we restricted our analysis to SNPs where the “derived” allele ( the one that has arisen as a new mutation as determined by comparison to chimpanzee ) is found in Neandertals and occurs in the tested population at a frequency of <10% . The justification for this frequency threshold is two-fold . First , the signal of Neandertals being more closely related to non-Africans than to Africans is substantially enriched at SNPs below this threshold ( Section S1 in Text S1 ) . Second , under the model of recent gene flow , such SNPs have an increased probability of having arisen due to mutations on the Neandertal lineage; we estimate that about 30% of them will have arisen on the Neandertal lineage under a model of history that we fitted to the data . This ascertainment enriches the class of informative SNPs by a factor of ten ( Section S1 in Text S1 ) . Our simulations show that restricting to this class of SNPs yields accurate estimates of the time of gene flow for a wide range of demographic histories consistent with patterns of human variation ( Section S2 in Text S1 ) . To assess how useful this statistic is for measuring admixture LD , we performed coalescent simulations of 100 regions of a million base pairs each , for a range of demographic histories chosen to be plausible for Neandertals , West Africans and non-Africans ( these histories were constrained by the observed population differentiation between west Africans and Europeans as measured by their FST and the quantitative extent to which Neandertals share more derived alleles with Europeans than with Africans ) . The simulation results , which we discuss at length in Section S2 of Text S1 , and summarize in Table 1 , show that we obtain accurate and relatively unbiased estimates of the number of generations since admixture ( never more than 15% from the true value ) for ( 1 ) constant-sized population scenarios , ( 2 ) demographic models that include population bottlenecks as well as more recent admixture after the gene flow , ( 3 ) hybrid models of ancient structure and recent gene flow , and ( 4 ) mutation rates that differ by a factor of 5 from what we use in our main simulations ( see Figure 2 ) . Two other SNP ascertainment schemes yield qualitatively consistent findings but the ascertainment we used provides the most accurate estimates under the range of demographic models considered ( Section S5 of Text S1 and Table 1 ) . The simulations also show that in the absence of gene flow ( including in the scenario of ancient subdivision ) , the dates obtained are always at least 5 , 000 generations for scenarios of demographic history that match the constraints of real human data . Thus , an empirical estimate of a date much less than 5 , 000 generations likely reflects real gene flow . We applied our statistic to data from Pilot 1 of the 1000 Genomes Project , which discovered polymorphisms in 59 West Africans , 60 European Americans , and 60 East Asians ( Han Chinese and Japanese from Tokyo ) [37] . We binned pairs of SNPs by the genetic distance between them using the deCODE genetic map . We considered all pairs of SNPs that are at most 1 cM apart . We computed the average LD over all pairs of SNPs in each bin and fit an exponential curve to the decay of LD ( from 0 . 02–1 cM in 0 . 001 cM increments ) . Figure 3 shows the extent of LD for pairs of SNPs where both SNPs have a derived allele frequency <10% . This figure shows that the extent of LD is larger in Europeans and East Asians than in West Africans , both when the Neandertal genome carries the derived and when it carries the ancestral allele . Empirical features of these LD decay curves show that , for alleles derived in the Neandertal genome , the pattern in Europeans and East Asians is reflecting “admixture LD” . LD in West Africans is less extensive when Neandertals carry the derived allele than when they carry the ancestral allele , while the reverse is seen in Eurasians . To understand this , we note that in the absence of gene flow , polymorphic sites where Neandertals carry the derived allele must have arisen from mutations that occurred prior to Neandertal-human divergence so that they are old and recombination will have had a lot of time to break down the LD , while sites where Neandertals carry the ancestral allele mutations will include mutations that have arisen since the Neandertal-human split and thus LD will be expected to be more extensive , exactly as is seen in West Africans . In contrast , if gene flow occurred , then LD can be greater at sites where Neandertals carry the derived allele as is observed in Europeans and East Asians . This signal persists when we stratify the LD decay curves by the frequency of the ascertained SNPs ( Figure S8 in Text S1 ) . Thus the scale of the LD at these sites must be conveying information about the date of gene flow . A concern in interpreting the extent of LD in terms of a date is that all available genetic maps ( which specify the probability of recombination per generation between all pairs of SNPs ) are likely to be inaccurate at the scale of tens of kilobases that is relevant to our analysis . We confirmed that errors in genetic maps can bias LD-based date estimates by simulating a gene flow event 2 , 000 generations ago using a model in which recombination was localized to hot spots [38] but where the data were analyzed assuming a genetic map that assumed homogeneous recombination rates across the genome . This led to a date of 1 , 597 generations since admixture . We developed a statistical model of the random errors that relate the true and observed genetic maps ( see Methods ) . The precision of the map is modeled using a scalar parameter α . A unit interval of the observed genetic map corresponds to an interval in the true map of expected unit length and variance 1/α . To validate this error model , we estimated the map error in these simulations ( α ) by comparing the true and the observed genetic maps . Theoretical arguments ( Section S3 in Text S1 ) show that we can obtain a corrected date ( tGF ) from the uncorrected date in generations ( λ ) using the equation tGF = α ( eλ/α-1 ) . We applied this correction to obtain a date of 1 , 926 generations . While this error model appears to provide an adequate description of random errors in a genetic map , it does not account for systematic biases . To apply this statistical correction to real data , we estimated the error rate α in the genetic map by comparing the genomic distribution of a set of cross-over events from 728 meioses previously detected in a European American Hutterite pedigree [39] to what would be expected if the map were perfect . Unfortunately , the map that we would ideally want to use for estimating the date of Neandertal admixture is not the genetic map that applies to Hutterites today , but the time-averaged genetic map that applied between the present and the date of gene flow . Obviously , such a map is not available , but we hypothesize that by performing our analyses using a genetic map that is built from samples more closely related to the Hutterite pedigree than the map that we would like to analyze ( the deCODE pedigree map built in Icelanders ) as well as a genetic map that averages over too long a period of time ( the European LD Map , which measures recombination over approximately five hundred thousand years ) , we can obtain some sense of the robustness of our inferences to uncertainties in how the European genetic map has changed over time . Table 2 shows the estimates of λ , α and tGF in Europeans obtained using the two genetic maps . The estimates of tGF are in 1 , 805–2 , 043 over the deCODE and European LD maps . We also estimated λ in East Asians using the “East Asian LD map” . We find that λ in East Asians based on the East Asian LD map is 1 , 253–1 , 287 , similar to the 1 , 159–1 , 183 in Europeans based on the European LD map , although the similarity of these numbers does not prove the Neandertal genetic material in Europeans and East Asians derives from the same ancestral gene flow event . While a shared ancestral gene flow event is plausible , the gene flow events could in principle have occurred in different places at around the same time [40] . We also cannot reliably estimate the recombination rate correction factor α for the East Asian map because we do not have access to cross-over events in an East Asian pedigree , and hence we do not present an estimate of tGF in East Asians and focus on Europeans in the rest of this paper . To convert the date estimates in generations to date estimates in years , we use an average generation interval which has been estimated to be 29 in diverse modern hunter gatherer societies as well as in developing and industrialized nation states [41] . We assume a uniform prior probability distribution of generation times between 25 and 33 years per generation for the true value of this quantity and integrate this with the uncertainty of λ and α , and obtain an estimate of last gene exchange between Neandertals and European ancestors of 47 , 334–63 , 146 years for the deCODE map , and 49 , 021–64 , 926 years for the European LD Map ( 95% credible intervals ) . Taking the conservative union of these ranges , we obtain 47 , 000–65 , 000 years BP . In our simulations of ascertainment strategy , we found demographic models that can produce biases in the date estimates that could be as large as 15% ( Section S2 in Text S1 ) . To be conservative , we applied this to the uncorrected dates from each of the maps and then applied the relevant map correction . The union of the resulting intervals leads us to conclude that the true date of gene flow could be as young as 37 , 000 years BP or as old as 86 , 000 years BP . We considered the possibility that our results might be biased by natural selection , which is known to affect patterns of human genetic diversity and to have had a much larger effect closer to genes [42] , [43] . We estimated the time of gene flow stratifying the SNPs by their distance to the nearest exon , dividing the data into 5 bins such that each bin contained 20% of all the SNPs . Using the deCODE map , we obtain λ = 1 , 145–1 , 301 in all bins ( Table S8 in Text S1 ) . This estimate is concordant with the λ = 1 , 201 obtained without stratification , and suggests that our inferences are not an artifact of LD generated by directional natural selection .
The date of 37 , 000–86 , 000 years BP is too recent to be consistent with the “ancient African population structure” scenario , and strongly supports the hypothesis that at least some of the signal of Neandertals being more closely related to non-Africans than to Africans is due to recent gene flow . These results are concordant with a recent paper by Yang et al [44] that analyzed joint allele frequency spectra in Africans , non-Africans and Neandertals , to reject the ancient structure scenario . After the present paper was accepted , Eriksson and Manica [45] showed , using an Approximate Bayesian Computation approach , that models of ancient substructure can produce a signal of Neandertals sharing more derived alleles with non-Africans than with Africans ( that is , they can account for the observation that D-statistics are significantly different from zero ) . The same observation was made in our earlier papers on the draft Neandertal and Denisovan genomes where we introduced D-statistics [13] , [14] . However , the new statistics we focus on here as well as the statistics focused on by Yang et al [44] show that ancient structure alone cannot explain these signals . One possibility that we have not ruled out is that both ancient structure and gene flow occurred in the history of non-Africans . In the simulations reported in Table 1 , we show that in this scenario , the ancient structure will tend to make the date estimate older than the truth but by not more than 15% , so that the date of 37 , 000–86 , 000 should still provide a valid bound while the less conservative estimate of 47 , 000–65 , 000 years should be interpreted as an upper bound on the date of gene flow . Further , we have not been able to differentiate amongst variants of the recent gene flow scenario: a single episode or multiple episodes of gene flow or continuous gene flow over an extended period of time . Our date has a clear interpretation as the time of last gene exchange under a scenario of a single instantaneous gene flow event . In the other scenarios , the date is expected to represent an average over the times of gene flow and should be interpreted as an upper bound on the time of last gene exchange . While recent gene flow from Neandertals into the ancestors of modern non-Africans is a parsimonious model that is consistent with our results , our analysis cannot reject the possibility that gene flow did not involve Neandertals themselves , but instead populations that were more closely related to Neandertals than any extant populations are today . Thus , the date should be interpreted as the last period of time when genetic material from Neandertals or an archaic population related to Neandertals entered modern humans . Genetic analyses by themselves offer no indication of where gene flow may have occurred geographically . However , the date in conjunction with the archaeological evidence suggests that the two populations likely met somewhere in Western Eurasia . An attractive hypothesis is the Middle East , where archaeological and fossil evidence indicate that modern humans appeared before 100 , 000 years ago ( as reflected by the modern human remains in Skhul and Qafzeh caves ) , Neandertals expanded around 70 , 000 years ago ( as reflected for example by the Neandertal remains at Tabun Cave ) , and modern humans re-appeared around 50 , 000 years ago [29] . Our genetic date estimates , which have a mostly likely range of 47 , 000–65 , 000 years ago ( and are confidently below 86 , 000 years ago ) , are too recent to be consistent with the appearance of the first fossil evidence of modern humans outside of Africa—that is , our date makes it unlikely that the Neandertal genetic material in modern humans today could arise exclusively due to the gene flow involving the Skhul/Qafzeh modern humans—and instead point to gene flow in a more recent period , possibly when modern humans carrying Upper Paleolithic technologies expanded out of Africa .
Our procedure computes a statistic based on the LD observed between pairs of SNPs . For all pairs of ascertained SNPs at a genetic distance x , we compute the statistic:Here S ( x ) denotes the set of all pairs of ascertained SNPs that are at a genetic distance x , and D ( i , j ) denotes the classic signed measure of linkage disequilibrium , D , at the SNPs i , j . The sign of D ( i , j ) is determined by computing D using the derived alleles ( defined relative to the chimpanzee base ) at SNPs i and j . Under the gene flow scenario , we expect the contribution of introgression to to have an exponential decay with rate equal to the time of gene flow , provided the gene flow is more recent than the Neandertal-modern human split ( Section S1 and Appendix A of Text S1 ) . We pick SNPs that contain a derived allele in Neandertal ( defined relative to the chimpanzee base ) and are polymorphic in the target population with a derived allele frequency <10% . Further details can be found in Text S1 , along with simulations exploring the performance of the statistic and demonstrating its properties under various demographic models and ascertainment schemes . We used the 1000 Genomes Pilot 1 genotypes to estimate the LD decay . For each of the panels that were chosen as the target population in our analysis , we restricted our analysis to polymorphic SNPs . The SNPs were polarized relative to the chimpanzee base ( panTro2 ) . For the set of ascertained SNPs , we compute as a function of the genetic distance x and fit an exponential curve using ordinary least squares for x in the range of 0 . 02 cM to 1 cM in increments of 0 . 001 cM . The standard definition of D requires the availability of haplotypes . We instead computed D ( i , j ) as the covariance between the genotypes observed at SNPs i and j [46] . Simulations show that dates estimated using this definition of D on unphased genotypes are very similar to the estimates obtained from haplotypes ( Section S2 . 1 . 1 of Text S1 ) . We were concerned that the complicated method used in the 1000 Genomes Project for determining genotypes , which involved statistical imputation and probabilistic calling of genotypes based on LD , might in some way be biasing our inferences based on LD . Thus , we also computed D ( i , j ) for all pairs of SNPs that passed our basic filters ( SNPs that contain a derived allele in Neandertal and are polymorphic in the target population with derived allele frequency <10% as estimated from the reads ) by computing LD directly from the reads , again using the SAMtools package [47] , and obtain qualitatively consistent results ( Section S7 of Text S1 ) . Further , simulations to mimic the low power to call rare SNPs in the 1000 genomes data show that our estimates are not sensitive to the deficit of rare alleles ( Section S6 of Text S1 ) . We have a genetic map G defined on m markers . Each of the m-1 intervals is assigned a genetic distance gi , i = 1 , . . m-1 . These genetic distances provide a prior distribution for the true underlying ( unobserved ) genetic distances Zi . A reasonable prior on each Zi is then:where α is a parameter that is specific to the map . This implies that the true genetic distance Zi has mean gi and variance gi/α . Thus , large values of α correspond to a more precise map . A motivation for the choice of the gamma prior over Zi is that this prior has the key invariance property Z1+Z2∼Γ ( α ( g1+g2 ) , α ) . Thus , α is a property of the map and not of the specific markers used . Given this prior on the true genetic distances , fitting an exponential function to pairs of markers at a given observed genetic distance g involves integrating over the exponential function evaluated at the true genetic distances given observed genetic distance g , that is:where λ is the rate of decay of as a function of the observed genetic distance g and can be estimated from the data as described in the previous section , tGF denotes the true time of the gene flow and the expectation is over the unobserved true genetic distance Z . We can use this equation to solve for tGF as ( see Appendix B in Text S1 ) :To estimate α for a given genetic map , we propose a statistical model that relates the true unobserved genetic map to the observed map and to crossover events found in a pedigree . We estimate the posterior distribution of α by Gibbs sampling ( Section S3 of Text S1 ) . To obtain estimates of the time of gene flow taking into account all sources of error , we formulated a Bayesian model that relates λ , tGF , and yGF ( the time in years ) ( Section S4 of Text S1 ) to the observed LD decay curve . Further , we assume a uniform prior distribution on the number of years per generation of 25–33 years , based on a recent survey of generation intervals , which are similar in diverse hunter-gatherer societies and in undeveloped as well as industrialized nation states . Assuming a flat prior on each of λ , tGF , and yGF , we use Gibbs sampling to obtain samples from the posterior distributions of each of these parameters . We then report the posterior mean and 95% Bayesian credible intervals . We will make the data and programs available at http://genetics . med . harvard . edu/reichlab/Reich_Lab/Datasets . html on publication .
|
One of the key discoveries from the analysis of the Neandertal genome is that Neandertals share more genetic variants with non-Africans than with Africans . This observation is consistent with two hypotheses: interbreeding between Neandertals and modern humans after modern humans emerged out of Africa or population structure in the ancestors of Neandertals and modern humans . These hypotheses make different predictions about the date of last gene exchange between the ancestors of Neandertals and modern non-Africans . We estimate this date by measuring the extent of linkage disequilibrium ( LD ) in the genomes of present-day Europeans and find that the last gene flow from Neandertals into Europeans likely occurred 37 , 000–86 , 000 years before the present ( BP ) , and most likely 47 , 000–65 , 000 years ago . This supports the recent interbreeding hypothesis and suggests that interbreeding occurred when modern humans carrying Upper Paleolithic technologies encountered Neandertals as they expanded out of Africa .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"haplotypes",
"mathematics",
"genetic",
"polymorphism",
"statistics",
"statistical",
"methods",
"population",
"genetics",
"biology",
"evolutionary",
"biology",
"gene",
"flow",
"evolutionary",
"genetics"
] |
2012
|
The Date of Interbreeding between Neandertals and Modern Humans
|
Cell penetrating peptides ( CPPs ) are those peptides that can transverse cell membranes to enter cells . Once inside the cell , different CPPs can localize to different cellular components and perform different roles . Some generate pore-forming complexes resulting in the destruction of cells while others localize to various organelles . Use of machine learning methods to predict potential new CPPs will enable more rapid screening for applications such as drug delivery . We have investigated the influence of the composition of training datasets on the ability to classify peptides as cell penetrating using support vector machines ( SVMs ) . We identified 111 known CPPs and 34 known non-penetrating peptides from the literature and commercial vendors and used several approaches to build training data sets for the classifiers . Features were calculated from the datasets using a set of basic biochemical properties combined with features from the literature determined to be relevant in the prediction of CPPs . Our results using different training datasets confirm the importance of a balanced training set with approximately equal number of positive and negative examples . The SVM based classifiers have greater classification accuracy than previously reported methods for the prediction of CPPs , and because they use primary biochemical properties of the peptides as features , these classifiers provide insight into the properties needed for cell-penetration . To confirm our SVM classifications , a subset of peptides classified as either penetrating or non-penetrating was selected for synthesis and experimental validation . Of the synthesized peptides predicted to be CPPs , 100% of these peptides were shown to be penetrating .
Cell penetrating peptides ( CPPs ) , also referred to as “Trojan” peptides , protein transduction domains , or membrane translocation sequences , are typically hydrophobic linear arrangements of 8–24 amino acids able to cross the lipid bi-layer membrane that serves as the cell’s outer barrier and gain access to the interior of the cell and its components [1] . Penetratin , an Antennapedia derived peptide , and the HIV derived Tat peptide were some of the first commonly studied CPPs , and along with transportan peptides ( derived from galanin receptor ligand proteins ) , make up three major families of CPPs . The remainder of CPPs are classified in a fourth , miscellaneous family [1] . Initially , cellular uptake of CPPs was believed to be through endocytosis or protein transporters , but some evidence suggested the mechanism may involve direct transport through the lipid bi-layer of the cell , which takes into account the hydrophobic properties of most of these peptides [2] . The current view is that CPP internalization is accomplished predominantly by endocytosis [2] . Historically , both flow cytometry and fluorescence microscopy have been used to study the uptake of CPPs into cells . Care must be used with these methods to avoid artifacts because traditional methodologies for these techniques can incorrectly show a high concentration of CPPs localizing to the cell nucleus or a higher than actual concentration of CPPs being taken into the cell [2] . Cell penetrating peptides capable of transporting other active molecules inside the cell have the potential to serve as drug delivery peptides . Drug delivery peptides and CPPs allow researchers to probe the mechanisms of peptide transport across a lipid bi-layer membrane and may allow customizable drug therapies for differing types of cells . Although there is some controversy regarding CPPs as drug delivery systems because of their lack of specificity for cell type , the general consensus among researchers is that both general CPPs and cell-specific CPPs will be developed into effective drug delivery systems in the future [3] , [4] . A classification system that can determine whether or not a unique peptide sequence can serve as a CPP , and thus possibly be a potential drug delivery peptide , can enable researchers to quickly screen candidate molecules for their potential viability for use in a customizable drug delivery regime . Much of the previous work in the prediction of CPPs has involved the use of a set of composite features assembled from primary biochemical properties through the use of principal component analysis [5] , [6] , [7] . These composite features , or z-scores , consist of a numerical value and an associated range . To predict cell-penetrating capability of a candidate peptide , the z-scores are computed for the peptide , and , if the z-scores fall within the range of known CPP z-scores , the peptide is classified as cell-penetrating [6] , [7] . While this method has a high accuracy ( >95% correct prediction of novel CPPs ) for generating novel CPPs [6] , it performs rather poorly ( 68% correct prediction ) when trying to distinguish known non-penetrating peptides that are closely related to known CPPs [7] and yields little information about exactly which biochemical properties contribute to the difference between these two classes . More recent work examines the use of quantitative structure-activity relationship ( QSAR ) derived features to predict penetration potential . The training process iteratively removes sequences that are difficult to classify and thus the classification accuracies reported are biased [8] . Further research into this topic is necessary to allow potential drug delivery peptides to be rapidly screened for usefulness . Using the basic biochemical properties of peptides as features instead of the widely used composite z-scores can potentially provide more insight into the differences between the class of CPPs and non-penetrating peptides when coupled with wrapper-based feature selection and classifier training using a machine learning technique such as a support vector machine . Additionally , once trained , these machine learning classifiers can then be used for rapid screening of candidate CPPs prior to their synthesis . This study examines the available information on known CPPs and their non-penetrating analogs in order to compile datasets used for training and testing of support vector machine classifiers using primary features derived from biochemical properties of each peptide and evaluates the accuracy of these classifiers . An experimental validation study was performed to determine the effectiveness of these classifiers using an avian tissue culture system .
Because of the sensitivity of classifiers to unbalanced classes [9] , our first challenge was to generate datasets for training and testing . A set of 111 known CPPs were identified from the literature [6] , [7] , [11] . However , only 34 negative examples could be found and many of these are analogs of known CPPs [6] , [7] . Unbalanced datasets present a number of different problems for machine learning methods [9] . When only a comparatively small number of examples are available for one class , the machine learning algorithm will not have sufficient information to learn a function to distinguish the classes . Reporting of classification accuracy is also impacted by unbalanced datasets . For example , if a dataset of 100 peptides contains 80 CPPs and 20 non-CPPs , a classification accuracy of 80% can be obtained by classifying all peptides as positive . Most previous work in CPP prediction has ignored this problem [7] , [8] . We designed an experiment to investigate the effect of unbalanced datasets on CPP prediction and to find methods to address the problem to evaluate classifier accuracy with precision . For the CPP prediction problem , there are many more positive examples than negative examples available . Five different approaches were used to generate training datasets for investigating this issue: The performance of all classifiers on the training data sets is based on 10-fold cross validation . The confusion matrices for classifiers trained using datasets based on approaches 1–4 are shown in Table 1 and the classifier statistics are shown in Tables 2 and 3 . The classifier trained on the unbalanced dataset ( 111 positive examples and 34 negative examples ) has a classification accuracy of only 75 . 86% compared to the naïve approach of classifying all examples as positive which would result in a classification accuracy of 76 . 55% . The results for this dataset in Table 1 show that the resulting classifier predicts almost all examples to be positive . This highlights the problems encountered when using an unbalanced dataset . The classifier cannot distinguish positive and negative examples because the dataset contains so many more positive examples than negative examples and because many of the negative examples are analogs of the positives . The classifiers trained using both the dataset balanced with random peptides for negatives ( approach 2 ) and with biological peptides for negatives ( approach 3 ) had classification accuracies of 95 . 95% and 94 . 14% respectively , indicating that both classifiers exhibit a high degree of accuracy in discriminating between known cell-penetrating peptides and randomly generated or biological peptides assumed to be negative . The confusion tables for these classifiers on the training data sets ( Table 1 ) show that most of the mistakes are false negatives ( CPPs incorrectly classified as non-CPPs ) . The weakness of these training approaches is that some of the assumed negative examples may in fact be cell penetrating and known non-cell penetrating analogs of CPPs were not used as negative examples . When we used these trained classifiers to evaluate the known non-penetrating cell penetrating analog peptides ( our unbalanced test data set ) these classifiers obtained accuracies of 80 . 69% and 79 . 31% respectively . For both classifiers , approximately one third of the known non-penetrating peptides are classified as cell-penetrating . Most of the mistakes made by these two classifiers on the test data seem to be false positives , that is classifying a peptide with no cell penetrating potential as a CPP , and this classification of known non-penetrating cell penetrating analogs demonstrates that while these classifiers are very accurate distinguishing the features strongly predictive of cell penetrating potential from the vast majority of non-penetrating peptides , the features used for classification do not serve to distinguish between peptides more similar to CPPs that do not penetrate and those peptides that can act as CPPs . The classifier trained on the data set constructed using approach 4 ( random sampling with replacement from the known negative examples ) has a classification accuracy of 88 . 74% on the training data set when evaluated with 10-fold cross validation . When compared to the classification accuracy of the dataset generated using the unbalanced dataset , these results show that it is possible to classify a set of CPPs and a set of known non-penetrating peptides using our SVM based method when care is used to construct balanced datasets . Tables 2 and 3 show that 60% of the errors are false positives ( non-CPPs incorrectly classified as CPPs ) . When we evaluated the unbalanced test set on this classifier , an accuracy of 91 . 72% was obtained . The classifiers trained on the smaller datasets using approach 5 have an average classification accuracy of 78 . 82% using 10-fold cross validation . Approach 2 using randomly selected biological peptides as the negative examples gives the best 10-fold cross validation accuracy while approach 4 with random selection from the negative examples gives the best accuracy for the unbalanced training set . This suggests use of a two step process for screening . In the first step , a classifier trained with random biological peptides as the negative examples would be used for preliminary bulk screening . As a second step , peptides predicted to be CPP in step 1 would be screened by a classifier trained using approach 4 that is more accurate in distinguishing non-penetrating analogs from CPPs . Approach 4 also provides more insight into the rational design of novel CPP analogs as the negative examples used in this approach are generally constructed by the modification of a known CPP sequence . In Hällbrink et al . ( 2005 ) , the authors describe a method of CPP prediction based on scoring a candidate peptide according to z-score descriptors , features compiled through PCA , and report an 84 . 05% accuracy in the prediction of 53 CPPs and 16 non-functional CPP analogs [6] . A follow-up to this study , utilizing both more known CPPs ( 65 ) and more non-functional CPP analogs ( 20 ) , reports a 68% prediction efficiency using the same z-score descriptor based prediction method [7] . More recently , these z-score descriptors were utilized alongside quantitative structure-activity relationship features in an artificial neural network ( ANN ) to predict cell penetrating potential for a set of 101 peptides ( 77 CPPs , 24 non-penetrating CPP analogs ) and report a classification accuracy of 83% for the general ANN model constructed [8] . However , it should be noted that the data set utilized is composed of unbalanced classes , and an accuracy of 76 . 24% can be achieved by classifying every peptide encountered as a CPP . A comparison of these previously published prediction methods and our approach is presented in Table 4 . The models constructed using our approaches and their high classification accuracies indicate that using the primary biochemical properties of a peptide as features instead of synthesized feature values compiled using PCA allows for a more informative analysis of which properties determine whether a given peptide is cell-penetrating . In contrast to PCA approaches where complex features are constructed in both the feature selection step and again for classifier construction , our consistent use of the SVM for wrapper-based feature selection and for classifier construction , allows predictive models to be constructed to provide for more rapid and elucidative screening of cell-penetrating potential than previous predictive methods . For each classifier constructed , feature selection was conducted using a scatter search approach through feature space [12] where the “wrapped” classifier was the same type of SVM used for classifier construction . The classifier is a sequential minimal optimization SVM [13] using the Pearson Universal Kernel [14] . Table 5 lists the features selected for datasets 1–4 above . Because the number of training/testing samples for dataset 5 was so small , we generated ten different datasets using this approach . The features selected from these ten datasets are listed in Table 6 . The features selected for the datasets constructed using approaches 1–5 contain a number of properties previously shown to aid in the prediction of CPPs . These include net charge , positive charge , negative charge , the net donated hydrogen bonds , and the water-octanol partition coefficient . The low number of features selected for the datasets constructed using approach 5 indicates over-fitting of these small datasets by the classification algorithm . Therefore our detailed examination of features selected focused on datasets generated using approaches 1–4 . The primary amino acid composition features , the number of a given amino acid and the percent a given amino acid contributes to the whole peptide sequence , indicates no predictive function arising from the non-polar amino acids leucine and isoleucine , the polar amino acid glutamine , and the negatively charged amino acid glutamate . At least one of the amino acid composition features was selected for the remaining amino acids , with the most notable of these being the positively charged amino acids lysine , arginine , and histidine , and the negatively charged amino acid aspartate . In addition , the group of aromatic amino acids were selected to a notable degree , and the presence of some aromatic amino acids in the peptide sequence has been previously reported to be required for cell-penetrating potential [15] . To experimentally validate our feature selection methodology and classifiers , 250 random peptides were generated using a 0th order Markov model based on the chicken predicted proteome and were classified as penetrating or non-penetrating using the classifier trained on the dataset constructed using random peptides as negative examples . From these classifications , four peptides predicted to be cell-penetrating and two peptides predicted to be non-penetrating were selected for synthesis and FITC-labeling along with three known cell penetrating peptides used for positive controls , three peptides consisting respectively of only polar amino acids , only non-polar amino acids , and only of mixed polar and non-polar amino acids to serve as negative controls . In addition , a known non-penetrating peptide ( TP13 , a transportan analog [15] ) was selected for synthesis to serve as a minor validation for our set of known non-penetrating peptides .
A database of cell-penetrating peptides was constructed from the literature and from commercial vendor product lines [6] , [7] , [11] . A total of 111 cell-penetrating peptide ( CPP ) sequences were identified and used to create a database of positive examples ( Table 7 ) [6] , [7] , [11] . The average amino acid lengths of these CPPs ranged from 12 to 26 . Because very few peptides have been experimentally validated to be non-penetrating , it was more challenging to construct a database of negative examples . Five different strategies were used . Because our experimental system is avian , we have used the composition of the chicken proteome as the basis for two of our datasets . Previous research has demonstrated the importance of using a balanced training sets where there are approximately equal numbers of positive and negative examples [9] . For each dataset , we generate a set of basic biochemical properties of each peptide ( e . g . mass , size , charge , secondary structure , etc ) and other features previously shown to be useful in the prediction of CPPs ( e . g . steric bulk and net donated hydrogen bonds ) [7] . The full list of the initial 61 features is shown in Table 9 . We use these features directly in our machine learning algorithm rather than using composite features such as features derived by principle component analysis ( citation ) . We feel this approach will be more informative in the rationale design of CPPs cell penetrating peptides . Because the data values for each feature within a dataset vary greatly , NV normalization was used to scale the numeric range of all features in the range [0 , 1] [16] . The WEKA Machine Learning Toolkit Version 3 . 6 . 1 , a freely available software package containing a number of machine learning algorithms for data mining , was used for feature selection , classifier construction , and classifier evaluation [17] . We conducted feature selection to reduce the dimensionality of the feature vectors . Empirical evaluation of a number of different feature selection methods was conducted and the best performance was obtained using a wrapper-based method . The wrapper-based method uses a parallel scatter search algorithm [12] to evaluate feature subsets based on classifier performance . Scatter search is an evolutionary algorithm , but unlike other evolutionary algorithms ( e . g . genetic algorithms ) , the search for a local optimum is guided through the use of a reference set that acts to intensify and diversify the resulting features [12] . Local searches of features generated from the reference set are conducted , and informative and diverse features from these local searches are used to update the reference set until a terminating condition is met [12] . Our classifier is a support vector machine ( SVM ) trained via a sequential minimal optimization ( SMO ) algorithm used in conjunction with the Pearson VII universal kernel [13] , [14] . SVMs are supervised learning classifiers generally used for solving two class problems , and in their simplest form can be thought of as a classifier separating two classes mapped onto a 2-dimensional plane by generating a line through the plane that optimizes the distribution of each class on either side of the line [13] . The SMO algorithm is a modification to the original SVM learning algorithms that replaces a numerical quadratic programming step with an analytical quadratic programming step , allowing the algorithm to spend a greater portion of time on the decision function instead of the quadratic programming step . This greatly increases the speed of the SVM for classification and allows scaling for large datasets [13] . We chose to utilize SMO-based SVM classifiers because of their speed and performance for our two class problem of determining if given peptide is cell-penetrating or non-penetrating . A kernel function used in conjunction with an SVM allows the classifier to examine non-linear relationships between features by mapping the initial non-linear features into a highly dimensional space where the solution can be represented by a linear classification [14] . We chose the Pearson VII universal kernel ( PUK ) for our SMO-based SVM because PUK has been shown to provide either equal or better mapping than traditional SVM kernels , while serving as a robust and generic alternative to other kernel functions [14] . Accuracy for all classifiers was evaluated using 10-fold cross-validation . A 0th order Markov chain based on the amino acid frequency of the IPI Chicken Proteome ( ipi . CHICK . v3 . 56 ) [10] was used to generate 250 peptides . The classifier trained on our biologically based random peptide dataset was then used to classify each of these peptides . From these classification results , four peptides predicted to be cell penetrating and two peptides predicted to be non-cell penetrating were selected for synthesis and experimental validation . In addition , three peptides known to be cell-penetrating ( HIV-Tat [18] , Antennapedia [19] , and Pep-1 [20] ) were chosen to be positive experimental controls . Three other peptides , one of all polar amino acids , one of all non-polar amino acids , and one of a mix of polar and non-polar amino acids , were chosen as negative experimental controls because their lack of charged and aromatic R-groups make it unlikely they would cross a cellular membrane . One peptide ( TP13 [7] , [15] ) was randomly selected for synthesis from the list of known non-penetrating cell penetrating peptide analogs . All peptides selected for synthesis are shown in Table 10 . Peptides were synthesized ( >95% purity ) and N-terminally labeled with FITC , a fluorescent tag , by Biomatik . During the peptide synthesis , one of our chosen negative controls , negative-2 ( GLALLGIAVAILVVL-NH2 ) was unable to be synthesized to our desired purity levels due to insolubility issues and is not considered further . The lyophilized peptides were reconstituted using 1 mL of 4∶1 dd H2O sterile filtered 0 . 45 µm and acetonitrile ( EMD OmniSolv ) . Two avian cell lines , Quail SOgE muscle cells [21] and a primary culture of Chicken embryonic fibroblasts ( CEF ) , were grown in tissue culture flasks in Dulbecco’s minimal essential medium containing 10% fetal bovine serum with penicillin ( 200 IU/mL ) , streptomycin ( 200 µg/mL ) , amphotericin B ( 0 . 5 µg/mL ) ( MP Biomedicals ) , and non-essential amino acids at 37°C in a 5% CO2 atmosphere . Approximately 100 , 000 cells per well ( both CEFs and SOgEs ) were plated onto 12-well tissue culture plates approximately 2 days prior to the experiment and allowed to reach confluency . The cells were changed to serum free media and incubated for 60 minutes prior to experimentation . The cells were then washed with two 1 mL washes of PBS , after which they were exposed to 300 µL of 10 µM peptide in serum free media for 30 minutes , with three replicates per peptide per cell line . The cells were then washed with two 1 mL washes of PBS , and lightly trypsinated to remove any external peptides that may have been attached to the cellular membrane and facilitate the detachment of cells from the tissue culture flask . Centrifugation of the cells was performed at 250 x G for 4 min , and the supernatant aspirated off . Cells were then lysed with 250 µL of 0 . 1% Triton-X in PBS at 4° C for 10 minutes . A 100 µL aliquot of the cell lysate and a 100 µL aliquot of the 10 µM peptide in serum free media were pipetted onto a 96-well plate . Fluorescence was measured on a Dynex Fluorolite 1000 plate reader at 485/530 nm . The samples were compared to the fluorescence of the added amount of peptide and t-tests ( p≤0 . 05 ) were performed for each experimental sample against an untreated control . The SOgE cells were seeded onto glass tissue microscopy slides ( approximately 50 , 000 cells/well ) , and allowed two days to reach confluency . The cells were changed to serum free media and incubated for 60 minutes prior to experimentation . The cells were then washed with two 1 mL washes of PBS , after which they were exposed to 300 µL of 10 µM peptide in serum free media for 30 minutes . The cells were then washed with two 1 mL washes of PBS , and then fixed using UltraCruz™Mounting Medium ( Santa Cruz Biotechnology ) containing a DAPI nuclear stain . The fluorescence was examined using a Nikon Eclipse TE2000-U Inverted Research Microscope with the MetaMorph microscopy imaging software .
|
Cell penetrating peptides ( CPPs ) are peptides that can potentially transport other functional molecules across cellular membranes and therefore serve a role as drug delivery vehicles . The properties of a given peptide that make it cell penetrating are unclear , and the rapid screening of potential CPPs aids researchers by allowing focus on those peptides most likely to be utilized in a therapeutic capacity . This paper shows that basic features representing primary biochemical properties of these peptides can be used to train a classifier that can accurately predict cell penetrating potential of peptides and provide insight into the biochemical properties associated with cell penetration .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"algorithms",
"computer",
"science",
"biology",
"computational",
"biology",
"molecular",
"cell",
"biology"
] |
2011
|
Prediction of Cell Penetrating Peptides by Support Vector Machines
|
Appropriate nutrient response is essential for growth and reproduction . Under favorable nutrient conditions , the C . elegans nuclear receptor DAF-12 is activated by dafachronic acids , hormones that commit larvae to reproductive growth . Here , we report that in addition to its well-studied role in controlling developmental gene expression , the DAF-12 endocrine system governs expression of a gene network that stimulates the aerobic catabolism of fatty acids . Thus , activation of the DAF-12 transcriptome coordinately mobilizes energy stores to permit reproductive growth . DAF-12 regulation of this metabolic gene network is conserved in the human parasite , Strongyloides stercoralis , and inhibition of specific steps in this network blocks reproductive growth in both of the nematodes . Our study provides a molecular understanding for metabolic adaptation of nematodes to their environment , and suggests a new therapeutic strategy for treating parasitic diseases .
The evolutionary success of nematodes is derived from their ability to adapt different developmental pathways depending on environmental conditions . The best studied of these pathways exists in the free-living nematode Caenorhabditis elegans ( C . elegans ) . After hatching from eggs when enviromental conditions are favorable , C . elegans larvae continually develop through four stages ( L1–L4 ) , which eventually mature into reproductive adults . In contrast , in unfavorable environments , C . elegans larvae interrupt their reproductive growth by arresting at an alternative L3 stage termed dauer , which is characterized by developmental quiescence , stress resistance and a substantial extension of lifespan . Once conditions become favorable , the L3-dauers exit the developmental diapause and rapidly progress into the L4 stage through a series of metabolic and developmental changes that are governed by a coordinated transcriptional network [1] . Through this developmental adjustment , C . elegans is able to maximize its reproductive advantage under diverse environmental conditions [2] . Similar to free-living species like C . elegans , parasitic nematodes also alter their larval development based on environmental conditions [3 , 4] . Before host infection occurs , larvae of developing parasitic nematodes , such as hookworms and Strongyloides stercoralis ( S . stercoralis ) , arrest their reproductive growth at a dauer-like stage called infectious L3 ( iL3 ) . Then , upon infection of their hosts where environmental conditions favor the completion of the parasite’s life cycle , the arrested iL3 larvae resume reproductive growth and develop into fertile egg-laying adults . At the molecular level , the nematode development program is controlled by a hormonal signaling pathway initiated by insulin/IGF-I and TGF-β , which eventually converges in the activation of a nuclear receptor called DAF-12 [2 , 5 , 6] . In C . elegans , favorable environments stimulate insulin/IGF-I and TGF-β pathways that induce expression of DAF-9 , a cytochrome P450 enzyme that catalyzes the synthesis of steroid-like hormones , called dafachronic acids [2 , 5 , 6] . Dafachronic acids serve as ligands that bind and activate DAF-12 [5–7] , which in turn commit the nematode to reproductive growth . Conversely , in unfavorable environments , the insulin/IGF-I and TGF-β pathways remain inactive , preventing dafachronic acid synthesis , which in turn allows DAF-12 to interact with DIN-1 , a strong co-repressor that is required for dauer formation [5 , 6 , 8] . In C . elegans larvae lacking DAF-12 , this repressor activity is absent , causing a dauer-defective phenotype that would be expected to decrease viability in an unfavorable environment [2 , 9] . In parasitic nematodes , the insulin/IGF-I/DAF-12 signaling pathway controlling development appears to be conserved [10–15] . Similar to C . elegans , in parasitic nematodes ligand-free DAF-12 is required for formation of the dauer-like iL3 , whereas ligand-activated DAF-12 is required for reproductive growth [15] . In C . elegans , larvae undergoing reproductive growth or dauer diapause display distinct patterns of energy metabolism . L2–L4 larvae in reproductive growth exhibit aerobic energy metabolism by converting dietary energy sources ( carbohydrates and lipids ) into acetyl-CoA , which is then fed into the TCA cycle and oxidative phosphorylation [2 , 16 , 17] . This aerobic metabolism produces sufficient energy to support rapid , energy-demanding reproductive growth . In contrast , aerobic energy metabolism is greatly reduced in dauer larvae , which instead exhibit a slower rate of anaerobic energy metabolism . Paradoxically , however , anaerobic metabolism also utilizes fat metabolism to meet the nematode’s energy needs for survival during privation [18–21] . The pathways involved in anaerobic energy metabolism are the glyoxylate cycle , a variant of the TCA cycle that converts acetyl-CoA to malate , and malate dismutation , a fermentation pathway that metabolizes malate for energy production [2 , 16 , 17] . The tendency towards the lower rate of anaerobic metabolism in dauer larvae prevents premature depletion of energy reserves and facilitates extended lifespan [22] . Thus , two distinct types of metabolism are employed to produce energy during reproductive growth and diapause stages , raising the question as to how the two pathways are differentially regulated . Although the study of metabolism in parasitic nematodes is hampered by the difficulty in obtaining sufficient numbers , it is known that during reproductive growth in their hosts certain species of parasitic larvae migrate through the circulatory system , lungs and trachea , where aerobic conditions are high [3 , 4] . Furthermore , iL3 larvae of the hookworm Ancylostoma caninum are suggested to use fat reserves as an energy source [23] , and there is an inverse correlation between oxygen consumption and iL3 longevity in parasitic nematodes [24] . These findings suggest that similar mechanisms control developmental energy metabolism in free-living and parasitic nematodes . In the present study , we show that in addition to governing expression of developmental genes required for entry and exit from dauer diapause , DAF-12 is required for activating a metabolic network that is required for the normal progression to reproductive maturity . Efforts to elucidate the molecular targets of DAF-12 have focused mainly on the identification of heterochronic and microRNA genes that ensure the correct developmental decision is made during entrance and exit from dauer [25–28] , and on longevity genes that are repressed in long-lived mutant worms [29–31] . Notably , however , a role for DAF-12 in energy homeostasis has not been well documented . Utilizing a combination of biochemical and genetic approaches , we show that DAF-12 is a key transcriptional regulator of developmental energy metabolism . In C . elegans , DAF-12 induces expression of a gene network that is responsible for aerobic fat utilization during reproductive growth . Further , this DAF-12-dependent metabolic network is conserved in the parasitic nematode , S . stercoralis . This work provides a molecular understanding of how nematodes adjust energy metabolism to assure successful reproduction in wide-ranging environments , and it suggests a therapeutic strategy for treating parasitic diseases by inhibiting fat utilization .
To investigate the potential role of DAF-12 in regulating energy metabolism in C . elegans , we used a dauer defective double-null mutant lacking both the DAF-12 co-repressor din-1 and the dafachronic acid-synthesizing enzyme daf-9 [8] . Employing a mutant that lacks both din-1 and daf-9 commits C . elegans to constitutive reproductive growth even in the absence of dafachronic acids . The advantage of the din-1;daf-9 mutant is that it permits evaluation of the direct effects of DAF-12 activation on metabolism while at the same time minimizing effects due to the developmental switching that would otherwise occur in the single null mutant of daf-9 . We found that treating din-1;daf-9 larvae with the high affinity endogenous ligand , Δ7-dafachronic acid ( DA ) decreased triglyceride levels in a dose dependent manner ( Fig . 1A ) . This decrease was not due to reduced dietary nutrient uptake , since DA treatment had no effect on pharyngeal pumping rates of the larvae ( Fig . 1B ) but rather slightly increased dietary fatty acid uptake ( Fig . 1C ) . In contrast , DA treatment increased the fatty acid oxidation and oxygen consumption in din-1;daf-9 larvae ( Fig . 1D , E ) . At the same time , DA treatment did not significantly change either triglyceride levels , fatty acid oxidation , or oxygen consumption in mutants that lack DAF-12 ( din-1;daf-12 ) or in wild type N2 larvae ( Fig . 1A , D , E ) , indicating that the effects of DA on metabolism were DAF-12 dependent . We also examined whether DAF-12 activation affects reproduction . As shown in Fig . 1F , progression from L4 to the young adult stage , when reproductive organs become well-developed [32] , occurred earlier in din-1;daf-9 larvae treated with DA compared to vehicle in a DAF-12 specific manner . DA treatment also advanced the onset of egg laying , another marker of reproductive maturity ( Fig . 1G ) . Together , these findings demonstrate that DAF-12 activation induces aerobic energy metabolism and accelerates larval reproductive growth . To gain insight into the molecular mechanism underling the DAF-12-regulated fatty acid metabolism , we evaluated global changes in C . elegans gene expression by comparing vehicle and DA treated L3 larvae . Microarray analysis identified 796 genes that were up-regulated and 985 genes that were down-regulated ( >2-fold change and FDR<5% ) in response to DAF-12 activation ( S1 Table ) . The DA-regulated transcriptome was then grouped into several functional categories based on gene ontology ( DAVID , ref . [33] , S1 Table ) . In addition to the expected changes in expression of heterochronic and molting genes ( e . g . , dre-1 ) that coordinately regulate developmental and reproductive pathways [26] , DA governed expression of a distinct cadre of genes involved in the metabolism of lipids ( S1 Table ) . In contrast , no changes were observed in the expression of genes required for metabolizing glucose . We then compared the microarray data from our DA responsive genes with that of genes that have been shown to be regulated during the exit from dauer [1] , which is another process where DAF-12 is activated . As expected , the transcriptome of DA up-regulated genes significantly overlapped the trancriptome of genes up-regulated during dauer recovery ( S1A Fig . ) . These data indicate that DAF-12 engages a gene network that governs metabolism and growth during both reproductive development and dauer recovery . We also compared the DAF-12 transcriptome with genes that are regulated by the transcription factor DAF-16 [34] . Whereas DAF-12 activation suppresses dauer , activation of DAF-16 is known to promote dauer [2] . As expected by this reciprocal regulation of dauer , there was no statistically significant overlap between genes regulated by DAF-12 and DAF-16 ( S1B Fig . and S1C Fig . ) . Although this comparison was between the transcriptomes from different stages of worms ( L3 for DAF-12 vs . adult for DAF-16 ) , these results suggest that DAF-12 and DAF-16 regulate distinct gene networks to coordinate entry and exit from dauer diapause , and the initiation of metabolic pathways that promote reproductive development . To further investigate the metabolic network governed by DAF-12 , we used qPCR to confirm changes in expression of fatty acid metabolic genes identified in our microarray study , as well as other candidate genes known to be involved in fatty acid metabolism [19 , 20] . A total of 69 genes were evaluated ( S2 Table ) . A hallmark of the DA-regulated metabolic pathway that correlates to reproductive growth is the induction of aerobic fatty acid oxidation ( Fig . 1 ) . DA increased expression of genes involved in every aspect of aerobic fatty acid utilization , including lipolysis , transport , esterification , and oxidation in both peroxisomes and mitochondria ( Fig . 2A-E; S2 Table ) . To provide an additional objective assessment of DAF-12’s role in regulating energy metabolism , we compared the DA-regulated lipid metabolic gene profile to changes observed in response to fasting . In addition to reproductive growth , fasting is another physiological process that mobilizes and utilizes fatty acids [19 , 20] . However , in contrast to reproductive growth , fasting utilizes anaerobic metabolism marked by reduced metabolic rates [21] and activation of the glyoxylate cycle ( through icl-1 expression ) [19 , 20] . Of the 69 fatty acid metabolic genes tested above ( S2 Table ) , 20 were increased by DAF-12 and 37 were increased by fasting ( Fig . 2F , S2 Table ) . Importantly , there was no significant overlap ( based on hypergeometric distribution ) in the number of genes that were either up- or down-regulated under both conditions , demonstrating that DAF-12 and fasting engage distinct gene networks for fatty acid utilization . DA decreased the mRNA levels of icl-1 ( Fig . 2G ) , the bi-functional enzyme with isocitrate lyase and malate synthase activities that is unique to the glyoxlate cycle and thus serves for an indicator of anaerobic fatty acid utilization . Taken together with the biochemical measurements of fatty acid oxidation ( Fig . 1 ) , these results suggest that DAF-12 activation selectively governs the pathways that lead to energy mobilization and utilization by increasing expression of genes involved in aerobic lipid metabolism . We also investigated the signaling pathways that govern metabolism of DA . Interestingly , expression of daf-28 ( insulin/IGF-I-like ) , daf-7 ( TGF-β-like ) and daf-9 ( the DA biosynthesis enzyme ) were specifically suppressed by exogenous DA treatment , while expression of strm-1 , which quenches DAF-12 ligand synthesis [35] , was induced ( S2 Fig . ) . The ability of DAF-12 to repress its own activity is reminiscent of a classic endocrine feedback circuit that functions to maintain homeostasis . To determine whether the up-regulation of gene transcription by DA was through the direct action of DAF-12 , we analyzed the promoters of several of the DA-induced genes that were confirmed by qPCR . In the absence of a DAF-12 antibody to perform chromatin immunoprecipitation experiments , we used bioinformatics to search for the consensus DAF-12 DNA binding element [36] in the promoters/introns of five representative DAF-12-induced genes . Selection of these genes was based on their representation of different metabolic processes , their high levels of expression , their response to DA , and their distinct chromosomal locations ( i . e . , genes not likely to be in a gene cluster sharing a common promoter ) . Our analysis revealed 33 putative DAF-12 response elements ( S3 Table ) . We found that DAF-12 bound efficiently to 13 of these elements ( Fig . 3A , B; S3 Table ) and activated transcription in a standard cell-based reporter assay through four of them ( Fig . 3C; S3 Table ) . These four DAF-12 response elements corresponded to three ( K08B12 . 1 , acs-1 and acs-3 ) of the five genes originally selected for analysis . Consistent with these genes being direct transcriptional targets of DAF-12 , K08B12 . 1 and acs-3 expression was induced rapidly within 30 to 60 minutes after treatment of din-1;daf-9 larvae with DA ( Fig . 3D ) . These data suggest that at least a portion of the genes regulated by DAF-12 are likely to be direct targets . We next asked whether aerobic fatty acid utilization is required for DAF-12 to promote reproductive growth by inhibiting aerobic fatty acid utilization with etomoxir , a specific inhibitor of the carnitine palmitoyltransferases that mediate fatty acid transport into mitochondria [37] . Etomoxir treatment significantly decreased fatty acid oxidation and increased fat storage in C . elegans ( S3 Fig . ) , demonstrating the effectiveness of the drug in inhibiting this pathway in nematodes . A further consequence of etomoxir treatment was that it completely blocked the earlier onset of egg laying that is dependent on DA , which is a marker for reproductive growth ( Fig . 4A ) . Etomoxir treatment also prevented DA-mediated rescue of reproductive growth in the daf-7 and daf-9 mutants ( Fig . 4B , C ) and delayed the rescued growth in the daf-2 mutant ( Fig . 4D ) . In this latter mutant , the inability of etomoxir to inhibit growth completely is likely due to compensatory anaerobic fatty acid utilization that is known to occur in the daf-2 mutants [34 , 38] . Consistent with the data shown in Fig . 1F , DA treatment did not affect reproductive capacity in wild type N2 worms or in mutants lacking DAF-12 expression ( din-1;daf-12 , daf-9;daf-12 , and daf-7;daf-12 ) , regardless of the absence or presence of etomoxir ( S4 Fig . ) . In sum , these data demonstrate that aerobic fatty acid metabolism is required for DAF-12 to promote growth from larvae to reproductive adults in C . elegans . Like C . elegans , many species of parasitic nematodes such as S . stercoralis also use the conserved insulin/IGF-I and DAF-12 signaling pathways to regulate their development [10–15] . We therefore asked whether the role of DAF-12 in promoting fatty acid utilization is conserved during reproductive growth of S . stercoralis . To test this , we confirmed the presence of fat utilization genes in S . stercoralis and then examined whether they are regulated by DA ( Fig . 5A , S4 Table ) . As in C . elegans , expression of genes encoding a lipase ( Ss_F28H7 . 3 ) , acyl-CoA synthase ( Ss_acs-1 ) and a gene involved in acyl-CoA transport ( Ss_acbp-3 ) was induced , while the key glyoxylate cycle gene ( Ss_icl-1 ) was repressed by DA treatment . Expression of the carnitine palmitoyltransferase gene , Ss_W03F9 . 4 , required for mitochondrial β-oxidation , was also increased by DA ( Fig . 5A; S4 Table ) . In contrast , expression of genes involved in peroxisomal β-oxidation ( Ss_acox-2 , Ss_acox-3 and Ss_ech-8 ) was repressed by DA ( Fig . 5A; S4 Table ) . This profile of gene regulation supports the notion that DAF-12 activation induces aerobic fat utilization in S . stercoralis similar to that observed for C . elegans ( Fig . 2 ) . We also examined the effect of etomoxir on DA-induced reproductive growth in the post-free-living larvae of S . stercoralis , which typically arrest at the dauer-like iL3 stage . Although DA is less potent as an agonist for DAF-12 in S . stercoralis compared to C . elegans [15] , DA treatment was able to induce >85% maturation of S . stercoralis larvae to the L3-L5 stages ( Fig . 5B ) . Notably , co-treatment of etomoxir with DA resulted in a significant decrease in the number of L3-L5 larvae ( Fig . 5B ) . Administration of etomoxir to DA-treated worms also led to a marked increase in lethality of these L3-L5 larvae ( Fig . 5C ) . Importantly , treatment with etomoxir alone did not kill S . stercoralis iL3 larvae ( which are in an arrested developmental stage ) , demonstrating that the lethality observed in L3-L5 worms is due to the specific effects of etomoxir on the reproductive developmental program induced by DA ( Fig . 5D ) . These results reveal that aerobic fatty acid metabolism is required for DAF-12-dependent reproductive growth of S . stercoralis and suggest that control of this metabolic pathway by DAF-12 might be a promising strategy for regulating development in these important pathogens .
The nuclear receptor DAF-12 plays an essential role in C . elegans in linking nutritional status to developmental programs , including the transition from second- to third-stage larval development and the progression into and out of dauer diapause . In this report , we show that activation of DAF-12 by its hormonal ligand DA directly regulates energy homeostasis by inducing aerobic fatty acid utilization , which is required for reproductive growth . DAF-12 was shown previously to regulate heterochronic genes , which determine the stage-specific timing of cell type differentiation [25–27] . Interestingly , the DAF-12 gene regulatory network is distinct from that of DAF-16 , which is active in the absence of DA signaling to promote dauer and inhibit reproductive growth . Thus , DAF-12 controls the balance between reproductive growth and dauer both by controlling the expression of genes directing development , and by regulating the flux of nutrients required to fuel the developmental program . Our findings provide a molecular explanation for the longstanding observation that C . elegans switches to aerobic metabolism once reproductive growth has been initiated [2 , 16] . Under favorable environmental conditions , which promote the biosynthesis of DA , we showed that DAF-12 stimulates aerobic fatty acid utilization in larvae as evidenced by decreased storage of triglycerides and increased oxygen consumption and fatty acid oxidation ( Fig . 1A , D , E ) . Although a previous study originally suggested DAF-12 might increase rather than decrease fat storage [8] , it has been shown since that the lipid staining assay used in this previous study is non-specific and ineffective for determining fat content [39] . DA treatment also stimulates progression from L4 to the young adult stages and advances the onset of egg-laying activity ( Fig . 1F , G ) . Thus , DAF-12 coordinates the release of energy required to support the rapid , energy-intensive growth of larvae to reproductive maturity . At the molecular level , DA induces metabolic genes that regulate fatty acid utilization at multiple steps , including fatty acid mobilization , esterification , and peroxisomal and mitochondrial β-oxidation . The rate-limiting step enzyme in fatty acid oxidation , carnitine palmitoyltransferase , is encoded by several homologs in C . elegans , two of which are DAF-12 targets ( Fig . 2E ) and inhibition of the enzyme’s activity blocks DA-stimulated reproductive growth ( Fig . 4 ) . Concomitant with its regulation of genes involved in oxidative metabolism , DA represses the expression of icl-1 , which encodes a bifunctional enzyme in the glyoxylate cycle that is essential for anaerobic fatty acid catabolism ( Fig . 2G ) . These results support the role of DAF-12 in promoting reproductive growth of C . elegans by acting as a key controller of energy homeostasis in response to nutrient supply . The adaptive response to fasting is a process that also mobilizes fat storage to maintain energy homeostasis . However , in contrast to the metabolic pathway involved in reproductive growth , fasting results in an increase in anaerobic metabolism . Fasting decreases metabolic rate [21] and utilizes the glyoxylate cycle to provide energy from fatty acids [18–20] , which is diametric to the action of DA . The alternate role of DA in aerobic metabolism is supported by the finding that the metabolic gene networks induced by DAF-12 and by the fasting response are distinct ( Fig . 2G ) . Interestingly , the metabolic response to fasting is mediated by another nematode nuclear receptor , NHR-49 [19 , 20] . Analogous to the role of DAF-12 in the dauer diapause , NHR-49 is required for the entry and exit of adult reproductive diapause , a process that preserves reproduction in C . elegans during starvation [40] . Thus , DAF-12 and NHR-49 appear to control two separate gene networks that alternatively use aerobic and anaerobic fatty acid utilization to ensure successful reproduction under varying environmental conditions . Another key finding of the current study is that DAF-12-mediated regulation of energy homeostasis is conserved in the human parasite , S . stercoralis . As in C . elegans , DAF-12 activation stimulates the expression of genes involved in aerobic fatty acid utilization ( Fig . 5A ) . While genes involved in peroxisomal β-oxidation were induced in C . elegans and repressed in S . stercoralis , we note that peroxisomal β-oxidation can support either aerobic or anaerobic fatty acid catabolism . Importantly , however , blocking aerobic fatty acid utilization with etomoxir inhibited DAF-12-induced reproductive growth in S . stercoralis ( Fig . 5B-D ) . Inhibiting this metabolic pathway in other parasitic species using this type of strategy has recently been suggested [41] . To date , there is a very limited armamentarium of anthelmintic drugs that is effective against S . stercoralis , which can cause disseminated strongyloidiasis and multi-organ failure in infected humans . Although further studies in relevant host species are needed , our results suggest targeting metabolic enzymes may lead to a therapeutic approach for treating diseases caused by S . stercoralis and possibly other parasites [42] . To that end , it is interesting that etomoxir and other drugs that were originally developed to regulate fatty acid metabolism [37] as a means for treating diabetes and metabolic disease might be repurposed for treating parasitism . In summary , our studies reveal a novel facet of DAF-12 activity in both C . elegans and parasitic nematodes , namely the regulation of fatty acid catabolism and energy homeostasis . In this regard , DAF-12 is similar to the PPAR subfamily of nuclear receptors , which coordinately regulate fatty acid homeostasis and energy balance in vertebrates in response to nutrient availability . Our results provide a molecular explanation for how nematodes adjust energy homeostasis in response to changes in environmental conditions for reproduction . Moreover , they suggest a new strategy for developing new classes of anthelmintic drugs .
Δ7-DA was synthesized as described [6] . C . elegans strains din-1;daf-9 ( dh127;dh6 ) , din-1;daf-12 ( dh127;rh61rh411 ) and daf-9 ( dh6 ) were from Dr . Adam Antebi ( Max-Planck Institute for Aging ) ; wild type ( N2 strain ) , daf-2 ( e1368 ) and daf-7 ( e1372 ) worms were from C . elegans Genome Center ( University of Minnesota ) . daf-7;daf-12 ( e1372;rh61h411 ) mutant was made by crossing daf-7 hermaphrodites with hemizygous daf-12 males and was screened for dauer defective F2 progenies . The wild type ( UPD ) strain of S . stercoralis , used for developmental switching studies , and an iso-female line ( PV001 ) of this parasite , used for qPCR studies , were maintained as described [43] . Vehicle or Δ7-DA was mixed with 5× concentrated OP50 bacteria culture and loaded on NGM-agar plates . L1 larvae prepared by egg synchronization were cultured on these plates at 25°C for 22 . 5 h and the resulting L3 larvae were collected and washed in M9 buffer for the indicated assays . For triglyceride ( TG ) content , worms were sonicated and the resulting lysates were centrifuged at 13000×g at 4°C . From supernatants , total glyceride ( TGs plus free glycerol ) and free glycerol were measured by Infinity TG Reagent ( Thermo Sci ) and Free Glycerol Reagent ( Sigma ) , respectively . TG levels were calculated by subtracting free glycerol from total and were normalized to protein amounts in the lysates . Fatty acid oxidation was measured as described by the production of H2O from fatty acid [44] . Briefly , the L3 larvae from different treatment groups were incubated with a mix of cold and [3H]-palmitic acid ( Perkin Elmer ) complexed with fatty acid-free BSA ( Sigma ) , and incubated in a shaker at 25°C for 1 h . The reaction was terminated by adding 10% TCA followed by centrifugation at 13000×g for 5 min to obtain supernatant . Remaining [3H]-palmitic acid was deprotonated by adding 5N NaOH and PBS and removed by ion exchange column ( Dowex 1×8 200–400 mesh Cl , strongly basic , Sigma ) . [3H]2O left in the supernatant was measured by scintillation counting . For oxygen consumption , worms were mixed with antibiotic-killed OP50 bacteria , and then transferred to Oxygen Biosensor Plates ( BD Bioscience ) for oxygen measurement . Oxygen consumption was expressed as the increase of fluorescence units ( ΔFU ) and was normalized by protein amounts of the worms . For fatty acid uptake , worms were mixed with OP50 bacteria and 250 nM of fluorescent tracer ( C1-BODIPY-C12 fatty acid , Invitrogen ) . Following 1 h incubation at 25°C , worms were washed , mounted , and photographed under fluorescence microscopy . Fluorescence density units ( FU ) of each worm were quantified by the software Image-J . Pharyngeal pumping rates were measured as described [45] . Briefly , din-1;daf-9 L3 larvae were transferred to a fresh NGM plate with OP50 bacteria lawn and were videotaped through a stereomicroscope . Pumping rates were measured by counting the grinder movements and presented as pumps per minute . For each treatment , 10 L3s were assayed . Reproductive growth of C . elegans was measured by L4-young adult transition or by egg laying assays . For L4-YA transition assay , synchronized L1 larvae from were cultured on NGM-agar plates pre-loaded with 5× concentrated HT115 bacteria culture . Worms were grown at 20°C and young adults were counted at indicated time points . Data were presented as the percentage of young adults in whole populations . For the egg-laying assay , synchronized din-1;daf-9 L1 larvae were grown at 25°C on NGM-agar plates pre-loaded with 5× concentrated OP50 bacteria culture . At the indicated time points , 10~15 worms were transferred to fresh plates with bacterial lawns for 2 . 5 h and laid eggs were counted . Data were presented as numbers of the eggs laid by each worm per hour . For C . elegans experiments , synchronized L1 larvae were treated with vehicle or Δ7-DA for 22 . 5 h at 25°C , or synchronized wild type L1 larvae were cultured to L4 stage and then harvested as fed worms or were deprived of food for an additional 12 h to obtain fasted worms . For S . stercoralis experiments , iL3s worms were treated with or without Δ7-DA in M9 buffer at 37°C and 5% CO2 in air for 24 h . Total RNA from worms was extracted with TRIzol Reagent ( Life Technologies ) , and analyzed by qPCR . Relative mRNA levels were normalized to expression of reference genes inf-1 or ama-1 ( C . elegans ) or 18S ribosomal RNA ( S . stercoralis ) . Data were presented as fold changes of relative mRNA levels in DA versus vehicle treated worms or in fasted versus fed worms . Total RNA was also subjected to the C . elegans Genome Array ( Affymetrix ) for whole genome gene expression analysis . Briefly , gene expression values were log2 transformed and genes with >10-fold difference between replicates in either of the treatments were removed from our analysis . To identify the differentially expressed genes , we applied Significance Analysis of Microarrays ( SAM ) analysis using the R package samr [46] . Genes with median false discovery <5% and fold changes >2 . 0 were considered differentially expressed . DAF-12 proteins were prepared with TNT Quick-Coupled Transcription/Translation System ( Promega ) and blocked with poly-[dI-dC] and non-specific single-stranded oligos . The DAF-12 proteins were then incubated with [32P]-end-labeled dsDNA probes ( S3 Table ) at room temperature for 30 min and binding to DAF-12 was analyzed by 5% PAGE followed by autography . For competitive binding experiments , 20- or 200-fold excesses of unlabeled DNA probes were also included in the binding reaction . Co-transfection and luciferase reporter assays were performed as described in HEK 293 cells [15] . Eight hours post-transfection , cells were treated with vehicle or 1 μM Δ7-DA , and luciferase and β-galactosidase activities were then measured 16 h later . Relative luciferase units ( RLU ) were normalized to β-galactosidase activity . Reporter plasmids were constructed by inserting DAF-12REs and their 10-bp genomic flanking sequences into a TK-luc reporter plasmid . Synchronized din-1;daf-9 L1 larvae were grown in 5×concentrated OP50 in liquid suspension and shaken at 25°C for 15 h . Resulting L2 larvae were treated transiently with vehicle or 500 nM Δ7-DA and harvested in ice-cold M9 buffer . Cell nuclei were extracted and incubated with ATP , CTP , GTP and 5’-Bromo-UTP ( BrUTP ) at 30°C for labeling of nascent RNAs with BrUTP ( BrUTP-RNAs ) . The BrUTP-RNAs were then enriched with anti-BrUTP agorase beads ( Santa Cruz ) and quantified by qPCR . Development switching assays were performed as described [5 , 15] . Briefly , synchronized L1 larvae ( C . elegans ) or eggs ( S . stercoralis ) were grown on NGM-plates pre-loaded with etomoxir and phenotypes were observed after 60 h incubation at 25°C . Data from three independent experiments were pooled and significance was determined by Fisher’s exact test . S . stercoralis homologs were identified as reported [43] by a TBLASTN ( NCBI ) search of C . elegans versus S . stercoralis ( 6 December 2011 draft; ftp://ftp . sanger . ac . uk/pub/pathogens/HGI/ ) databases , followed by annotation to RNA-seq data ( ArrayExpress accession number E-MTAB-1164 ) . Phylogenetic tree analyses were constructed to resolve gene homology . S . stercoralis genes with 1:1 homology to C . elegans genes were identified as homologous genes . Unless otherwise stated , data were expressed as mean ± SD or SEM and significance tests between vehicle- or DA-treated groups were performed by Student’s t-test . The statistic tests of overlap between two gene sets were based on hypergeometric distribution and calculated by the R function “phyper ( ) ” ( https://stat . ethz . ch/R-manual/R-patched/library/stats/html/Hypergeometric . html ) .
|
Animals adjust their internal biological processes in response to their environments . In this study , we report that in a nutrient rich environment the free-living nematode , Caenorhabditis elegans , induces an energy-generating metabolic pathway to govern its reproductive growth by activating the nuclear receptor , DAF-12 . By responding to its endogenous ligands , called dafachronic acids , DAF-12 induces oxidation of lipids to produce the energy necessary to support growth and reproduction; and likewise , in the absence of dafachronic acids , DAF-12 prevents activation of this pathway . Through gene expression analysis , we show that DAF-12 regulates a network of genes involved in energy homeostasis and lipid metabolism . Given that dafachronic acids are produced only in well-fed worms , we conclude that DAF-12 functions as an environmental sensor that coordinately governs energy homeostasis . Through analogous studies in the incurable human parasite , Strongyloides stercoralis , we demonstrate that this pathway is conserved and that blocking it compromises the viability of the parasites . These findings elucidate a molecular mechanism for how nematodes govern their energy needs in response to the environment , and provide a potential new strategy for treating nematode parasitic diseases .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
The Nuclear Receptor DAF-12 Regulates Nutrient Metabolism and Reproductive Growth in Nematodes
|
Schistosomiasis is endemic to several parts of the world . Among the species that affect humans , Schistosoma mansoni is one of the most common causes of illness . In regions where schistosomiasis mansoni is endemic , reinfection is responsible for the emergence of hepatosplenic schistosomiasis ( HSS ) with portal hypertension in about 10% of infected individuals . Regardless of its etiology , portal hypertension may bring about the formation of arteriovenous fistulas and pulmonary vascular dilation , thus constituting a pulmonary shunt and its presence has been associated with the occurrence of neurological complications . The objective of this study was to identify pulmonary shunt using TTCE in patients with HSS and esophageal varices , and to compare the abdominal ultrasound and endoscopy findings among patients with and without pulmonary shunt . In this case series , a total of 461 patients with schistosomiasis mansoni were prospectively evaluated using abdominal ultrasound and endoscopy and 71 presented with HSS with esophageal varices . Fifty seven patients remained in the final analysis . The mean age of the patients was 55 ± 14 years , and 65% were female . Pulmonary shunts were observed in 19 ( 33 . 3% ) patients . On comparing the groups with and without pulmonary shunt , no significant differences were observed in relation to the abdominal ultrasound and endoscopic findings . When comparing the two subgroups with pulmonary shunts ( grade 1 vs grades 2 and 3 ) , it was observed that the subgroup with shunt grades 2 and 3 presented with a significantly higher frequency of an enlarged splenic vein diameter ( >0 . 9 cm ) , and an advanced pattern of periportal hepatic fibrosis ( P = 0 . 041 and P = 0 . 005 , respectively ) . None of the patients with pulmonary shunts had severe neurological complications . Our findings suggest that in HSS with esophageal varices the pulmonary shunts may be present in higher grades and that in this condition it was associated with ultrasound findings compatible with advanced HSS .
It is estimated that more than 200 million people in the world are infected with Schistosoma and more than 700 million remain at risk of infection , according to World Health Organiztion reports [1 , 2] . Of the species of Schistosoma that infect humans , the two that most frequently cause illness are Schistosoma haematobium , found in Africa and the Middle East , and Schistosoma mansoni , found in parts of Africa , the Middle East , and the Americas [3] . In Brazil , Schistosoma mansoni infection affects all of the Northeastern states and parts of the North , South , Southeast , and Mid-West , and it is endemic in nine states [4 , 5] . In these regions , reinfection is frequent and around 10% of the infected individuals develop severe forms of the disease , such as hepatosplenic schistosomiasis ( HSS ) , while in hospital-based samples this percentage may reach higher values [6] . Generally , HSS presents with elevated pressure in the portal venous system , secondary to the deposition of Schistosoma mansoni eggs in the intrahepatic branches of the portal vein , which promotes an intense inflammatory process followed by periportal liver fibrosis ( Symmers’ fibrosis ) , splenomegaly and gastroesophageal varices [7 , 8] . In portal hypertension , regardless of the etiology , an imbalance in the hepatic production of angiogenic and vasoactive substances , coupled with a possible genetic predisposition , promotes the formation of pulmonary vascular dilations and pulmonary arteriovenous fistulas [9–11] . These may be responsible for the deviation of part of the blood from the pulmonary arteries directly into the pulmonary veins and the left chambers of the heart [12 , 13] . This diverted blood reaches the systemic circulation without passing through the pulmonary capillaries , thus not undergoing the filtering function that the capillaries perform [12 , 13] . This situation is characterized by non-physiologic pulmonary shunting , which may be responsible for complications due to paradoxical embolization , such as transient ischemic attack , ischemic stroke , brain abscess , or hemoptysis and hemothorax . The latter two being secondary to intrabronchial or intrapleural rupture of the fistulas [14–16] . This type of shunt differs from the so-called physiologic pulmonary shunt , which is not associated with any clinical complications , due to the deviation of a minimal volume of blood from the bronchial arteries directly into the pulmonary veins without passing through the pulmonary capillaries , in addition to the coronary venous blood drained directly into the left ventricle [13] . In schistosomiasis , the pathophysiological mechanism of non-physiological pulmonary shunt has yet to be elucidated . It is speculated herein that in patients with HSS and esophageal varices , pulmonary shunts may occur through the same mechanism observed in patients with portal hypertension secondary to other causes , with the formation of pulmonary vascular dilatations and arteriovenous fistulas . Another mechanism could be the migration of eggs from Schistosoma mansoni to the pulmonary vessels through portosystemic collateral vessels or through dilatation of the liver sinusoids [17–18] . This may result in local necrosis and the formation of pulmonary arteriovenous fistulas , as described in a single one case report [19] . Non-physiological pulmonary shunts are frequently investigated through transthoracic contrast echocardiography ( TTCE ) in portal hypertension due to hepatic cirrhosis as part of the assessment for liver transplantation [20 , 21] . In hereditary hemorrhagic telangiectasia this is performed to determine the grade of intensity with the aim of preventing neurological complications , but it has been poorly studied in patients with schistosomiasis [16] . The objective of this study was to identify pulmonary shunts in patients presenting with HSS and esophageal varices using TTCE , and to compare the abdominal ultrasound and endoscopic findings between patients with and without pulmonary shunt . The secondary objective was to compare the abdominal ultrasound and endoscopic findings between the subgroups of patients with different shunt grades .
The study was approved by the research ethics committee of the Federal University of Pernambuco Center for Health Sciences ( Protocol no . 396/2010 ) . All participating patients signed the informed consent forms . In this case series , patients previously diagnosed with Schistosoma mansoni infection ( referred from primary health care services from all municipalities in the Pernambuco State ) were prospectively screened for pulmonary shunts at a schistosomiasis mansoni outpatient clinic at a tertiary hospital ( Hospital das Clínicas at the Universidade Federal de Pernambuco , HC-UFPE ) , between December 2010 and December 2012 . Patients initially underwent an abdominal ultrasound in order to define the clinical form of schistosomiasis mansoni . Patients presenting with HSS were referred to undergo upper gastrointestinal endoscopy . The exclusion criteria for this phase of the study were: seropositivity for hepatitis B or C virus , alcohol over the limit of 60 g/day for males and 40 g/day for females during the previous six months , concomitant diagnosis of liver cirrhosis , splenectomy , previous diagnosis of ventricular dysfunction with a left ventricular ejection fraction ≤35% using echocardiogram and presence spirometry with a severe obstructive disorder or severe restrictive disorder . Patients with HSS and esophageal varices who met the inclusion criteria were referred for TTCE .
Four hundred and sixty one patients infected with schistosomiasis mansoni were initially assessed by abdominal ultrasonography . Three hundred and ninety were excluded either because they did not present with HSS or because they fulfilled at least one of the exclusion criteria . Seventy-one patients with HSS and esophageal varices were selected for TTCE . After performing TTCE , 14 patients were excluded: 4 because of a poor acoustic window , 1 for presenting with an atrial septal defect , 1 because of a patent foramen ovale , and 8 failed to complete the study . Therefore , the final study sample comprised 57 patients ( Fig 2 ) . The mean age of the patients was 55 ± 14 years , and the majority ( 64 . 9% ) were female ( Table 1 ) . Of the 57 patients , 35 ( 61 . 4% ) presented with previous gastrointestinal bleeding due to ruptured esophageal varices , 28 ( 50 . 9% ) with portal vein cross-sectional diameter > 1 . 2 cm , 31 ( 58 . 5% ) with splenic vein cross-sectional diameter > 0 . 9 cm , 24 ( 42 . 1% ) with collateral vessels , 40 ( 70 . 2% ) with advanced or very advanced patterns of periportal hepatic fibrosis , 36 ( 63 . 2% ) with medium or large caliber varices and 40 ( 70 . 2% ) with portal hypertensive gastropathy ( Table 1 ) . Pulmonary shunts were identified by TTCE in 19 ( 33 . 3% ) of the 57 patients , of whom , 8 ( 42 . 1% ) were classified as grade 1 , 8 ( 42 . 1% ) as grade 2 , and 3 ( 15 . 8% ) as grade 3 ( Fig 3 ) . No pulmonary shunt grade 4 was recorded . When the groups with and without pulmonary shunts were compared , no significant differences were observed in relation to the mean age , gender , previous gastrointestinal bleeding , or abdominal ultrasound and endoscopic findings ( Table 1 ) . The abdominal ultrasound and endoscopic findings were compared between the two subgroups of patients with grade 1 pulmonary shunts versus those with grades 2 and 3 . The mean values for the longitudinal diameter of the spleen , cross-sectional diameter of the portal vein , and cross-sectional diameter of the splenic vein were similar in both subgroups . There were also no significant differences in the presence of collateral vessels and in cross-sectional portal vein diameters >1 . 2 cm . There was a significantly higher frequency of splenic veins with a transversal diameter >0 . 9 cm and of an advanced or very advanced pattern of periportal liver fibrosis in the subgroup of patients with grade 2 and 3 pulmonary shunts when compared to the grade 1 subgroup ( P = 0 . 041 and P = 0 . 005 , respectively ) . There were no significant differences between the two subgroups in relation to the endoscopic parameters ( Table 2 ) . No neurological complications were recorded in the 19 patients with pulmonary shunts .
In schistosomiasis , echocardiography is often performed as a screening test for the diagnosis of portopulmonary hypertension , yet TTCE is not currently included in the guidelines for assessment of these patients [25 , 26] . In our study , we evaluated patients with HSS and esophageal varices , one of the significant findings was the diagnosis of pulmonary shunts in more than 30% of patients , and most of them ( 57 . 9% ) classified with grade 2 and 3 shunts . However , at the time of collecting data on each patient there was no record of neurological complications that could be related to the diagnosis of a pulmonary shunt . Studies evaluating hereditary hemorrhagic telangiectasia patients have observed severe neurological complications in about 20% of patients with higher grades of pulmonary shunts [14 , 27] . We emphasize that our case series study was a cross-sectional rather than a follow-up study , which may have impaired the observation of such complications . Although we have not reported severe neurological complications , the finding of grade 2 and 3 pulmonary shunts in HSS could indicate that consideration needs to be given to performing TTCE for diagnosing pulmonary shunts , in addition to screening for pulmonary hypertension . Another prominent finding was the observation of a significantly higher frequency of abdominal ultrasound findings compatible with more advanced clinical forms of HSS in the subgroup of patients with higher grades of pulmonary shunts . It is noteworthy that the findings of our study differ from those of study using scintigraphy to diagnose pulmonary shunts in patients with HSS , in which the presence of pulmonary shunts was not associated with abdominal ultrasound parameters compatible with more severe forms of HSS [28] . Given that patients with more advanced periportal fibrosis progress to higher levels of portal pressure [29–31] , our findings seem consistent with the current knowledge that associates portal hypertension with the presence of pulmonary shunts [32] . It is worth noting that abdominal ultrasound is not only used to diagnose the clinical form of schistosomiasis , but also to monitor progression of the disease in patients frequently exposed to water contaminated by Schistosoma mansoni [29 , 31] . Thus , abdominal ultrasound could also be a useful screening tool for TTCE , especially in areas with limited resources that do not have easy access to TTCE . This could be initially indicated for patients with advanced periportal fibrosis and an enlarged splenic vein . These abdominal ultrasound findings are common in patients with esophageal varices who need to be referred to more advanced medical centers to undergo specialized treatment of portal hypertension , centers where TTCE may also be available . Another point to be discussed in relation to the abdominal ultrasound findings of this study is the fact that no significant differences were observed in the frequency of portosystemic collateral vessels , neither when comparing patients with and without pulmonary shunts nor between the subgroups of patients with grade 1 pulmonary shunts versus grades 2 and 3 . In fact , the presence of collateral vessels was observed in 7 of the 19 patients with pulmonary shunts , and in 5 of the 11 patients with grade 2 and 3 pulmonary shunts ( Fig 4 ) . Indeed , this finding could be explained by the possibility that pulmonary shunts in patients with schistosomiasis may occur through a pathophysiological mechanism similar to that described in patients with portal hypertension secondary to other causes , and not by migration of parasite eggs through portosystemic collaterals . We also speculate that the formation of collateral vessels could work as a protective factor against the occurrence of pulmonary shunt , since it is a physiological mechanism that can reduce portal pressure levels in these patients . We must also consider the limitations of this study . The fact that this is a case series , even though it had an analytical character because we included as controls patients without shunts and then the subgroup of grade 1 shunts , is a limitation . This study design allows us to establish hypotheses , but not to draw definitive conclusions on a causal association between the abdominal ultrasound variables ( advanced periportal fibrosis and enlarged diameter of the splenic vein ) and the presence of pulmonary shunts . These hypotheses may be tested in further studies with an analytical design . Another limitation refers to patient selection , which was restricted to those with both HSS and esophageal varices . On the one hand , this inclusion criterion guaranteed the presence of portal hypertension in all patients . On the other , the presence of esophageal varices in all patients precluded the analysis of the association between the presence of esophageal varices and the pulmonary shunt . What we evaluated was whether the greater caliber of esophageal varices ( medium/large ) would be associated with the presence of the most severe grades of shunting ( grades 2 and 3 ) , which was not observed . Therefore , this inclusion criterion also limited the conclusions inferred by our findings to HSS with esophageal varices , and it would be not appropriate to extrapolate these results to all clinical presentations of schistosomiasis mansoni . A control group with HSS and no esophageal varices would also be needed to check for an association between the presence of esophageal varices and pulmonary shunts . Finally , one further limitation may have been the fact that there was no previous study that defines the grade of left atrial opacification that may be considered a physiological pulmonary shunt through the semi quantitative technique of TTCE specifically for schistosomiasis mansoni . We attempted to minimize this fact by dividing the patients with pulmonary shunts into subgroups for analysis , where the subgroup with the least intense shunt , grade 1 , was considered the control group . In conclusion , our findings suggest that in HSS with esophageal varices pulmonary shunts may be present in higher grades . These findings may stimulate a discussion on performing TTCE in HSS with esophageal varices . In areas with limited resources , abdominal ultrasound findings compatible with advanced HSS could be used as screening parameters to perform TTCE . Follow-up studies in patients with HSS and pulmonary shunts may be useful to observe if neurological complications occur and how often they are present , in addition to observing whether they are associated with the presence of higher grades of pulmonary shunting .
|
Among the species of Schistosoma that infect humans Schistosoma mansoni is one of the most common causes of illness . In the areas where schistosomiasis mansoni is endemic , around 10% of infected individuals develop hepatosplenic schistosomiasis ( HSS ) with portal hypertension . Portal hypertension may promotes an imbalance in the hepatic production of vasoactive substances , which may act on the lungs promoting the formation of arteriovenous fistulas and pulmonary vascular dilation , a condition that is called a pulmonary shunt . When the pulmonary shunt is of higher grades , small thrombus or septic emboli that would normally be filtered through the pulmonary capillaries reach the left heart and the systemic circulation , which can lead to neurological complications . We found pulmonary shunts in patients with HSS and esophageal varices and we also found that patients with higher grades of pulmonary shunts presented with a significantly higher frequency of advanced periportal fibrosis and an enlarged splenic vein diameter . No neurological complications were observed . Our findings suggest that pulmonary shunts may be present in patients with HSS and esophageal varices . The abdominal ultrasound findings compatible with advanced HSS could be used as screening to investigate pulmonary shunt .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"schistosoma",
"invertebrates",
"schistosoma",
"mansoni",
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"radiology",
"ultrasound",
"imaging",
"helminths",
"cardiovascular",
"anatomy",
"tropical",
"diseases",
"fibrosis",
"parasitic",
"diseases",
"animals",
"liver",
"diseases",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"veins",
"developmental",
"biology",
"endoscopy",
"heart",
"gastroenterology",
"and",
"hepatology",
"neglected",
"tropical",
"diseases",
"portal",
"hypertension",
"research",
"and",
"analysis",
"methods",
"cardiac",
"atria",
"imaging",
"techniques",
"blood",
"vessels",
"helminth",
"infections",
"schistosomiasis",
"radiology",
"and",
"imaging",
"portal",
"veins",
"diagnostic",
"medicine",
"anatomy",
"biology",
"and",
"life",
"sciences",
"abdominal",
"ultrasound",
"organisms"
] |
2017
|
Pulmonary shunts in severe hepatosplenic schistosomiasis: Diagnosis by contrast echocardiography and their relationship with abdominal ultrasound findings
|
Arabidopsis seeds rapidly release hydrophilic polysaccharides from the seed coat on imbibition . These form a heavy mucilage layer around the seed that makes it sink in water . Fourteen natural Arabidopsis variants from central Asia and Scandinavia were identified with seeds that have modified mucilage release and float . Four of these have a novel mucilage phenotype with almost none of the released mucilage adhering to the seed and the absence of cellulose microfibrils . Mucilage release was modified in the variants by ten independent causal mutations in four different loci . Seven distinct mutations affected one locus , coding the MUM2 β-D-galactosidase , and represent a striking example of allelic heterogeneity . The modification of mucilage release has thus evolved a number of times independently in two restricted geographical zones . All the natural mutants identified still accumulated mucilage polysaccharides in seed coat epidermal cells . Using nuclear magnetic resonance ( NMR ) relaxometry their production and retention was shown to reduce water mobility into internal seed tissues during imbibition , which would help to maintain seed buoyancy . Surprisingly , despite released mucilage being an excellent hydrogel it did not increase the rate of water uptake by internal seed tissues and is more likely to play a role in retaining water around the seed .
Polysaccharides released from the seed coat on imbibition form a sticky , gelatinous halo called mucilage around the seed . This property , termed myxospermy , was observed in cress ( Lepidium sativum ) by Darwin [1] and is found in the model plant Arabidopsis . In addition to the Brassicaceae , this trait has been noted in a hundred plant families including Solanaceae , Linaceae and Plantaginaceae [2] . During seed development in Arabidopsis , the epidermal cells of the seed coat undergo a complex differentiation process during which mucilage polysaccharides are accumulated [3]–[5] . The resulting epidermal cells of mature Arabidopsis seeds have a distinctive morphology with reinforced radial cell walls connected to a column of secondary cell wall material at their centre , called the columella , which is surrounded by dehydrated mucilage polysaccharides under a primary cell wall . Genes involved in the differentiation of these cells and the production of mucilage have mainly been identified through mutant phenotypes [2] . In Arabidopsis , seed mucilage forms two structurally distinct layers with the pectin domain rhamnogalacturonan I ( RG I ) the major component of each [4] , [6] , [7] . The outer mucilage layer is diffuse and water-soluble . In contrast , the inner layer adheres strongly to the seed surface and requires harsh chemical treatment , or enzyme digestion , to remove it from the seed coat [4] , [7]–[9] . RG I attachment to the seed coat requires cellulose , as cesa5 , fei2 and sos5 mutants implicated in cellulose synthesis have reduced adherent mucilage [10]–[12]; cesa5 is affected in a cellulose synthase catalytic subunit , fei2 is defective in a leucine-rich receptor kinase and sos5 carries a mutation in a fasciclin-like arabinogalactan protein with a glycophosphatidylinositol anchor . In cesa5 null mutants some cellulose was still observed within the reduced layer of adherent mucilage , implicating other CESA genes in its production . Precisely how pectin and cellulose interact to form the adherent mucilage layer has still to be determined . The ecophysiological role of mucilage production by seeds is ambiguous , diverse functions have been put forward , but none appears to be comprehensively applicable . The adhesive properties of mucilage led to proposals that it mediates long-distance seed dispersal by attachment to animals or that it prevents seed removal during soil erosion or by ants through fixation to soil particles [1] , [13]–[15] . Comparison of the formation of mucilage in Artemisia taxa associated the trait with dry habitats , as had previously been observed in Lamiaceae [16] , [17] . Nevertheless , a potential role of mucilage in modifying germination capacity [6] , [18]–[20] has not been consistently observed in tests with mutants defective for mucilage release [21]–[24] . Furthermore , differences in the composition and structure of mucilage layers could reflect specific physiological roles for each [25] . Naturally occurring genetic variation provides an alternative source of mutations for functional analysis and gene cloning to that of induced mutations . In Arabidopsis thaliana a large number of accessions are available that have been derived from seeds harvested in the wild in a variety of geographical locations . These have generally been exploited for quantitative trait locus ( QTL ) mapping of important agronomic traits [26] . In a previous study we identified a naturally occurring mutation in the Shahdara accession that affects the liberation of mucilage from the seed coat [25] . The Shahdara accession is defective in the MUM2 β-D-galactosidase . This enzyme trims galactan ramifications from RG I in seed mucilage , rendering it more hydrophilic , and increasing mucilage expansion on imbibition so that the outer cell wall breaks and mucilage is released [25] , [27] . Except for the mucilage extrusion defect , no other visible phenotype was reported for the Shahdara accession or other mum2 mutants . Although the genetic basis of this phenotype was elucidated , its ecological relevance was not resolved . Seeds from the Shahdara accession had been collected in Tajikistan , near to the Shokhdara River . The collection of accessions previously screened for mucilage release defects contained mainly European accessions with few from central Asia [25] . In this study , we analysed a larger panel of Arabidopsis accessions and identified a further nine genotypes that were defective in mucilage release; these were observed to float on the surface of water , unlike seeds with a thick layer of adherent mucilage that sink . A screen for seed flotation identified four other accessions whose seeds floated despite releasing mucilage , as they had little adherent mucilage . Analysis of the causal mutations responsible for the modified mucilage release observed in the variants indicated ten independent mutations that affected a minimum of four different loci and these had evolved in populations from two geographical zones , central Asia and Scandinavia . The investigation of water uptake by seeds using nuclear magnetic resonance ( NMR ) relaxometry showed that non-released mucilage polysaccharides contribute to the maintenance of buoyancy , and released mucilage does not improve imbibition , contrary to previous hypotheses .
During an expedition to central Asia , seeds were harvested from 25 plants at a site in Tajikistan believed to correspond to that where the original Shahdara population was collected , these were termed NeoShahdara ( Neo ) and have been shown to be closely related to the original Shahdara individual [28] . Interestingly , descendants from just eight of the twenty-five Neo plants tested had the mum2Sha deletion , identified as the causal mutation for mucilage retention in the original Shahdara [25] . Individuals from two Kyrgyzstan and six Scandinavian populations were also found to be defective in mucilage release [29] ( Table 1 ) ; the trait was fixed in 7 of the 9 populations . In 488 accessions examined representing the species-wide distribution of Arabidopsis , 71 exhibit the mucilage release defect [25] ( Table 1 ) . The sites of origin for these accessions were confined to central Asia and Scandinavia . Crosses were carried out between Shahdara and ten representative individuals from among the 70 new mutant accessions ( Table S1 ) . For Neo , a representative carrying the mum2Sha deletion , Neo-3 , and one that did not , Neo-6 , were used . Dja-1 and Sk-1-1 complemented the Shahdara mum2 mutant phenotype . Furthermore , reciprocal crosses between Dja-1 and Sk-1-1 showed that their mutations were not allelic . The gene affecting mucilage release in Dja-1 has been identified and codes the pectin methylesterase inhibitor PMEI6 [29] . A combination of mapping and whole-genome sequencing identified a point mutation in At3g50990 in the Sk-1-1 accession; this gene codes PEROXIDASE36 ( PER36 ) and the mutation causes the conversion of Tyr-262 to a stop codon . PER36 has recently been shown to be required for mucilage release [30] . All the remaining accessions failed to complement the Shahdara mum2 mutant allele . MUM2 gene polymorphisms were determined for the accessions affected in mucilage release and five causal mutations , distinct from the original Shahdara mutation , were identified in six of the accessions , four of these introduced premature stop codons ( Figure 1A , Table S2 ) . Segregation for mucilage release or non-release in the Ale and Sku populations always correlated with the absence or presence of the causal mutation , respectively . For the Lom3-1 accession no causal mutation was identified in MUM2 , despite genetic non-complementation demonstrating it to be a mutant allele ( Table S2 ) and no mutation was identified in 208 base pairs ( bp ) upstream of the ATG or 192 bp downstream of TGA . Analysis of MUM2 expression , by quantitative RT-PCR ( qRT-PCR ) , confirmed that Lom3-1 was affected in MUM2 , as RNA steady-state levels were extremely low , similar to those of the mum2-11 knockout mutant ( Figure 1B ) . Interestingly , MUM2 transcript abundance was markedly reduced in all the accessions where mutation introduced a premature stop codon ( Figure 1B ) . As these stop codons should only directly affect translation , the observed reduction in transcript levels indicates the possible intervention of nonsense mediated mRNA mechanisms to degrade aberrant transcripts [31] . As expected , MUM2 expression levels were similar to those observed for wild-type Col-0 for the accessions Nfro1-1 , Dja-1 and Sk-1-1 ( Figure 1B ) . The latter two are not mum2 mutants and Nfro1-1 contains an amino-acid substitution in MUM2 . The natural mucilage release mutants are therefore the result of 9 independent mutation events , with 7 in the same gene MUM2 . The identification of MUM2 loss-of-function alleles in two specific geographic regions could result from local selection pressures . We examined patterns of differentiation between local populations within geographical regions , using previously described patterns of neutral variation [32] ( Protocol S1 ) . Patterns of population differentiation at either the MUM2 locus or for mucilage release did not deviate from neutral expectations ( Table S3 ) . Although the repeated evolution of independent loss-of-function mutations in central Asia and Norway remains intriguing , there is no evidence for local adaptation for mucilage retention . Forced mucilage release by breaking the outer cell wall of seed coat epidermal cells with acid and alkali has previously shown that mum2-11 seeds form a thinner layer of adherent mucilage; branched RG I does not expand to the same degree as unbranched RG I when hydrated ( Figure S1A and S1B ) [11] . In accessions containing a premature stop codon , adherent mucilage was observed as a thin layer , like that of mum2-11 seeds , coherent with reduced MUM2 transcript abundance ( Figure 1C and Figure S1C to S1H ) . The adherent mucilage was reduced less for Nfro1-1 and Lom-3-1 consistent with the production of normal amounts of a hypofunctional enzyme or a reduced amount of functional enzyme , respectively ( Figure 1C ) . The adherent mucilage of Dja-1 and Sk-1-1 accessions had a different appearance to that of mum2 mutants , in agreement with their causal mutations being in PMEI6 and PER36 , respectively ( Figure S1I and S1J ) . In order to examine in more detail the ecological significance of the absence of mucilage release in Arabidopsis accessions , physiological characteristics were examined in more detail . For these studies , insertion mutants in genes affecting mucilage production were used as the natural variants contain other polymorphisms that could influence the phenotypes examined . Three mucilage-release mutants were used , all in the reference accession Col-0; mum2-11 accumulates normal amounts of polysaccharides , whereas myb61 and myb5-1 have reduced polysaccharide accumulation [6] , [25] , [33] , [34] . No consistent differences in germination were observed for different seed lots of wild-type , mum2-11 or myb61 seeds grown on high PEG concentrations ( Figure S2; Protocol S2 ) . As seeds for the same genotype from three independent cultures showed different germination capacities at high osmotic potential , this phenotype would appear to vary depending on the environmental conditions of mother plant culture . The effect of mucilage retention on seed imbibition was examined in more detail for mum2-11 and myb5-1 . Water mobility during seed imbibition was analysed by low-field NMR spectroscopy . The proton spin-spin relaxation times ( T2 ) depend on their molecular environment . NMR relaxation time measurements carried out on plant tissues are sensitive to the water content and localization of water in cells [35] , [36] and relaxation signals are generally described by a multi-exponential behaviour reflecting different water compartments [37] , [38] . T2 relaxation times were firstly attributed to water associated with different seed compartments using wild-type seeds ( Table 2 ) . Water transfer rates between compartments were determined from the evolution of signal amplitudes . Although , wild-type seeds showed the expected rapid water uptake by mucilage polysaccharides over the first 90 minutes ( Figure 2A ) , water transfer to internal seed tissues was slower for wild-type seeds than in mucilage release mutants ( Figure 2B and Figure S3 ) . This indicates that during early imbibition while water is transferred into the mum2-11 and myb5-1 seeds to interact with macromolecules , in the wild type it remains trapped outside the seed in mucilage . Furthermore , transfer of water to internal seed tissues was higher in myb5-1 than mum2-11 , with amplitude already at high levels after the 5 minutes that had elapsed before the first measurement ( Figure 2B ) . The presence of more non-released mucilage polysaccharides in mum2-11 , therefore , reduced water transfer inside the seed compared to myb5-1 . A previous study has shown that myb5-1 seeds have higher oil contents than wild type [39] . Determination of seed oil content for dry seeds ( Table 2 , A ( 3 ) +A ( 4 ) ) confirmed this finding ( 46 . 1%±0 . 71 for myb5-1 , 40 . 34%±1 . 35 , wild type and 42 . 62%±0 . 73 , mum2-11 ) . Mutations of MUM2 , PMEI6 and PER36 may result in modifications in other tissues where these genes are transcribed . Although MUM2 is almost exclusively expressed in vascular tissue in vegetative tissues [25] , the naturally occurring mutations are unlikely to increase fitness through improved water management as no difference was observed for mum2 mutants compared to wild type on water deficit ( Figure S4; Protocol S3 ) . Dry seeds of Plantago coronopus , a myxospermous species that grows in desert highlands , can be dispersed by run-on rainwater before sinking and adhering to soil by released mucilage [40] . In a similar manner , it was noted that when water was added to seeds of the mum2-11 mutant or accessions that do not release mucilage , the majority of seeds floated at the surface , whereas most wild-type Col-0 or mucilage releasing seeds sank ( Figure 3A ) . Wild-type Col-0 seeds floated for a few seconds before mucilage was released and increased seed specific weight , whereas mutant seeds continued to float for many hours and even germinated at the water surface ( Figure 3B ) confirming that floating seeds were imbibed . The floating phenotype was also observed for seeds of myb5 , ttg1 , gl2 and mum4 mucilage release mutants where mucilage accumulation is reduced [22] , [24] , [33] , [34] , [41] . Buoyancy maintenance was not due to differences in mutant dry seed surface area or weight ( Figure S5; Protocol S4 ) . Since flotation was observed in central Asian and Scandinavian accessions , we screened for seed flotation in an extra set of 53 Arabidopsis accessions from central Asia ( including Russia ) and Scandinavia ( Table S4 ) . Seeds were examined for flotation by imbibition in water followed by ruthenium red staining; seeds of four accessions floated and these originated from the Altaï Republic of Russia , close to the border with Kazakhstan . Interestingly , staining indicated that mucilage had been released from these accessions , and that the adherent layer of mucilage was greatly reduced ( Figure 4C compared to 4A ) ; these accessions were termed Floating Mucilage Releasing ( FMR ) accessions . Cellulose in adherent mucilage is necessary for fixing mucilage pectin domains to the seed; seeds of cesa5 , fei2 and sos5 mutants are affected in cellulose production and have a reduced amount of adherent mucilage ( Figure 4B ) [10]–[12] . Imbibed seeds of FMR accessions showed little or no cellulose labelling within the thin layer of adherent mucilage , in contrast to the wild-type Col-0 accession where cellulose was visible as rays originating from the tops of columella as well as diffuse labelling between rays ( Figure 4D and 4G ) . Labelling of FMR seeds was even more reduced than that observed with cesa5 ( Figure 4E and 4H ) . Nevertheless , in the FMR accessions columella and radial cell walls were visible , as well as outer cell wall fragments ( Figure 4F and 4I ) . The reduction of adherent mucilage in the cesa5 mutant has been associated with a redistribution of RG I sugars to the water-soluble mucilage layer [12] . The amount of sugars in water-soluble mucilage was , therefore , determined for the different FMR accessions compared to wild-type Col-0 and cesa5-1 ( Figure 5 ) . All four FMR accessions had higher amounts of water-soluble mucilage sugars than wild type , like cesa5-1 , indicating that FMR mutants also show redistribution of mucilage to the water-soluble layer .
Natural variation between populations is a useful tool for the identification of mutations that produce physiological changes . Previously a visual screen for defective mucilage release from seeds of Arabidopsis accessions had identified Shahdara and Dja-1 as natural mutants affected in the MUM2 and PMEI6 genes , respectively [25] , [29] . Here we have identified a further eight accessions affected in mucilage release and shown that their seeds float on water ( Table 1; Table S1; Figure 3 ) . An additional four accessions were also identified whose seeds float , but release mucilage , termed FMR accessions ( Figure 4; Table S4 ) . Seven of the eight accessions affected in mucilage release were mum2 mutants resulting from six different mutations that were distinct from the original Shahdara mutation ( Figure 1A ) . In the accessions that are mum2 mutant alleles , MUM2 transcript abundance and the extent of adherent mucilage swelling were in accord with the mutations identified ( Figure 1B and 1C; Figure S1 ) . The Sk-1-1 accession was found to contain a non-sense mutation in PER36 . The identification of only one natural mutant in PMEI6 and PER36 genes compared to eight variants with 7 causal mutations for MUM2 may be due to the size of the MUM2 gene region , 5751 bp , which would increase the probability of mutation compared to the 816 bp and 1451 bp of PMEI6 and PER36 , respectively . FMR accessions had an extremely reduced layer of adherent mucilage ( Figure 4C ) , whereas soluble mucilage amounts were higher than those of the Col-0 accession ( Figure 5 ) . Mutants affected in the synthesis of cellulose present in adherent mucilage , cesa5 , fei2 and sos5 , have similar mucilage phenotypes . Nevertheless , some cellulose is still observed in the adherent mucilage of these mutants as rays , even in knockout mutants ( Figure 4E and 4H ) [10]–[12] , and these are not observed in the FMR accessions ( Figure 4F and 4I ) . Furthermore , seeds of cesa5-1 mutants do not float . This indicates that the mutation ( s ) in FMR accessions have a more profound effect on the synthesis of cellulose present in seed mucilage . Identification of the gene ( s ) affected should further our understanding of cellulose production and the interactions between cellulose and pectin . Seed flotation enables long-distance dispersion on water . A previous study has observed seed dispersion by flotation on run-on rainwater for Plantago coronopus , which inhabits desert highlands . This species also releases mucilage from seeds and it was observed that dry seeds floated for between 10 to 44 minutes before sinking [40] . Similarly mucilage retention in the seed coat , or its release as mainly water-soluble mucilage , allowed Arabidopsis seeds to float , and even germinate on the water surface ( Figure 3 ) . Mutations causing natural Arabidopsis variants to float could , therefore , be a local adaptation to improve seed dispersal . In effect , the collection sites of Neo-3 , Neo-6 and Sus-1 accessions were near to rivers . The external surfaces of seed coats are remarkably diverse and features such as hairs or wings are often used to assist dispersal . Genetic variation in Arabidopsis seed dispersal has previously been observed , but was a maternally inherited trait controlled by plant architecture and expressed in high-density populations [42] , [43] . This contrasts with the genetic variants identified here that might modify seed dispersal independently of maternal effects . Although the existence of genetic variation for a characteristic that could influence dispersal reveals the potential for evolution of this trait , the advantage of floating due to modified mucilage properties remains to be demonstrated; at the species level , the footprint of natural selection was not observed . Seed mucilage polysaccharides represent 3% of dry seed weight [12] and their production is a significant metabolic investment for the mother plant . Several mucilage release mutants have been identified that accumulate reduced amounts of mucilage and tests using myb5 , ttg1 , gl2 and mum4 mutants showed that their seeds also floated . Yet , none of the accessions affected in mucilage release were affected in loci that reduce the accumulation of mucilage polysaccharides ( Figure 1C ) [25] , [29] . This suggested that the non-released polysaccharides present in the epidermal cells of the seed coat serve a function . Low-field NMR analysis of water uptake by seeds showed that rates of transfer inside the seed and in interaction with macromolecules in seed tissues were higher in myb5-1 than mum2-11 ( Figure 2; Figure S3 ) . The presence of more mucilage polysaccharides in the epidermal cell layer , therefore , decreased the speed with which the tissues within the seed were imbibed and would maintain mum2-11 seed buoyancy , whereas the floating capacity of myb5 seeds would be reduced . The four FMR accessions present a novel mucilage phenotype with almost no adherent mucilage ( Figure 4 ) while the amounts of soluble mucilage released were higher than those observed in wild-type Col-0 ( Figure 5 ) . In these accessions , the mucilage produced could not maintain buoyancy by reducing water uptake by internal seed tissues . Mucilage is also produced by root cap cells and , like seed coat mucilage , it is at the interface between plant cells and the external environment . One of the roles proposed for root cap mucilage has been the structuration of bacterial communities present in the rhizosphere through its metabolism [44] . In the same manner , a recent study has implicated the metabolism of seed mucilage by soil microorganisms in the promotion of Artemisia sphaerocephala seedling growth [45] . The soluble mucilage from hydrated seeds of FMR accessions would be released into the soil near the mother plant and could aid in the modulation of soil flora as observed for root mucilage [reviewed in 45] . Released mucilage has been proposed to enhance seed hydration and to enable germination under conditions of reduced water potential [6] . Although rapid uptake of water was observed for adherent mucilage polysaccharides in wild-type Col-0 seeds ( Figure 2A ) , water was transferred at a slower rate to internal seed tissues compared to seeds of mucilage release mutants ( Figure 3B; Figure S3 ) . This demonstrates that although mucilage polysaccharides are an excellent hydrogel , they do not increase transfer of water into seeds . Accordingly , seed germination on high concentrations of PEG was not consistently different between seeds that released mucilage and those that did not ( Figure S2 ) . The more rapid imbibition of internal seed tissues in mucilage release mutants might enable them to germinate more rapidly , but differences in germination reported for certain Arabidopsis mucilage mutants have indicated that the absence of released mucilage causes a germination lag [6] , [18]–[20] . Alternatively , as the adherent mucilage traps water around the seed , this could slow the rate of seed drying . Studies on the desert shrub Artemisia sphaerocephala have shown that DNA repair was improved in achenes with mucilage humidified by desert dew , compared to achenes where mucilage had been removed [46] . Seed mucilage could , therefore , prolong the imbibed state , providing more time for repair mechanisms to function and thereby improve seed longevity . In conclusion , the identification of fourteen natural Arabidopsis mucilage mutants that float has highlighted the occurrence of at least ten independent mutation events , affecting four different loci , in populations in central Asia and Scandinavia . These mutations lead to modifications in seed mucilage production that maintain buoyancy compared to seeds that release a thick adherent layer of mucilage . Whether this is an adaptive trait requires further study , as among the accessions studied here the population genetic signature of local adaptation was not detected . Genetic variation in mucilage properties promises to contribute to our understanding of the ecological relevance of its production and the populations where mucilage release is not a fixed trait will be an important tool for testing potential roles .
Accessions were obtained from the Versailles Arabidopsis Stock Centre ( http://publiclines . versailles . inra . fr/ ) except for the Norwegian and Swedish populations in Table 1 and Table S1 , which were kindly donated by Odd-Arne Rognli through NARC ( Norway ) . Seeds for new central Asia accessions , including Dja , Sus and Neo , were collected from their natural habitat ( Table S1 ) http://www . inra . fr/vast/collections . htm ) . Genetic differentiation of MUM2 was examined in 289 lines corresponding to 15 Norwegian , 6 central Asian , 7 Spanish and 13 French populations [32] . The mum2-10 , mum2-11 and cesa5-1 mutants ( SALK_011436 , SALK_110461 and SALK_125535 , respectively; Col-0 background ) were obtained in previous studies [25] , [47] . The ttg1 ( GK_286A06 ) , mum4 ( SALK_085051 ) , myb61 ( SALK_106556 ) and gl2 ( SM_3_16350 ) mutants , all Col-0 background , were obtained from the Nottingham Arabidopsis Stock Centre ( [48]–[50]; http://arabidopsis . info ) . The myb5-1 ( SALK_030942 ) and aba3-1 mutants [51] were a gift of C . Dubos and M . Koornneef , respectively . Crosses were performed between mum2-12 ( Shahdara accession ) and Dja-1 , Sus-1 , Neo-3 , Neo-6 , Ale-8 , Had-3-1 , Lom-3-1 , Nfro-1-1 , Sk-1-1 and Sku-20 , and between Dja-1 and Sk-1-1 accessions . As the seed coat is maternally derived , seed coats of progeny from F2 and F3 generations were examined; only seeds from crosses with Dja-1 and Sk-1-1 showed genetic complementation . The gene affecting mucilage release in Sk-1-1 was localised to an interval on the lower arm of chromosome 3 using a mapping population of 176 F2 individuals derived from a cross between Col-0 and Sk-1-1 accessions . DNA was extracted from 33 F2 progeny that did not release mucilage , using 10 seedlings from each . Whole-genome resequencing was carried out on this DNA by Genome Enterprise Ltd ( http://www . norwichresearchpark . com/parkdirectory/genomeenterpriselimited . aspx ) using an Illumina HiSeq 2000 sequencer . The SHOREmap software package with bwa aligner ( http://1001genomes . org/downloads/shore . html ) was used to map the Sk-1-1 polymorphisms against those of the Col-0 accession and identify mutations . The output of this pipeline was filtered for mutations in the mapping interval on the chromosome 3 . These polymorphisms were then filtered against polymorphisms present in the genomes of 17 accessions available from the 1001 genome project ( http://1001genomes . org/index . html ) and known to release mucilage; Alc-0 , Altai-5; Baz-0 , Blh-1 , C24 , Dja-1 , Gol-1 , Jea , Ler-0 , Neo-6 , Oy-0 , Qar-8a , Ri-0 , Sakata , Sus-1 , Ws-2 and Zal-1 . These sequence data were produced by the Weigel laboratory at the Max Planck Institute for Developmental Biology , the Ecker laboratory at the Salk Institute [52] , the Pennacchio laboratory at the DOE Joint Genome Institute , and the Mott laboratory at the Wellcome Trust Center for Human Genetics [53] . Of the 258 SNPs that were Sk-1-1 specific only 4 resulted in the introduction of a stop codon; one of these was in PER36 , which has been shown to be required for mucilage release [30] . All phenotypic analyses were carried out using seed lots or tissue obtained from plants cultured simultaneously . Phenotyping for mucilage release was carried out as described previously [25] or by imbibing seeds on filter paper ( 6 cm dia . ) hydrated with 700 µL of water for 1 h followed by staining with 200 µg mL−1 ruthenium red . Wild-type Col-0 , mum2-11 and accessions defective for release were treated with 0 . 05 N HCl then 0 . 3 M NaOH to release mucilage and stained with ruthenium red as described previously [25] . Seeds for 53 accessions ( Table S4 ) were obtained from 2 independent series of plants cultured in compost ( Tref Substrates , http://www . jiffygroup . com/ ) in a growth chamber ( 21°C day , 17°C night , 150 µmol m−2 s−1 light intensity , 16 h photoperiod , 65% relative humidity ) . A seed lot from each culture was examined for seed flotation by imbibition of seeds for 10 min in water , followed by staining in 500 µg mL−1 ruthenium red and observation with a light microscope ( Axioplan 2; Zeiss; http://www . zeiss . fr/ ) . DNA for PCR analysis was extracted from flower buds as described by Doyle and Doyle [54] . The full sequence of MUM2 in a subsample of 28 accessions was retrieved from public databases [55] . Otherwise , sequencing of the MUM2 gene was performed as described by Sullivan et al . [12] . The Arabidopsis lyrata sequence of MUM2 was obtained from the public database and used as outgroup ( http://www . phytozome . net/alyrata ) [56] . Genotyping with the MUM2Sha marker was carried out by PCR amplification of either a 162 bp ( wild-type ) or 118 bp ( mum2-12 ) DNA fragment with forward primer 5′-TGGTCGTTATTGGGTCTCGT-3′ and reverse primer 5′-TTAAGAACGCCCGAGGAATA-3′ . Two fragments of MUM2 were re-sequenced for 41 Arabidopsis accessions collected worldwide and for the 110 individuals collected in Norway and central Asia; the 5′ portion of ∼1000 bp was amplified using forward primer 5′-GAAGGAGGCATCGATGTGAT-3′ and reverse primer 5′-GGTGAGTTTGGTCCAGGAAA-3′ and the 3′ portion of ∼600 bp was amplified using forward primer 5′-CTGGAGCTTACATGGAGAGGA-3′ and reverse primer 5′-CAAGAGGATCACCTTCC-3′ . Developing siliques at 8 to 12 days after pollination were pooled and total RNA extracted and reverse transcribed as described previously [29] . Quantitative real-time PCR reactions were performed as described by Plessis et al . [57] . MUM2 specific primers were as follows , forward primer 5′-CAGCGGCATGGTTGGTCT-3′ and reverse primer 5′-CCAAGCAAACCCACCGAGT-3′ , and had been tested for their efficiency rates and sensitivity on a dilution series of cDNAs . EF1α4 primers have been described previously [57] . Water-soluble mucilage extracts were obtained from 200 mg of intact seeds as described previously [12] The uronic acid ( as GalA ) and total neutral sugar ( as Rha ) contents were determined by the automated m-hydroxybiphenyl and orcinol methods , respectively [58] , [59] . 1H NMR measurements were performed using a Time-Domain spectrometer ( Minispec BRUKER; http://www . brukeroptics . com/ ) operating at a resonance frequency of 20 MHz . The NMR system was equipped with a temperature control device connected to a calibrated optic fibre ( Neoptix Inc . ; http://www . neoptix . com/ ) allowing for ±0 . 1°C temperature regulation . For the assignment of T2 relaxation times to different water compartments , measurements were firstly obtained for three different wild-type Col-0 samples; intact imbibed seeds , intact dry seeds and imbibed seeds pre-treated to remove soluble mucilage , but with adherent mucilage; water-soluble mucilage was extracted from wild-type Col-0 seeds as described previously [12] and after lyophilisation , seeds without soluble mucilage were resuspended in water and analysed by NMR . The mature , “dry” Arabidopsis seeds ( approximately 8% water content ) were introduced into NMR tubes ( 10 mm dia . ) . In order to fill tubes to a 10 mm height , corresponding to the homogeneous region of the NMR radiofrequency , 200 to 220 mg of Arabidopsis seeds were used and imbibed with 150 to 220 µl of deionised water , tubes were then weighed and hermetically sealed . Acquisitions of T2 were carried out from 5 min to 24 h of imbibition and samples were analysed at ambient temperature ( 20°C ) . Two types of pulse sequences were used; proton free induction decays ( FID ) were acquired using the following parameters: a 90° pulse of 2 . 6 µs , a dwell time of 0 . 4 µs between two successive data points , 16 scans of 150 data points , and a recycle delay of 9 s between each scan , and the Carr–Purcell–Meiboom–Gill ( CPMG ) pulse sequence with a time echo of 0 . 2 ms . Sixteen scans were acquired for each genotype with 8000 or 16000 data points [60] . Transverse relaxation data were analysed with the following model:where T2i are the relaxation times of the mobile populations and Ai is the intensity of the mobile populations [60] . To ensure the accuracy of the data treatment , spin–spin relaxation decay curves were fitted using MEM [61] and a discrete method [62] . In imbibed intact seeds five T2 components were identified which could be assigned to protons of water populations with different mobility and ratio ( expressed in relative percentage ) ( Table 2 ) . The first and second components , T2 ( 1 ) and T2 ( 2 ) , were also observed in dry seeds and could be assigned , respectively , to the solid phase of the seed , notably macromolecules such as polysaccharides and proteins , and protons in exchange with the hydroxyl groups found in these macromolecules . The latter two components ( T2 ( 3 ) and T2 ( 4 ) ) were assigned to oil with a superposed contribution from water in both as intracellular water in the former and in adherent mucilage in the latter . Comparison of T2 ( 3 ) and T2 ( 4 ) values for dry seeds with those obtained in imbibed seeds with water-soluble and adherent mucilage or only adherent mucilage confirmed the latter attributions ( Table 2 ) . A fifth component , T2 ( 5 ) , was identified in imbibed wild-type seeds that corresponded to water in soluble mucilage . This component was also present in imbibed mutant seeds where it was attributed to water outside seeds ( Table 2 ) . In order to verify the assignment given to the multi-exponential NMR signals for water in seed compartments , at the end of NMR measurements , the water content of samples was estimated by weighing before and after drying in an oven at 103°C for 24 h . The amplitudes of the NMR signals from the seeds before and after imbibition were compared to the amplitude of the signal expected from distilled water . The signals expected from water were calculated for each sample and each compartment as the product of its mass , its water content and the signal of the distilled water per unit mass . The water content of each compartment was then calculated by deducing the signal contribution from oil . Analysis of the evolution of signal amplitudes for the different components allowed the water transfer rates between different compartments to be determined . The total water content was calculated using the T2 amplitude and was coherent with results using a gravimetrical method before and after drying seeds . Calculation of water content for each compartment confirmed that even after 24 h of imbibition , approximately 40% of water associated with wild-type seeds was trapped outside in mucilage ( Table 2 ) . Seeds were imbibed in water for 3 h then rinsed twice prior to staining . The released adherent mucilage was stained with ruthenium red as previously described [25] or Pontamine Fast Scarlet S4B in 150 mM NaCl [63] . Observations were carried out with a light microscope for ruthenium red ( Axioplan 2; Zeiss ) or a Zeiss LSM710 confocal microscope using a 561 nm diode laser line to excite Pontamine and detecting fluorescence emission between 570 and 650 nm . For comparison of signal intensity within a given experiment laser gain values were fixed .
|
Seeds of the model plant Arabidopsis release sticky mucilage on imbibition that is constituted of complex polysaccharides . In this study , we have identified and characterised natural Arabidopsis variants that do not release mucilage and found that their seeds float . The accumulation of unreleased polysaccharides in the seed coat reduced water uptake rates on imbibition and would maintain buoyancy . We subsequently identified additional floating natural variants where mucilage is released , but is not attached to the seed , apparently due to defective cellulose production . The different variants arise from at least ten independent unique mutations and were collected from two discrete geographical areas . Arabidopsis seed flotation has thus evolved several times due to modifications in mucilage release . Released mucilage was found to retain water , but did not improve imbibition of internal seed tissues , indicating a role in maintaining seeds hydrated . These findings highlight the physical and potential physiological effects of mucilage production by the seed coat .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"plant",
"science",
"plant",
"evolution",
"plant",
"genetics",
"biology",
"plant",
"physiology"
] |
2014
|
Local Evolution of Seed Flotation in Arabidopsis
|
Due to the omnipresent risk of epidemics , insect societies have evolved sophisticated disease defences at the individual and colony level . An intriguing yet little understood phenomenon is that social contact to pathogen-exposed individuals reduces susceptibility of previously naive nestmates to this pathogen . We tested whether such social immunisation in Lasius ants against the entomopathogenic fungus Metarhizium anisopliae is based on active upregulation of the immune system of nestmates following contact to an infectious individual or passive protection via transfer of immune effectors among group members—that is , active versus passive immunisation . We found no evidence for involvement of passive immunisation via transfer of antimicrobials among colony members . Instead , intensive allogrooming behaviour between naive and pathogen-exposed ants before fungal conidia firmly attached to their cuticle suggested passage of the pathogen from the exposed individuals to their nestmates . By tracing fluorescence-labelled conidia we indeed detected frequent pathogen transfer to the nestmates , where they caused low-level infections as revealed by growth of small numbers of fungal colony forming units from their dissected body content . These infections rarely led to death , but instead promoted an enhanced ability to inhibit fungal growth and an active upregulation of immune genes involved in antifungal defences ( defensin and prophenoloxidase , PPO ) . Contrarily , there was no upregulation of the gene cathepsin L , which is associated with antibacterial and antiviral defences , and we found no increased antibacterial activity of nestmates of fungus-exposed ants . This indicates that social immunisation after fungal exposure is specific , similar to recent findings for individual-level immune priming in invertebrates . Epidemiological modeling further suggests that active social immunisation is adaptive , as it leads to faster elimination of the disease and lower death rates than passive immunisation . Interestingly , humans have also utilised the protective effect of low-level infections to fight smallpox by intentional transfer of low pathogen doses ( “variolation” or “inoculation” ) .
The first encounter of a host with a particular pathogen often leads to the outbreak of the disease , yet a secondary exposure rarely causes illness , due to the immunological memory of the host . Whereas immune memory in vertebrates is well appreciated [1] , the phenomenon of an individual developing specific immunity against a subsequent pathogen exposure—referred to as immune priming—has only recently been described in invertebrates , both within the lifetime of an individual [2]–[8] and in transgenerational protection of offspring ( [8]–[12] , but see [13] ) . In contrast to vertebrates , the underlying mechanisms are not yet understood in invertebrates [14] , [15] . In addition to this immunological memory at the level of individuals , a similar phenomenon occurs at the colony level in insect societies [16]–[18] . Society members act collectively , similar to cells in a body , and work as a superorganism [19] , [20] in multiple aspects , including anti-pathogen defence [21] . For instance , an initial pathogen contact of a colony due to the presence of exposed individuals has been shown to lower the susceptibility of their nestmates to infection when they are later exposed to the same pathogen [16]–[18] . In addition to this physiological “social immunisation , ” the collectively performed hygiene behaviour that complements individual defences in social insects [22]–[24] is also affected . Allogrooming of exposed individuals by their nestmates occurs more frequently in colonies with previous experience with this pathogen than in naive colonies [25] , [26] . In contrast to individual immune priming , social immunisation thus refers to a protection of naive individuals of a colony after social contact to exposed individuals . The phenomenon of social immunisation occurs broadly in insect societies—in unrelated social host species ( ants and termites ) and against divergent pathogen taxa ( fungi [17] , [18] and bacteria [16] ) —yet the mechanisms underlying this effect are largely elusive ( but see [16] ) and have only been hypothesised upon for fungal pathogens [3] , [17] , [18] , [27] . In this study , we therefore aimed to determine the underlying causes of social immunisation in colonies of the ant Lasius neglectus after exposure of single individuals to the entomopathogenic fungus Metarhizium anisopliae , a common natural pathogen of ants [28] , [29] . In this system , we have previously described that 5 d of social contact to an individual exposed to fungal conidia ( conidiospores; [30] ) led to a lower susceptibility of nestmate ants when challenged with a high fungal dose after this period [18] . It remained open , however , which social interactions may trigger this effect and how they elicit changes in nestmate immunity . The observed protection in nestmates of exposed ants may be caused by the active upregulation of their own immune systems following social contact to the fungus-exposed individual . Alternatively , social transfer of immune mediators produced by colony members may lead to passive protection of nestmates without requiring the activation of their own immune systems ( as outlined by [3] , [17] , [27] ) . The active and passive route to social immunisation may also act in concert . Active upregulation of the nestmates' immune system may be caused by perception of a trigger signal elicited from the exposed individual , possibly of behavioural or chemical nature . In humans , mere visual perception of sick individuals was recently shown to cause preventive stimulation of the immune system [31] . Similarly , in plants , herbivory defence was promoted by perception of volatile chemical cues elicited by an attacked neighbouring plant [32] . Active stimulation of the immune system can also be caused by low-level infections [3] , [8] , [33] , [34] , which may result from social transfer of the pathogen from the exposed individual to its nestmates ( as suggested by [3] ) , occurring during “normal” social interactions , or as a byproduct of collective sanitary behaviour such as allogrooming of the exposed individual by its nestmates [22] , [35] . Passive immunisation may result from a social exchange of antimicrobials produced by the exposed individuals and transferred to their nestmates . Possible transfer pathways include the “external route” over the body surface or the “internal route” by exchange of body fluids [16] . The external body surface ( cuticle ) of ants is covered with antimicrobial substances produced in an ant-specific gland ( metapleural gland [36] , [37] ) and nestmates could easily pick up these substances and apply them on their own bodies by allo- and self-grooming . Immune effectors produced inside the body of infected individuals may be exchanged during the common social feeding behaviour of regurgitation and feeding of trophallactic droplets [16] , [38] , as has recently been suggested as a mechanism for social immunisation of ant colonies after bacterial exposure [16] . Whereas bacterial infections are typically orally transmitted [39] , entomopathogenic fungi are externally transmitted , making distinct disease dynamics of these pathogen taxa likely . In this study , we applied a multi-level approach to determine the functional mechanism of social immunisation of ant colonies against a fungal pathogen . We analysed the behavioural interaction rates between group members and determined whether social contact may lead to exchange of the pathogen or immune effectors , or whether social immunisation may be triggered by social signals . We determined both the physiological immunity of fungus-exposed individuals and their nestmates , as well as their immune gene expression . Lastly , we developed an epidemiological model to explore long-term colony-level effects of social immunisation depending on the underlying mechanisms .
We have previously shown that social contact to a Lasius worker exposed to conidia ( dispersal form , conidiospores; [30] ) of the entomopathogenic fungus M . anisopliae , but not to control-treated ants , increased the survival of previously naive nestmates when challenged with the same M . anisopliae strain 5 d later [18] . We now directly assessed the immune function of nestmates with a novel and sensitive “antifungal activity assay . ” We incubated ant tissue with blastospores ( within-host infection form; [30] ) of the fungus to measure the ability of ants to inhibit fungal growth . We found a significantly higher antifungal activity in nestmates of fungus-exposed as compared to nestmates of control-treated individuals ( Figure 1 ) . This was true not only after 5 d of social contact to an exposed individual , but already after 3 d ( GLM , F = 3 . 859 , df = 3 , p = 0 . 017; treatment type [fungus treatment versus sham control]: F = 10 . 634 , df = 1 , p = 0 . 002; time [3 versus 5 d post-treatment]: F = 0 . 001 , df = 1 , p = 0 . 973; interaction [Treatment Type×Time]: F = 0 . 942 , df = 1 , p = 0 . 338 ) . To understand the mechanism behind increased antifungal defence in nestmates of exposed ants , it is important to study the behaviour of group members . First , behavioural changes of individuals after fungal exposure may be a signal to their nestmates to upregulate their immune system . Second , the social interactions define the routes and opportunities for potential exchange of immune effectors [40] , [41] or the pathogen itself [42] . Compared to control-treated ants , which did not elicit social immunisation in their nestmates , fungus-exposed ants did not show significantly changed rates of either brood care behaviour [18] or self-grooming activity ( LVU , unpublished data ) . Similarly , other studies found that pathogen exposure had no effect on self-grooming [26] or only when doses present in the colony were very high [25] . This makes it unlikely that nestmates may have perceived a trigger signal by social interaction or potential observation of the individual behaviour of exposed ants . To obtain information on possible pathways for transfer of the pathogen or immune mediators , we analysed the social interactions between colony members in more detail . As in our original experimental setup we grouped five naive nestmates with a single treated Lasius worker that had either received infectious M . anisopliae conidia ( fungus treatment ) or the same treatment without the pathogen ( sham control ) . We observed three types of social interactions between group members . Antennation behaviour—that is , nestmate recognition behaviour by antennal contact [43]—occurred extremely rarely ( 6 . 6% of all interactions ) . Moreover , rates did not differ between treated and nestmate ants or among nestmates , for both fungus treatment and sham control ( Generalised Linear Model [GLM] with negative binomial errors , LR χ2 = 1 . 969 , df = 3 , p = 0 . 579; data not shown ) . All other social interactions observed between group members consisted of ( a ) allogrooming ( i . e . , cleaning the body surface of another ant ) and ( b ) trophallaxis behaviour ( i . e . , exchange of regurgitated liquid food droplets ) [43] . Both may be important pathways for social immunisation [3] , [16] , [17] , [27] . It is well known that nestmates actively contact exposed individuals and remove infectious material with their mouth by allogrooming , which is a very efficient social sanitary behaviour [43] , [44] increasing survival of pathogen-exposed individuals , but typically not compromising the survival of the nestmates [25] , [35] , [45] , [46] . Still , the grooming ant may contract the pathogen if it is not able to kill all infectious material in its mouth ( infrabuccal pockets; [47] , [48] ) or gut [49] , or if it unintentionally rubs off conidia with other body parts than the mouth during this intimate social interaction . In addition , allogrooming may lead to uptake of antimicrobial substances from the body surface of an exposed individual similar to exchanges of cuticular waxes important for nestmate recognition [50] . In our experiment , allogrooming rates between treated individuals and their nestmates were higher than among nestmates , but independent of the treatment type ( fungus versus sham control; Figure 2A; GLM with negative binomial errors , LR χ2 = 15 . 134 , df = 3 , p = 0 . 002; ant pairing [treated-nestmate versus nestmate-nestmate]: Wald χ2 = 14 . 501 , df = 1 , p<0 . 001; treatment type [fungus versus sham control]: Wald χ2 = 0 . 006 , df = 1 , p = 0 . 939 ) . Upregulation of grooming frequency not only against individuals treated with infectious material but also with sham control solutions is known from previous studies [29] , [51] and indicates that ants are very sensitive to applications on the bodies of their group members . Despite the lack of difference between the two treatment types , intensive grooming towards treated individuals provides a potential route for transfer of either the pathogen itself or external immune effectors . One important factor is the timing of allogrooming expression during the infection course of M . anisopliae . Entomopathogenic fungi like M . anisopliae infect their hosts by external adhesion onto and active penetration of the cuticle [52] . After contact to the insect cuticle , the conidia first adhere loosely to the body surface within several hours and then germinate and form a penetration plug to actively enter the host body within approximately 24 to 48 h [46] , [53] . Infection of the host and onset of an active immune response therefore occurs with a time delay of 2 to 3 d after exposure [54] , [55] . Allogrooming in the first 1 to 2 d would therefore allow for pathogen transfer , whereas after this time exposed ants lose their infectiousness [26] . Intensified allogrooming 3 or 4 d after exposure would instead indicate exchange of external antimicrobial substances . We analysed the time course of allogrooming frequency between treated individuals and their nestmates and found no change over time in the control treatment ( GLM with repeated measures , time: F = 0 . 973 , dfHuynh-Feldt = 3 . 648 , p = 0 . 416 ) . Allogrooming between nestmates and fungus-exposed individuals , however , was significantly higher in the first 2 d compared to later phases of the experiment ( Figure 2B; time: F = 4 . 006 , dfHuynh-Feldt = 3 . 306 , p = 0 . 006 [day1 versus day2: p = 0 . 178; day1 versus day3: p = 0 . 041; day1 versus day4: p = 0 . 001; day1 versus day5: p = 0 . 014] ) . Based on these data we suggest that if a transfer between group members occurs via allogrooming , it more likely involves a transfer of conidia , detachable early after exposure , than immune effectors , which can only be upregulated and transferred to the cuticle after infection of the individual 24–48 h after exposure . Social feeding via regurgitation and transfer of a trophallactic droplet may promote transfer of internal antimicrobial substances [16] . However , we found no differences in the rates of trophallaxis among all four groups , that is , neither between treated ants and their nestmates nor among the nestmates in either the fungus treatment or the control group ( Figure 2C; GLM with negative binomial errors , LR χ2 = 2 . 555 , df = 3 , p = 0 . 465 ) . Our data show that fungal exposure does not alter trophallaxis rates between exposed individuals and their nestmates , making passive immunisation by transfer of internally produced antimicrobial substances rather unlikely in our model system . Our findings after fungal exposure contrast with observations that trophallaxis rates between individuals injected with dead bacteria or bacterial cell wall components ( but also wounding controls ) were increased compared to trophallaxis rates among untreated individuals ( [16] , [56] , but see [57] ) . Taken together , our behavioural observations strongly suggest exchange of the fungal pathogen between the fungus-exposed ant and its nestmates during intensified , early grooming as the most likely mechanism for the observed anti-fungal protection in the nestmates . We therefore determined if fungal conidia indeed were transferred from the exposed individual to its untreated nestmates by direct tracing of fluorescently labelled conidia . We applied conidia of M . anisopliae labelled with red fluorescent protein ( RFP ) onto the exposed ant and determined their presence or absence on the cuticle of all group members after 2 d of social contact . We expected maximum pathogen transfer to have occurred at this time as ( a ) grooming activity between exposed ants and their nestmates is most intense in the first 30 h ( Figure 2B ) and ( b ) conidia are no longer transferable after this time [26] , [53] . As expected we found high amounts of conidia on all directly exposed individuals ( 15/15 ) and furthermore detected low numbers of conidia on the cuticles of 37% ( 17/45 ) of nestmates ( Figure S1; for negative controls see Materials and Methods ) . Interestingly , not only the quantity but also the location of conidia differed: whereas directly exposed individuals carried them mostly in areas likely difficult to reach by grooming such as joints and the antennal grooves , conidia on nestmates were rather attached to antennae and legs ( Figure S1 ) , suggesting that nestmates pick up the pathogen from the fungus-exposed individual during grooming . We can thus confirm pathogen transfer to the nestmates . In a next step we determined if the transferred conidia successfully established an infection in the nestmates . To quantitatively determine the infection load of directly fungus-exposed individuals and their nestmates over the course of the experiment , we sterilised their body surface to destroy all remaining conidia , dissected the ants , and plated their body contents on agar plates to count emerging fungal colony forming units ( CFUs ) . We used morphological determination , as well as PCR [58] , to confirm that outgrowing CFUs were indeed M . anisopliae , which was the case for all CFUs ( see Figure S2 as an example ) . None of the 30 negative controls ( see Materials and Methods ) and none of the individuals measured within 24 h after exposure ( 0/10 fungus-treated , 0/14 nestmates; Figure S3 ) showed fungal growth , confirming that we effectively sterilised the ants and measured only live fungus from inside the body . Three as well as five days after exposure , CFUs grew from the body content of nearly all directly exposed ants ( 80% [8/10] and 90% [9/10] ) and a similarly high number of nestmates ( 64% and 64% [each 9/14]; Figures 3 , S3; Fisher's exact test; day 3 , p = 0 . 653; day 5 , p = 0 . 341 ) . These data show that fungal infections in nestmates were more common than estimated from external pathogen transfer using labelled conidia . This may either indicate that we did not detect all conidia or that an additional infection route via the infrabuccal pocket in the mouth or the gut system occurred , for instance if groomed-off conidia were not completely prevented from germinating [47]–[49] . Fungal infection load in nestmates revealed that their infections were “low-level infections . ” The number of CFUs growing out of their bodies when infected was significantly lower than those growing from directly exposed ants at both day 3 ( Figures 3A , S3; Mann-Whitney U-test: n1 = 8 , n2 = 9 , U = 4 . 0 , p = 0 . 002 ) and day 5 ( Figures 3B , S3: n1 = 9 , n2 = 9 , U = 0 . 0 , p<0 . 001 ) . On average , the infection load of infected nestmates was 8 ( 4 . 4 versus 36 . 0 ) and 12 ( 8 . 1 versus 102 . 4 ) times smaller than that of directly exposed individuals on days 3 or 5 , respectively . Even if low-level infections occurred in the majority of nestmates , only 2% ( 3/150 ) died from a M . anisopliae infection after 5 d of social contact with the exposed individuals ( who showed death rates of approximately 50% due to application of an LD50 ) . This confirms that the effects of M . anisopliae infections are highly dosage dependent ( [35] and MKo and STr , unpublished data ) . To determine if the observed increase in antifungal activity of nestmates was a direct cause of these low-level infections , we established low-level infections in individuals in the absence of social interactions . To this end , we exposed isolated ants with a conidia dose that led to the same death rate ( LD2 ) and infection level as observed in the socially exposed nestmates . We found that low-dose , directly exposed ants had a significantly increased antifungal activity 3 d after exposure compared to control-treated ants ( Figure 4 ) . Interestingly , directly exposed individuals with a high dose ( LD50; as used for exposure of the single ants in our experiment above ) showed a significantly decreased capacity to inhibit fungal growth ( Figure 4; ANOVA: F = 10 . 361 , df = 2 , p<0 . 001; post hoc Protected Fisher's LSD tests all pairwise: sham control versus LD2: p = 0 . 046 , sham control versus LD50: p = 0 . 021; LD2 versus LD50: p<0 . 001 ) . This immune-suppressive effect of a high-dose infection is likely caused by the immune-interference and toxicity of M . anisopliae or by the fact that the immune responses had been depleted [41] , [59]–[61] . Immune stimulation of low-level infections has previously been described for both vertebrates and invertebrates [3] , [8] , [33] , [34] , and its protective effect yielded clinical application in humans [62] , [63] and poultry health management [64] . We have established that low-level infections , caused by social contact or direct low-dose exposure , lead to increased antifungal activity . Yet this does not exclude that nestmates with social contact to an exposed individual may also obtain signals that could actively trigger their antifungal immunity ( similar to [31] , [32] ) . To test this , we performed a “spatial-separation experiment” in which body contact and pathogen transfer to the exposed individual were prevented , whereas exchange of visual signals or volatile chemicals was still possible . The antifungal activity of nestmates of fungus-exposed individuals did not differ from that of nestmates of control-treated ants after 3 d of this constrained contact ( t test: t = −0 . 376 , df = 18 , p = 0 . 711 ) . These data suggest that a visual or volatile signal alone—at least one that acts over distance—is not sufficient to promote antifungal activity in the nestmates . Non-volatile chemical signals , such as cuticular hydrocarbons [65] that are part of the ants' cuticle , may in theory still play an additional role . However , their perception would always require body contact , which promotes pathogen transfer at the same time . We conclude that low-level infections alone provide a sufficient explanation for an active social immunisation of nestmates . We then tested if it may be complemented by a passive transfer of antimicrobial substances among nestmates . We performed a “temporal-separation experiment” and allowed the exposed ant to interact with its nestmates for 48 h . In this period , the pathogen ( a ) lost its ability to be transferred ( for confirmation see Materials and Methods ) and ( b ) established an infection in the ants , likely triggering an immune response [53]–[55] . After this time , we separated the treated individual and its “early nestmates” and added five “new nestmates” to both ( see Figure 5A , B ) . Three days later , we measured the antifungal activity of the new nestmates . We found no difference between new nestmates of control-treated versus fungus-exposed ants ( Figure 5A; t test: t = −0 . 159 , df = 18 , p = 0 . 876 ) or between new nestmates of early nestmates to a control-treated versus exposed individual ( Figure 5B; t test: t = −1 . 273 , df = 18 , p = 0 . 219 ) . This reveals that nestmates do not show an increase in antifungal activity if pathogen transfer is excluded . Passive transfer of antimicrobials among the group members thus seems very unlikely as an explanation for social immunisation . However , such transferable substances might be upregulated in infected individuals and simply failed to elicit immunisation of nestmates in our experiment . We therefore also analysed both the fungus-exposed ant and its nestmates directly for the presence of potentially transferable antimicrobials 3 d after treatment . Although allogrooming rates among nestmates were low in both sham control and fungus-treated groups ( Figure 2A ) , and trophallaxis rates were completely independent of treatment ( Figure 2C ) , infected nestmates may be important in transferring antimicrobial substances , as their antifungal activity is higher than that of directly exposed ants , which suffer a much higher infection level ( Figure 4 ) . We tested whether transferable substances of fungus-exposed individuals or their nestmates had higher antifungal activity than those of control-treated individuals and their respective nestmates . For externally transferable substances via allogrooming , we measured the antifungal activity of ( a ) the cuticle and ( b ) the thorax containing the metapleural gland content , which is known to have antimicrobial function and to be secreted onto the cuticle [36] . We also measured the antifungal activity of ( c ) the trophallactic droplet that is produced in the ant's body and is transferred via social feeding . We found that neither the cuticles nor the thoraxes containing the metapleural gland nor the trophallactic droplets of fungus-exposed individuals showed a different antifungal activity than the respective body parts of control-treated individuals ( Figure 5C; t tests; cuticle: t = 1 . 064 , df = 10 , p = 0 . 312; thorax: t = 0 . 224 , df = 10 , p = 0 . 828; trophallactic droplets: t = −0 . 594 , df = 18 , p = 0 . 560 ) . The same was true for the nestmates ( Figure 5D; t tests; cuticle: t = 0 . 107 , df = 18 , p = 0 . 916; thorax: t = 0 . 894 , df = 18 , p = 0 . 383; trophallactic droplets: t = −0 . 717 , df = 18 , p = 0 . 482 ) . This result was not an artifact caused by a potential effect of the control treatment , as the antifungal activity in these individuals was not different from completely untreated ants ( Materials and Methods ) . Taken together , we found no evidence for ( a ) a potential protective effect of nestmates in the absence of pathogen transfer and ( b ) potential upregulation of socially transferable antimicrobials in exposed colonies . This contrasts observations that trophallactic droplets obtained from bacteria-exposed ants had higher antibacterial activity than that of controls [16] , making passive immunisation a likely mechanism involved in social immunisation of ant colonies after bacterial exposure [16] , but not after fungal exposure . Instead , we documented that social interaction , most likely allogrooming , leads to pathogen transfer and sublethal low-level infections in the majority of nestmates of fungus-exposed individuals and that low-level infections are necessary and sufficient to induce an increased antifungal activity . To directly assess the effect of low-level infections on the immune response , we measured immune gene expression in nestmates using quantitative real-time PCR . We chose three immune genes known to be involved in the humoral and cellular defences of ants: ( 1 ) the antimicrobial peptide ( AMP ) defensin [66] , [67] , a soluble mediator that most closely resembles termicin , an antifungal peptide in termites [68] , [69]; ( 2 ) prophenoloxidase ( PPO ) , a key mediator of immune function in ants [70] , [71] that is essential for the process of melanization upon infection by a variety of pathogens , including entomopathogenic fungi [72] , [73]; and ( 3 ) cathepsin L , a lysosomal protease expressed in hemocytes [74] , which has both antibacterial [75] and antiviral activity [76] , but has not been implicated in antifungal responses . In Camponotus pennsylvanicus , another cathepsin ( cathepsin D ) was found to occur in higher amounts in the trophallactic droplets of ants after injection of heat-killed bacteria or LPS [16] , suggesting the involvement of cathepsins in antibacterial responses in ants . We confirmed that our host ant , L . neglectus , also responds to bacterial infection with cathepsin upregulation . Septic injury with Bacillus thuringiensis led to upregulation of cathepsin L gene expression , but not PPO , or defensin expression , compared to pricked controls ( Figure S4; defensin: t test; t = 0 . 186 , df = 4 , p = 0 . 862; PPO: t test; t = −1 . 448 , df = 4 , p = 0 . 221; cathepsin L: t test; t = −3 . 695 , df = 4 , p = 0 . 021; gene expression standardised to the housekeeping gene 18s rRNA ) . The choice of these three immune genes in this study therefore allowed us to examine the specific effects of social immunisation against the fungus M . anisopliae on immune pathways involved in insect defences . We compared mRNA levels of the three genes in nestmates of fungus-exposed individuals versus nestmates of control-treated individuals on day 3—that is , the first day that we observed an increase in their antifungal activity ( Figure 1 ) . After normalising to a housekeeping gene ( 18s rRNA ) , elevated expression was observed in nestmates of fungus-exposed individuals relative to nestmates of control-treated individuals for both defensin and PPO ( Figure 6; defensin: Welch's t test; Welch t = −2 . 348 , df = 26 , p = 0 . 032; PPO: t test; t = −2 . 923 , df = 26 , p = 0 . 007 ) , whereas cathepsin L showed no difference ( t test; t = −0 . 094 , df = 26 , p = 0 . 926 ) . This reveals an active upregulation of immune gene expression in nestmates of fungus-exposed ants and suggests the induction of a specific immune response distinct from immune responses to bacteria ( Figure S4; [16] ) . Similar specific immune upregulation after fungal infection is known to occur in Drosophila [77] . To determine if the observed specificity in our candidate gene approach , which is limited to a small set of genes , reflects specificity at the functional level , we tested the nestmates' capacity to inhibit growth of the bacterium Arthrobacter globiformis in an “antibacterial activity assay . ” We found that nestmates exhibited similar antibacterial activity for fungus and control treatment ( Figure 7; t test: t = −0 . 644 , df = 18 , p = 0 . 528 ) , revealing that social immunisation after fungal exposure of the colony is specific and does not lead to a protective effect against bacteria . We developed an epidemiological model to explore the adaptive value and colony-level long-term effects of social immunisation . We compared the effect of active versus passive immunisation in our ant-fungus system by extending classical SIS and SIR ( Susceptible-Infectious-Recovered/Removed ) models , which describe the progress of epidemics over time using the simplification that the diversity in the population can be reduced to a few states . Possible states in SIR models include individuals susceptible to the disease outbreak ( S ) , infectious individuals ( I ) , and recovered or dead individuals ( R; [78] , [79] ) . We included an active or passive immunisation mechanism by constructing a SIRM ( Susceptible-Infectious-Removed-iMmune ) model , in which ants can take five different states . Healthy nestmates are defined as susceptible ( S ) individuals , pathogen-exposed individuals as infectious ( I ) ones , and individuals dying from the disease are removed ( R ) from the model . Successful immunisation ( by active or passive immunisation ) leads to initially immune ( Mi ) individuals that may persist to create late-stage immune individuals ( Ml; Figure 8 ) . We describe the mean number of ants in each state by ordinary differential equations ( ODEs; for details , see Text S2 ) . We have thereby chosen a simple approach focusing on the comparison of active versus passive immunisation , but not taking into account spatial effects on epidemiology in societies that have been modelled elsewhere by cellular automata [27] , [80] , [81] or pair-wise approximations models [82] . Ants can change their state by social interactions with each other and depending on their infection state ( Figure 8A , B ) . Allogrooming reduces the fungus load of infectious ( I ) , changing them to susceptible ( S ) , but at the same time can increase the fungus load of the susceptible individuals ( S ) , changing them to infectious ( I ) . Active immunisation can occur when individuals receive a low-level infection and actively build up immunity , changing from infectious ( I ) to immune ( Mi ) with a given active immunisation rate . Under passive immunisation , susceptible ( S ) individuals change directly to the immune state ( Mi ) with a passive immunisation rate when receiving antimicrobial substances from infectious ( I ) individuals . Under the active immunisation scenario , initially immune ants ( Mi ) may then either die ( R ) if infection levels are too high and lead to the disease or enter into the later stage of immunity ( Ml ) . Under passive immunisation , all initially immunised individuals become late-stage immune . Late-stage immune ants ( Ml ) can then lose their immunisation and become susceptible individuals ( S; see Figure 8A , B and Text S2 ) . Each transition is governed by a transition rate , which in total were fixed to similar ranges in order to allow easy model comparison . The following qualitative results did not depend on the precise rate values , so that we report only representative outcomes of our simulations in Figure 8C , D . We found that more individuals typically reach the immune state ( Mi , and turn into Ml ) after passive immunisation ( Figure 8C ) , as a single infectious individual may immunise multiple susceptible nestmates , whereas actively immunised ants need to first be in the infectious state themselves . Yet we found that infections die out ( I becomes 0 ) more quickly under active immunisation ( Figure 8D ) , leaving only a very small reservoir for individuals to become immunised . Moreover , active immunisation leads to a lower number of dead individuals ( R ) . This is despite the fact that contraction of disease through pathogen transfer can only occur in the active route ( with a risk of dying similar to our experimental outcome ) . Increasing this risk leads to higher death rates and lower immunisation in a linear relationship ( simulations not shown ) . Taken together , active immunisation via pathogen transfer seems beneficial , as it allows more rapid disease elimination and produces lower death rates in colonies , except if the pathogen requires only a very low exposure dose to establish lethal infections in its host . In this study , we identified active immunisation as the underlying mode of social group-level immunisation in ant societies after fungal exposure of single individuals . Social contact to a fungus-exposed individual led to low-level infections in the majority of previously naive nestmates ( Figures 3 , S1 , S3 ) and to a higher capacity to inhibit fungal growth ( Figure 1 ) . We found that these low-level infections per se , even in the absence of social contact , are necessary and sufficient to explain the increased antifungal activity of nestmates ( Figure 4 ) . We found no evidence for visual or volatile chemical cues acting as additional trigger signals for the immune stimulation of the nestmates . Furthermore , neither ant behaviour ( Figure 2 ) nor physiology ( Figure 5C , D ) gave an indication for passive nestmate immunisation via transfer of antimicrobials from either exposed ants or their nestmates to the other group members . Finally , experimental elimination of the active route resulted in the absence of protective antifungal activity in nestmates ( Figure 5A , B ) . The increased immune activity of nestmates of fungus-exposed individuals correlates with an increased expression of immune genes such as the antimicrobial peptide defensin and the enzyme , prophenoloxidase ( PPO , Figure 6A , B ) , which both have known antifungal properties [55] , [83] . Cathepsin L , a lysosomal protease rather involved in antibacterial and antiviral responses ( [75] , [76]; Figure S4 ) , was not expressed at higher levels in nestmates of fungus-exposed compared to control-treated ants ( Figure 6C ) . In addition to the specific immune gene upregulation revealed by our candidate gene approach , we also found in a functional assay that nestmate immunity is not generally increased , but acts against Metarhizium fungus ( Figure 1 ) and not Arthrobacter bacteria ( Figure 7 ) . Precisely how specific social immunisation is at both the functional and gene expression levels remains to be addressed , and will be facilitated by the emerging genomic information on ants and other social insects [84]–[87] . To our knowledge , our study provides the first mechanistic explanation for the phenomenon of reduced susceptibility of nestmates after social contact to a fungus-exposed individual , that is , social immunisation , described for both ants [18] and termites [17] . Whether group-level immunisation in termite societies follows the same principle as in Lasius ants remains to be shown . Interestingly , our study on fungal exposure contrasts with findings of the suggested mechanisms of social immunisation of ants after bacterial exposure , where transfer of antimicrobial substances from the exposed individual via social feeding seems to elicit protection of nestmates [16] . We suggest that distinct infection modes of bacterial and fungal pathogens underlie these differences . Bacterial infections typically occur via oral uptake [39] , so that bacteria-exposed individuals do not carry socially transferable spores on their cuticle , as is the case with entomopathogenic fungi . Moreover , the long delay between exposure and infection is not common in bacterial infections , allowing for faster production of immune effectors in the exposed individuals and an earlier potential onset of immunisation . Social immunisation may not be limited to the highly eusocial insect societies but could similarly occur in other societies or at the family level . If also detected in vertebrates , the underlying mechanisms may be very different , as vertebrates have the additional adaptive/acquired immune component and do not rely solely on the innate immune system that characterises invertebrate immunity [1] , [21] . Humans have used the intentional transfer of low-level infections—referred to as “variolation” or “inoculation”—in an attempt to fight smallpox and frequently succeeded in creating long-term protection against this otherwise often deadly disease [62] , [63] . In humans , the technique was later replaced by less risky immunisation with attenuated strains as soon as these became available [88] , but variolation is still used for , for example , poultry disease management [64] . It is still unclear whether acquiring the protective low-level infections in ants is also an active strategy or , rather , an unintentional byproduct of social contact similar to “contact immunity” occurring in human societies , for example , after live strain polio or smallpox vaccination , where vaccinated individuals became spreaders and vaccinated their family members [89] , [90] . It is interesting that allogrooming by the ants is not restricted to single individuals , which would be a good strategy to avoid infecting the whole colony , but is rather performed by many colony members , all of which pick up a low-level infection . This may hint at social immunisation by low-level infections being an adaptive evolutionary strategy . Our epidemiological modeling indeed suggests that active immunisation is a beneficial strategy for ant colonies , as it allows for faster disease elimination and therefore leads to lower death rates than passive immunisation would . This is particularly true if exposure to low pathogen levels confers a low risk of mortality , as is the case with Metarhizium fungus , which requires relatively large doses to elicit a deadly course of disease . We therefore predict that social transfer of pathogens with higher infectivity [91] would not be an advantageous strategy for societies . A comparative analysis of mechanisms employed by social insects against pathogen types differing in their virulence and transmission would thus be highly interesting . Moreover , it seems likely that active immune stimulation following low-level infections may induce individual immune priming and , thereby , a longer lasting protection of colony members than if they simply received immune effectors . The long-lived societies of social insects [43] are at especially high risk of re-encountering the same pathogens multiple times during their lifespans [21] , and could greatly benefit from a persistent , rather than transient , social immunisation , particularly against common pathogens such as the fungus Metarhizium . To fully understand long-term epidemiological dynamics at the society level it will be indispensable to learn more about the mechanisms involved at the individual level—for example , to better understand if immune priming plays a role in social immunisation .
The unicolonial ant species Lasius neglectus [92] , [93] was sampled from four populations ( Jena , Germany; Volterra , Italy; Seva and Bellaterra , both Spain; for details on sample locations , see [94] ) and reared in the laboratory as described in Ugelvig and Cremer ( 2007 ) [18] . Behavioural observations were performed on workers collected in 2006 from all four populations , whereas all further experiments used L . neglectus workers collected in 2008 from Jena , Germany . Ants were kept at a constant temperature of 23°C with 75% humidity and a day/night cycle of 14 h light/10 h dark during the experiments . Experiments were performed in petri dishes with a plastered floor and 10% sucrose solution as food . We used the entomopathogenic fungus Metarhizium anisopliae var . anisopliae ( strain Ma 275 , KVL 03-143; obtained from Prof . J . Eilenberg , Faculty of Life Sciences , University of Copenhagen , Denmark ) to expose the ants in our experiments . To determine inhibition of fungal growth by ant material ( antifungal activity assay , see below ) and the transfer of conidia to the cuticle of nestmates traced by fluorescence microscopy , we used the RFP ( Red Fluorescent Protein ) labelled strain 2575 ( [95]; obtained from Prof . M . Bidochka , Brock University , Canada ) . For exposure of ants , we applied the fungal conidia ( conidiospores ) —that is , the dispersal form that is produced in a natural infection cycle from dead insect cadavers [30]—on the ants , whereas we used blastospores—that is , a single cell spore stage produced inside the body of the infected host [30] , [52]—for measuring the antifungal activity . Multiple aliquots of conidia of each strain were kept at −80°C and were grown on malt extract agar at 23°C for 2–4 wk prior to each experiment . Conidia were harvested by suspending them in 0 . 05% Triton X-100 ( Sigma ) and stored at 4°C for a maximum of 3–4 wk . All conidia suspensions had a germination rate of >98% as determined directly before each experiment . We produced liquid cultures of blastospores following an adjusted protocol by Kleespies and Zimmermann ( 1994 ) [96] , though growing the spores at 23°C . Blastospores were harvested by sieving them through a sterile 41 µm nylon net filter ( Merck Millipore ) . We exposed individual ant workers by applying a 0 . 3 µl droplet of a suspension of 109 conidia/ml in 0 . 05% Triton X solution ( fungus treatment ) , which corresponds to the lethal dose ( LD ) 50 for isolated ants . To obtain low-level infections in the same order as those picked up by the nestmates during social contact ( as confirmed by comparison of internal infection load of the socially transferred and directly applied group ) , we exposed the ants to 0 . 3 µl of a 105 conidia/ml suspension ( LD2 ) and kept them isolated . For the sham control , we treated the ants with a 0 . 3 µl droplet of a 0 . 05% Triton X solution only . Subsequently , the ants were dried on a piece of filter paper for several minutes . We grouped six workers ( 1 treated individual and 5 naive nestmates , to be distinguished by colour marking [Edding 780] ) and three larvae of L . neglectus in a petri dish ( Ø = 5 . 5 cm ) with a dampened plaster floor and a piece of filter paper ( 1×1 cm ) moistened with 10% sucrose solution as food supply . The treated individual received either a sham control or a fungus treatment as described above . Our experimental setup is equivalent to the experiment described in more detail in Ugelvig and Cremer ( 2007 ) [18] , which either led to a social immunisation of nestmates ( fungus treatment ) or not ( sham control ) after 5 d of social contact . We used this setup for observations of ant-ant interactions , obtaining physiological immune measures and conidia transmission analysis , yet made some measurements already after 1 , 2 , or 3 d of social contact . We changed this general setup for two experiments . First , to determine if signal transfer alone may be sufficient to elicit social immunisation in nestmates , we prevented direct social contact between the treated ant ( n = 10 for sham control and fungus treatment , respectively ) and its nestmates . This was done by keeping the treated individual in a plastic tube ( 200 µl , Ø of opening = 0 . 7 cm , containing cotton wool moistened with 10% sucrose solution ) , attached to the main petri dish , but separated by a double-layered nylon mesh ( mesh size 20 µm ) . The setup prevented direct physical contact yet allowed exchange of visual or volatile chemical signals . After 3 d , nestmates were frozen and subjected to the antifungal activity assay as described below . In a second setup , we excluded both signal and pathogen transfer from the exposed individual to its nestmates occurring in the first 2 experimental days , only allowing for potential later exchange of antimicrobial substances . To this end , we removed the exposed individual 2 d after fungal exposure from its “early nestmates” and placed it with “new nestmates” ( Figure 5A ) , the latter being tested for their antifungal activity after 3 d with the treated individual ( n = 10 replicates for sham control and fungus treatment , respectively ) . The new nestmates therefore only had contact to an exposed nestmate after conidia had firmly attached to the host's cuticle , and no longer could be transferred to nestmates ( as experimentally confirmed by absence of colony forming units [CFUs] in the new nestmates , see below ) . When removing the treated individual , we added five new nestmates to the five early nestmates ( Figure 5B ) to test if early nestmates may transfer immunity to the new nestmates in the form of antimicrobial substances . New nestmates were frozen after 3 d of social contact to the early nestmates of either the control-treated or fungus-exposed individual , and their antifungal activity measured as described below . All workers in the observed ant groups were individually colour marked . We then conducted 10 daily behavioural scan samples for each individual in each of six ant nests ( replicates ) from each of the four study populations ( total n = 24 ant groups per treatment , i . e . 288 ants ) over the 5 d of the experiment ( as described in [18] ) . We were interested in the behavioural interactions between different individuals , which we analysed separately for interactions between the treated individual ( total interactions n = 240 per treatment ) and its nestmates and among nestmates only ( total interactions n = 480 per treatment ) . The following types of interactive behaviours could be recorded: antennation ( recognition behaviour ) , allogrooming ( mutual cleaning of the body surface ) , and trophallaxis ( exchange of regurgitated liquid food; [38] ) . For statistical analysis of the behavioural data , see the statistics section below . We developed a sensitive antifungal and antibacterial assay ( MS , unpublished ) that reveals the antimicrobial activity of ant tissue via the growth inhibition of a pathogen culture ( as reduced absorbance in a spectrophotometer ) compared to a pathogen growth control without an ant sample . For each assay , we first determined the required ratio of pathogen , ant sample , and buffer to be in the linear range of the growth curve in which antimicrobial activity could be detected . We measured growth inhibition against blastospores of M . anisopliae by using either complete ants ( n = 10 replicate samples for each group ) , specific ant body parts ( gaster cuticle and thorax; n = 6 replicate samples for each group ) , or the trophallactic droplet ( n = 10 replicate samples for each group ) of treated ants ( sham control and fungus treatment ) and their respective nestmates . Most measurements were taken 3 d ( i . e . , 72 h ) after treatment of the single individual . Nestmates of control and exposed ants were also analysed on day 5 ( i . e . , 120 h ) after treatment . Bacterial growth inhibition against vegetative cells of A . globiformis was determined for the nestmates of fungus-exposed and control-treated individuals ( n = 10 replicates each ) . In all cases , the body parts or exudates from five individuals were pooled to obtain a single replicate sample . Both antifungal and antibacterial activity was determined as the reduction of either M . anisopliae fungal blastospore or A . globiformis bacterial vegetative cell growth , measured as absorbance in a spectrophotometer ( SpectraMax M2e , Molecular Devices , similar to [97] , [98] ) , after incubation of ant samples with the fungal or bacterial suspension . For detailed information , see Text S1 , and for statistical analyses , see below . We set up 15 experimental groups each consisting of five nestmates and one individual exposed to RFP-labelled conidia . After 2 d of social contact all ants were removed and frozen at −20°C . The cuticles of three random nestmates per group—that is , 45 nestmates in total—and cuticles from the 15 directly exposed individuals were examined for the presence of RFP-labelled conidia using a fluorescence microscope ( Leica MZ16 FA; Software: Leica Application Suite Advanced Fluorescence 2 . 3 . 0; Filter Cube: ET DsRed ) . Each ant was screened for the presence of conidia for a maximum duration of 30 min . In addition we checked the cuticle of 15 naive ants as negative control using the same method . We did not detect any structures resembling RFP-labelled conidia on any of the naive ants . We exposed 30 ants , kept them in individual petri dishes , and randomly assigned them to either of the three groups ( n = 10 ants each ) : ants that were frozen ( −20°C ) after 1 , 3 , or 5 d post-exposure . On day 1 post-exposure 10 of 10 ants were alive , 3 d post-exposure 8 of 10 ants survived , and 5 d post-exposure 4 of 10 ants survived . In addition , we set up 21 experimental groups , each consisting of five nestmates and one fungus-exposed individual , which were also frozen ( in equal numbers ) 1 , 3 , or 5 d post-exposure . None of the nestmates had died at this time point . All individually kept , directly exposed ants ( i . e . , 10 per day ) and two randomly chosen nestmates per experimental group ( i . e . , 14 per day ) were surface-sterilised in ethanol and sodium hypochlorite ( as described in [18] ) to destroy all fungal material on the cuticle prior to dissection under a stereomicroscope ( Leica S6E ) . For each ant , all contents of the gaster ( abdomen ) without the cuticle were removed and dissolved in 30 µl of Triton X . The body contents were then plated on selective medium agar plates ( containing: chloramphenicol 100 mg/l , streptomycin 50 mg/l , dodin 110 mg/l ) and kept at 23°C . After 2 wk of cultivation , the number of colony forming units ( CFUs ) per plate was determined . We identified CFUs as pure M . anisopliae cultures by morphological fungal determination and amplification of specific M . anisopliae genes by PCR ( see Text S1 ) . For statistical analysis , we used both presence/absence of CFUs for each individual and the number of CFUs growing out of infected ants ( for details , see statistical analysis section below ) . For method development , we performed the following negative controls: ( a ) 15 completely untreated ants and ( b ) 15 ants that were exposed to conidia but were surface-sterilised after 3 h ( i . e . , before the fungus could penetrate the cuticle and reach the inside of the ant ) . We did not detect any fungal growth from these 30 ants . Moreover , we could confirm that pathogen transfer did not occur towards the new nestmates of either directly exposed ants or early nestmates ( n = 14 replicates each ) . We set up 30 experimental groups consisting of five nestmates and one fungus-exposed individual each . After the 5 d of social contact to the exposed individuals , each nestmate was isolated in a single petri dish for another 12 d . During the whole experimental period of 17 d , the survival of nestmates was checked daily . Dead nestmates were surface-sterilised as above and put on moist filter paper in a petri dish at constant temperature , 23°C . Cadavers were checked for a period of 3 wk for the growth of M . anisopliae . The bacterium Bacillus thuringiensis ( strain NRRL B-18765 , obtained from the permanent strain collection of the Northern Research Laboratory , U . S . Department of Agriculture , Peoria , Illinois , USA ) was precultured in LB medium and grown to an OD600 of 0 . 1 . We centrifuged 1 ml of the suspension at a speed of 3 , 000× g for 5 min and discarded the supernatant to obtain a concentrated bacterial pellet as in [99] . Ants were immobilized and pricked ventrally between the 2nd and 3rd gaster sternite with a sterilized needle ( minutien needles , Sphinx V2A 0 . 1×12 mm , bioform ) dipped in either LB medium ( sham control ) or the concentrated bacterial pellet ( n = 10 ants per treatment , replicated three times; i . e . , total n = 30 ants per treatment ) . The ants were frozen for gene expression analysis 12 h after pricking . Ants were analysed either individually ( nestmates of Metarhizium-exposed ants ) or in pools of 10 ants ( bacterial septic injury ) by qPCR for gene expression of three immune genes and the housekeeping gene , 18s rRNA . For immune genes , we chose the antimicrobial peptide defensin [68] , [69] , the enzyme prophenoloxidase ( PPO [72] , [73] ) , and the lysosomal protease cathepsin L [74] , [76] . For details of the procedures on RNA extraction , cDNA preparation , and qPCR , please see Text S1 and the statistical analysis section below . We always tested the distributions underlying our data and chose the corresponding tests . If data were not normally distributed even after transformation , we applied models with specified error structures or non-parametric tests . Reported p values are two-sided . All statistical analyses were carried out in IBM SPSS Statistics version 19 . 0 or Sigma Stat 3 . 5 ( Systat Software Inc . ) . All figures are based on raw data . For the behavioural observations , we first analysed all behaviours overall over the 5 experimental days . Due to the nature of the data ( overdispersed count data ) , generalised linear models ( GLM ) with negative binomial errors and a log link function were employed using the following factors: treatment type ( fungus treatment versus sham control ) , ant pairing ( treated-nestmate versus nestmate-nestmate ) , and the interaction between them . As neither nests within populations nor populations behaved differently , they were not included in the final models . We give the likelihood ratio ( LR ) χ2 to test if our overall model explains the data better than a model with only the intercept . As we detected significant differences for allogrooming , we performed a second test to analyse the effect of time in the interactions between treated individuals and their nestmates for the two treatment types separately ( n = 240 ) using a GLM with repeated measures . Simple contrasts with day 1 as reference were employed to test the differences between day 1 and the succeeding days ( Figure 2B ) . For statistical analysis of the antifungal and antibacterial activity , the absorbance values ( optical density ) of the different treatment groups were compared by one-way ANOVAs or t tests as data were normally distributed or could be transformed to obtain normality . For the antifungal activity of nestmates of exposed versus control nestmates , we applied a GLM to analyse the effects of treatment type ( fungus treatment versus sham control ) and time ( day 3 versus day 5 post-treatment ) , as well as their interaction ( Figure 1 ) . For analysis of pathogen load , we compared directly exposed and nestmate ants for ( a ) the proportion of individuals that were infected ( i . e . , showed at least a single CFU; Fisher exact test ) and ( b ) the number of CFUs in the individuals that showed an infection ( Mann Whitney U test; Figure 3 ) . As the experimental grouping did not influence the number of CFUs found in nestmates from the same ant group , this factor could be excluded from statistical analysis comparing treated individuals and nestmates ( GLM with negative binomial errors , LR χ2 = 112 . 362 , df = 34 , p = 0 . 000; Replicate , Wald: χ2 = 21 . 273 , df = 17 , p = 0 . 214 ) . Gene expression analyses were run in two to three technical replicates . Normalised gene expression values ( the average of technical replicates , standardised to the housekeeping gene ) were either a priori normally distributed or could be normalised by transformation and were analysed using t test or—in the case of unequal variances between groups ( defensin , Figure 6A ) —Welch's t test for unequal variances [100] . We applied ordinary differential equations ( ODE ) to extend classical SIR modeling ( Susceptible-Infectious-Removed ) with an immunised state to a SIRM model ( Susceptible-Infectious-Removed-iMmune ) , in which the immune individuals were further separated into an initial and a late phase of immunity . See Figure 8A , B for the model and how we calculated state changes and Text S2 for model construction and simulations .
|
Close social contact facilitates pathogen transmission in societies , often causing epidemics . In contrast to this , we show that limited transmission of a fungal pathogen in ant colonies can be beneficial for the host , because it promotes “social immunisation” of healthy group members . We found that ants exposed to the fungus are heavily groomed by their healthy nestmates . Grooming removes a significant number of fungal conidiospores from the body surface of exposed ants and reduces their risk of falling sick . At the same time , previously healthy nestmates are themselves exposed to a small number of conidiospores , triggering low-level infections . These micro-infections are not deadly , but result in upregulated expression of a specific set of immune genes and pathogen-specific protective immune stimulation . Pathogen transfer by social interactions is therefore the underlying mechanism of social immunisation against fungal infections in ant societies . There is a similarity between such natural social immunisation and human efforts to induce immunity against deadly diseases , such as smallpox . Before vaccination with dead or attenuated strains was invented , immunity in human societies was induced by actively transferring low-level infections ( “variolation” ) , just like in ants .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"mycology",
"zoology",
"immunity",
"immunology",
"biology",
"evolutionary",
"biology",
"microbiology"
] |
2012
|
Social Transfer of Pathogenic Fungus Promotes Active Immunisation in Ant Colonies
|
Kaposi’s sarcoma-associated herpesvirus ( KSHV ) is the etiologic agent of Kaposi’s sarcoma ( KS ) . It is endemic in a number of sub-Saharan African countries with infection rate of >50% . The high prevalence of HIV-1 coupled with late presentation of advanced cancer staging make KS the leading cancer in the region with poor prognosis and high mortality . Disease markers and cellular functions associated with KS tumorigenesis remain ill-defined . Several studies have attempted to investigate changes of the gene profile with in vitro infection of monoculture models , which are not likely to reflect the cellular complexity of the in vivo lesion environment . Our approach is to characterize and compare the gene expression profile in KS lesions versus non-cancer tissues from the same individual . Such comparisons could identify pathways critical for KS formation and maintenance . This is the first study that utilized high throughput RNA-seq to characterize the viral and cellular transcriptome in tumor and non-cancer biopsies of African epidemic KS patients . These patients were treated anti-retroviral therapy with undetectable HIV-1 plasma viral load . We found remarkable variability in the viral transcriptome among these patients , with viral latency and immune modulation genes most abundantly expressed . The presence of KSHV also significantly affected the cellular transcriptome profile . Specifically , genes involved in lipid and glucose metabolism disorder pathways were substantially affected . Moreover , infiltration of immune cells into the tumor did not prevent KS formation , suggesting some functional deficits of these cells . Lastly , we found only minimal overlaps between our in vivo cellular transcriptome dataset with those from in vitro studies , reflecting the limitation of in vitro models in representing tumor lesions . These findings could lead to the identification of diagnostic and therapeutic markers for KS , and will provide bases for further mechanistic studies on the functions of both viral and cellular genes that are involved .
Kaposi’s sarcoma ( KS ) is a multi-focal pleomorphic , highly vascularized tumor that is one of the AIDS defining illnesses . KS mostly manifest as cutaneous lesions , but it can also be detected in the oral mucosa , lymph nodes and all visceral organ systems . There are four types of KS: iatrogenic , classic , endemic and epidemic/AIDS-related KS [1–4] . Endemic and epidemic KS are the most common forms in sub-Sahara Africa . Endemic KS occurs in HIV-1 negative African patients of both genders . The presentation in both genders at varying ages differentiates this form of KS from classical KS found mostly in elderly Mediterranean men . Epidemic KS occurs in patients who are HIV-1 positive , and along with iatrogenic KS , which results from chemical immunosuppression , highlights the role of immune dysregulation in KS development . Depending on the location of lesions , KS patients may display signs and symptoms such as nausea , vomiting , dysphagia and dyspnea . Cutaneous lesions may also disfigure and lead to social stigma for KS patients [5 , 6] . The prognosis of KS patients in sub-Sahara Africa is generally poor , often due to delayed diagnosis and late initiation of treatment [7] . Even after receiving anti-retroviral drugs therapy ( ART ) , the mortality of epidemic KS patient is often high due to advanced KS disease staging , simultaneous advanced HIV-1 diagnosis and both HIV-1 and KS-immune reconstitution inflammatory syndromes [8] . The etiologic agent associated with KS is the Kaposi’s sarcoma-associated herpesvirus ( KSHV ) or human herpesvirus type 8 ( HHV-8 ) which is a member of the gamma-herpesviridae . It is an enveloped virus with a double-stranded viral DNA of approximately 170-Kb . The KSHV genome encodes more than 80 genes that are expressed in a regulated transcriptional program that promotes latency with very limited viral expression , or supports lytic replication with the production of progeny virions . The cellular tropism of KSHV includes epithelial , endothelial , B cells , and more recently has been expanded to include neurons [9–17] . Depending on the host cell environment encountered upon infection , the virus may establish latency and only express latency related genes such as latency-associated nuclear antigen ( LANA ) . LANA expression helps KSHV prevent programmed cell death and circumvent host immune surveillance [18–22] . During reactivation , the entire KSHV genetic repertoire is expressed enabling virus replication and propagation . KSHV has a high seroprevalence in sub-Sahara Africa , where over 50% of the population has detectable anti-KSHV antibodies [23] . The virus can be transmitted by saliva and sexual routes [24–28] . In addition to KS , KSHV is also the etiological agent of other lymphoproliferative disorders such as multicentric Castleman’s disease ( MCD ) and primary effusion lymphoma ( PEL ) [15 , 29 , 30] . KS can be an opportunistic malignancy commonly presenting in untreated HIV-1 infected patients [31] . Due to the high HIV-1 prevalence in sub-Sahara Africa countries such as Zambia and Tanzania , KS is the leading cancer in men and the third leading cancer in women in Africa [32 , 33] . In contrast to the drastic reduction of KS in resource-rich regions , widespread implementation of ART in sub-Saharan Africa , has only modestly reduced KS incidence among the resource-limited countries of the region [34 , 35] . There are currently no existing targeted immunotherapies against viral or host proteins in KS , and there is no vaccine against KSHV infection . Indeed the correlates of KS development or protection are ill-defined . A more complete understanding of changes in host and viral gene expression in KS lesions compared with non-tumor tissues could provide insights into the interactions between KSHV and the host that drive development of KS . Moreover , it would allow exploration for any potential biomarkers that are unique to KS neoplastic cells . These biomarkers could make early detection of KS possible and perhaps help to eliminate one of the major malignancies in sub-Saharan Africa . In addition , gene expression profiles of KS lesions may also lead to identification of pathways required for KS neoplasia . Inhibition of such essential pathways could potentially provide therapeutic benefit for KS patients . RNA-Seq is an extremely high density , unbiased sequencing technique that allows high-throughput assessment of total cellular transcription . It exhibits a broader dynamic range and specificity than microarray analyses [36 , 37] . Thus RNA-Seq is an ideal tools for defining the gene expression profiles or transcriptome of KS lesions . RNA-Seq has been utilized in transcriptomic analysis of various forms of cancers to identify pathways that could provide potential targets for future diagnosis or therapy [38 , 39] . However , only a limited number of in vitro studies on KSHV infected cell lines or infected monocultures of specific lineages of cells , have utilized either RNA-Seq or microarray for transcriptomic analyses [40–47] . In the current study , epidemic KS lesion biopsies were collected from Zambia and Tanzania for comparative transcriptomic analysis using RNA-Seq . Ipsilateral or contralateral sampling of normal tissues from the same KS patients provided for direct intra-subject comparison of gene expression profiles . We found that KSHV has significantly impacted the cellular gene expression that were involved in the lipid and glucose metabolism disorder pathways . To our knowledge , this is the first study to utilize RNA-Seq to provide unbiased gene expression profile directly from KS lesions in comparison to that of non-neoplastic tissue from the same subject . Our study provides unique insights into the interactions between KSHV and the host and may point to previously unrecognized pathways in KS pathogenesis .
Permission to conduct this study was obtained from Tanzania National Institute for Medical Research , Ocean Road Cancer Institute Review Board , University of Zambia Biomedical Research Ethics Committee and the Institutional Review Board of the University of Nebraska-Lincoln . Written informed consents were obtained from all study participants and the study did not interfere with the routine clinical care of the participants as per institute’s guidelines . Histologically confirmed epidemic KS patients ( ≥ 18 years old ) were recruited from University Teaching Hospital ( UTH ) , Zambia and Ocean Road Cancer Institute ( ORCI ) , Tanzania . Whole blood was collected from each patient and separated into plasma and mononuclear cells using Lymphoprep according to the manufacturer’s protocol ( Stemcell technologies , Massachusetts , USA ) . A 4 mm cylindrical single-use punch was used to collect biopsy samples from a representative KS lesion and a second identical punch biopsy was collected from an uninvolved ipsilateral /contralateral site on the same individual . The collected tissues were treated overnight in RNALater ( Ambion ) to facilitate subsequent DNA/RNA extraction . RNALater was removed prior to freezing at -80°C . The HIV-1 status for the Tanzanian KS patient was determined using the Tanzania HIV Rapid Test Algorithm . Zambian KS patients’ HIV-1 status was determined using the Alere Determine HIV-1/2 Ag/Ab Combo test . The HIV-1 serostatus of all subjects was re-confirmed at the University of Nebraska-Lincoln with the HIV-1-2 . 0 First Response kit ( Premier Medical Corporation Limited , Daman , India ) . HIV-1 viral RNA was extracted from plasma using the QIAamp viral RNA mini kit following the manufacturer’s protocol ( Qiagen , Hildren , Germany ) . HIV-1 plasma viral load was quantified by RT-PCR ( QuantStudio 3 , Applied Biosystems , Carlsbad , CA ) using oligonucleotides HIV-1 LTR forward primer [5’-GCCTCAATAAAGCTTGCCTTGA-3’] , reverse primer [5’- GGGCGCCACTGCTAGAGA-3’] and probe [5’-FAM-CCAGAGTCACACAACAGACGGGCACA-BHQ_1–3’] under the following reaction conditions: 50°C for 15 min , 95°C for 2 min; 45 cycles at 95°C for 15 sec and 60°C for 30 sec . KSHV DNA was extracted from the patient plasma with QIAmp DNA mini kit ( Qiagen , Hilden , Germany ) according to manufacturer’s protocol . The extracted DNA was confirmed to be PCR negative for β-actin to demonstrate the absence of cellular DNA contamination . To detect the presence of KSHV DNA , nested-PCR for KSHV ORF26 was performed on the extracted DNA ( 1st round PCR primers: forward [5’-AGCCGAAAGATTCCACCAT-3’] and reverse [5’-TCCGTGTTGTCTACGTCCAG-3’] , 2nd round PCR primers: Forward [5’-CGAATCCAACGGATTTGACCTC-3’] and reverse [5’-CCCATAAATGACACATTGGTGGTA-3’] ) using the following conditions: 95°C for 5 min; 35 cycles at 95°C for 30 sec , 58°C for 30 sec , 72°C for 30 sec; 72°C for 7 min . The final PCR product was analyzed on a 1% TAE agarose gel versus a KSHV ORF26 DNA positive amplification product at 173 base pairs . Genomic DNA from the KSHV chronically infected BC3 cell line was used as positive control . Fresh frozen biopsies from KS lesions were homogenized by cryo-cracking with liquid nitrogen . Genomic DNA was extracted from the homogenized sample using Puregene genomic DNA purification kit ( Qiagen , Hilden , Germany ) following the manufacturer’s protocol . To quantify the KSHV DNA load from the lesion , the extracted DNA was analyzed by real-time PCR against KSHV ORF26 with the 2nd round PCR primers set listed above and the following probe [5’-FAM-CCATGGTCGTGCCGCACGCA-BHQ_1–3’] . A plasmid encoding KSHV ORF26 was used to create a standard curve to compare patient DNA amplification signals . The number of cells analyzed was quantified by comparing patient DNAs to a standard curve produced by real-time PCR for the β-globin gene . Genomic DNA from the human cell line 8E5 was used as a standard . All reactions were performed in triplicate under the following conditions: 50°C for 2 min , 95°C for 10 min; 40 cycles of 95°C for 15 sec , and 60°C for 1 min . Fresh frozen biopsies from KS lesions and ipsilateral/contralateral normal sites were homogenized by cryo-cracking in liquid nitrogen using a mortar and pestle . Briefly , the mortar and pestle were cleaned and chilled in a metal tray with liquid nitrogen . The sample was transferred from -80°C storage on dry ice into the pre-chilled mortar , followed by pulverize and homogenize using the pre-chilled pestle . The homogenized sample is then transferred into a 2-ml Eppendorf tube by a pre-chilled spatula for genomic RNA extraction . Genomic RNA was extracted from the homogenized sample with miRNeasy mini kit with on-column DNase I treatment according to the manufacturer’s protocol ( Qiagen , Hilden , Germany ) . Concentration of the extracted RNA was measured with a Qubit fluorometer using the Qubit RNA Broad-Range kit ( Invitrogen , Waltham , MA ) . Quality of the extracted RNA was assessed by an Agilent Bio-Analyzer system ( Agilent Technologies , Santa Clara , CA ) for fragment analysis . By measuring the 18s and 28s rRNA peaks within the electrophoretic trace , the RNA Integrity Number ( RIN ) was calculated to determine the level of RNA integrity . Only samples with an RIN >4 . 9 were used for library preparation and RNA-Seq . Library preparation was performed using the TruSeq™ RNA Library Prep Kit v2 , and RNA-Seq data were collected using an Illumina HiSeq2500 in single-read , 50 bp Rapid Run mode at the University of Nebraska Medical Center DNA Sequencing Core . The RNA-seq data was aligned against human ( hg19 genome , Ensemble v75 transcriptome ) or KSHV ( NC_009333 ) genomes and transcriptomes using the bowtie2 algorithm , and RSEM v1 . 2 . 31 software was used to estimate read counts and FPKM values at the gene level [48 , 49] . Raw counts were converted to log2 ( 5+count ) values and quantile normalized to use in Principle Component Analysis and gene expression heatmaps . Significance and fold change of differential expression between lesion and control samples was estimated using the DESeq2 method on raw values and genes with false discovery rate ( FDR ) <5% were considered as significant [50] . An additional threshold of 5-fold was used to enumerate a set of most-changed genes between conditions . Pearson correlation was used to test associations between KSHV transcript loads in tissues for significantly differentially expressed genes . Correlations with p<0 . 05 were considered significant . KSHV gene functional definitions were obtained from Arias et . al . [51] . Hierarchical clustering based on the KSHV gene expression was done on the transcripts with at least 10 counts in at least one sample using Spearman correlation distance for genes and Euclidean distance for patients using average linkage . The final gene clusters were defined using a distance threshold of 0 . 5 . Functions were then assigned for each cluster based on the function with the best enrichment ratio E = % genes in cluster / % genes total . Hierarchical clustering and principal component analysis was done on z-score converted normalized values using MATLAB R2016a ( v9 . 0 . 0 ) . Expression heatmaps were plotted in Microsoft Excel using normalized values centered versus average across all samples . Gene set enrichment analysis was done using Qiagen’s Ingenuity Pathway Analysis software ( IPA , QIAGEN Redwood City , www . qiagen . com/ingenuity ) using “Upstream Analysis” , “Regulator Effects” and “Diseases & Functions” options . Activation Z-scores ( Z ) , calculated by IPA are a result of combining mRNA expression changes from the experiment and known effect of the gene on function or upstream regulator on the target ( e . g every downregulated gene suppressing a function or upregulated gene activating a function will contribute to predictions as to whether the function is activated ) . |Z|>2 predictions were considered significant . Upstream regulators that were significantly differentially expressed at the mRNA level , had significantly overrepresented number of known targets ( p<10−5 ) and had a significantly predicted activation state ( |Z|>2 ) were considered . The only regulator effect network with consistency score >100 is reported ( the next best network had a score of 43 ) . Functions that passed p<10−15 threshold or P<10−5 and |Z|>2 threshold were considered . For analyzing relationships between genes involved in the lipid and glucose models , Ingenuity Knowledgebase was used to derive information for all considered functions . A combined model was built to demonstrate two aspects: overlap between lipid-related functions , and the cross-talk of lipid functions with glucose metabolism disorder . Functions that shared 100% of genes with another function were omitted ( e . g . all genes from metabolism and synthesis of triacylglycerol were in metabolism of acylglycerol and all genes from concentration of acylglycerol were in concentration of lipid ) . Three of the other 6 lipid-related functions had most , except for 1 or 2 ( >90% ) genes as members of a more general function and are depicted in the model as sub-functions . Information about genes’ effect on each function was considered and the potential cross-talk is annotated . Genes that did not have a consensus effect on any of the functions received an “affects” designation . If a gene was known to increase or decrease activity of any function , it was assigned to one or to both effect categories in cases of both increase and decrease calls . Additional information includes number of genes specific to the particular function ( not involved in any other considered functions ) and the overall number of genes from the 3 effect categories for all major functions . CIBERSORT software predictions were based on gene FPKM values using the LM22 cell signature model in “absolute” mode [52] . Cell abundance was then compared using a two-sample t-test . Significance was defined at p<0 . 05 . RNA-seq data from external GEO sets ( GSE62829 , GSE62344 , GSE84237 ) were processed as described above and tested for differential gene expression using DESeq2 . Microarray data from external sets ( GSE6489 , GSE45590 , GSE66682 ) was quantile normalized and log2-transformed before statistical evaluation by one-way ANOVA with correction for multiple testing to estimate FDR as previously described [53] . The original RNA-seq data from this study was uploaded to the GEO database ( https://www . ncbi . nlm . nih . gov/geo/ ) with the accession number GSE100684 . To verify the RNA transcript level changes observed from the transcriptomic profiling data , a two-step qRT-PCR methods was used to measure the RNA transcript level for selected host genes . First , DNase I-treated total RNA extracted from the tissues was subjected to cDNA synthesis using Superscript III reverse transcriptase primed with random 9-nonamers ( Invitrogen , Waltham , MA ) . The amount of cDNA was measured with Qubit ssDNA assay kit ( Invitrogen , Waltham , MA ) . Real-time PCR reactions were prepared using iQ™ SYBR Green supermix ( Biorad , Hercules , CA ) . Each PCR reaction contained 20 ng of cDNA with primer pairs for the respective host genes ( S1 Table ) . qRT-PCR was performed in a QuantStudio3 instrument ( Applied Biosystems , Waltham , MA ) under the following cycling parameters: 50°C for 2 min , 95°C for 10 min; 40 cycles at 95°C for 15 sec and 60°C for 1 min; followed by melt-curve parameters: 95°C for 15 sec , 60°C for 1 min , 95°C for 15 sec . Lesion tissues were measured in triplicate , while normal tissues were measured in duplicate due to reduced cDNA yield . Glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) was used as an internal standard for each target gene . The RNA transcript level was presented as relative quantification ( RQ ) using the comparative cycle threshold ( ΔΔCT ) method . Briefly , the ΔCT for each target gene was calculated by subtracting the CT for GAPDH in the same sample from the CT for target gene . For each target gene , their respective average ΔCT from the normal tissue was used as a reference for the calculation of ΔΔCT by subtracting ΔCT for normal tissue from the ΔCT for lesion tissue . Finally , the RQ for each target gene is defined as 2- ΔΔCT .
Four male HIV-1 positive epidemic KS patients 37 to 54 years of age were recruited from Zambia and Tanzania . Two of the four patients ( p32 and p83 ) had been on ART for over 2 years at the time of inclusion in the study . The other patients ( p22 and p23 ) had received ART for 2 to 3 months ( Table 1 ) . All patients were fully suppressed with undetectable HIV-1 plasma viral load . However , KSHV viral DNA was detected in the plasma of three out of four patients . Quantification of KSHV load in the extracted lesion tissue DNA revealed 1 . 14x105 and 8 . 17x105 copies/million cells in the long-term ART treated patients ( p32 and p83 respectively ) , while the recently ART treated patients ( p22 and p23 ) had a lower KSHV load at 6 . 93x103 and 5 . 44x104 copies/million cells . The extracted RNAs from lesions and control tissues were used for subsequent RNA-seq analysis . The RNA extracted from the lesion and control tissues were sequenced at a depth of over 12 million reads/sample ( Table 2 ) . After passing the standard quality control , reads with good quality scores , were aligned against the reference human and KSHV genomes and transcriptomes ( Fig 1 ) . The majority of reads aligned to the human transcriptome , with at least 79% and 72% alignment rate for the lesion and control tissues , respectively . As expected , there were more reads aligned to the KSHV transcriptome in KS lesions , ( range 718–17202 reads ) compared to control tissues ( range 0–60 reads ) ( Table 2 ) . Although low KSHV transcripts reads were detected in some control tissues , it represent less than 2% of the viral reads detected in the corresponding KS lesion and likely originated from the KSHV positive circulating peripheral blood mononuclear cells ( PBMC ) . After normalization to the total number of aligned reads , the KSHV RNA load within each tissue was estimated from the total read count aligned to the KSHV transcriptome . Gene expression profiles for both human and KSHV transcriptome were then determined for both the lesion and control tissues , as well as in correlation with the KSHV RNA load in the lesion . Genes that were demonstrated to be significantly up or downregulated were then selected for upstream regulator , functional and pathway analyses to explore potential mechanisms dysregulated by KSHV in KS lesions . Finally , genes that were differentially expressed in the lesions were comparatively overlapped with other published KSHV microarray or RNA-seq datasets . Alignment of the KSHV gene expression profile in the lesion and control tissues to the reference KSHV genome and transcriptome showed that nearly no KSHV expression was detected in the control tissues , except in patient p83 , where 60 reads in the control tissue aligned to the KSHV transcriptome ( Fig 2A ) . In contrast , robust KSHV gene expression was readily detected in all lesion tissues . Twenty viral genes were expressed at levels ≥ 6-fold than in the control tissues . These differences were statistically significant with a false discovery rate ( FDR ) of ≤ 5% with additional 15 genes passing nominal p ≤ 0 . 05 ( Fig 2A ) . Other viral genes were also detected in the lesion tissues , but differentials did not reach significance ( Fig 2B ) . Lack of statistical significance is a result of variant levels of expression of these viral genes among the four lesion samples . For example , K5 and vIRF3 gene expression was clearly highly upregulated in the lesion tissue of patient p23 compare to its corresponding control tissue , yet the significance of this difference is masked by the relatively low expression of the same gene in lesions from the other three patients . Next , we explored how the differences in viral gene expressions relate to their known or presumed functions . Viral genes were classified based on the functional categories of latency , expression and replication , structural proteins , immunomodulation and undefined functions . Normalized expression of genes showing at least 10 aligned reads was used for hierarchical clustering . The analysis yielded 6 major clusters of genes showing distinct patterns among the four patients ( Fig 2C ) . While all lesions tissues expressed latency genes at high levels , patients p83 and p23 had relatively higher expression of those genes ( cluster 4 ) . Relative expression in the lesion tissues from patients p22 and p23 was higher for genes from the expression and replication categories ( clusters 2 , 3 ) . KSHV structural clusters was defined by upregulation in patients p22 and p23 ( cluster 1 ) and patients p23 and p32 ( cluster 6 ) , suggesting a higher level of lytic replication in tissue . Immunomodulation related genes were expressed in all patients , with patient p23 displaying the highest expression level ( cluster 5 ) . After alignment to human transcriptome , relationships between sample were studied by Principle Component Analysis which indicated 40% and 17% sample variability ( the first and second principal component , respectively ) ( Fig 3A ) . We note that the first principal component mainly corresponds to the KSHV RNA load within the tissue . While the lesion and control tissues formed distinct clusters , the control tissues shows less variability . In addition , control tissue from patient p83 is unique as it associates more closely to the lesion tissues . This association is likely due to the presence of low KSHV RNA level in this control tissue , resulting in marginally similar transcriptional profile to those of lesion samples . Direct comparison of gene expression between the collective lesion and control samples , revealed 3589 significantly differentially expressed genes ( FDR < 5% ) , 1096 of which have ≥ 5 fold changes ( Fig 3B ) and 52 genes showing ≥ 20-fold change ( Fig 3C ) . A comparison of gene expressions among the 4 lesion/control tissue pairs is shown in S2 Table , demonstrating the highly similar expression pattern among the samples . Expression heatmap for the top 30 known most changed genes , regardless of their correlation with KSHV RNA load within the tissues , are shown for each tissue sample in S1 Fig . In addition , 311 genes in the lesion tissues were both significantly differentially expressed ( FDR < 5% ) and directly correlated with the KSHV RNA load within the tissues ( p < 0 . 05 ) with top 30 genes shown in Fig 4 . A subset of the gene expression results determined by the RNA-seq analysis were validated by real-time PCR . Six cellular genes , 4 upregulated and 2 downregulated genes , from the lesion tissues were selected . These genes were distributed across the range of expression changes between lesion and control tissues and each target was strongly correlated with KSHV RNA load in the tissues . The up or downregulation expression of these genes as quantified by real time-PCR was concordant with the results obtained through RNA-seq analysis . Patient p23 displayed the highest fold increases in the expression of the 4 upregulated genes examined ( Fig 5A ) . A similar pattern was observed among the downregulated genes , where the patient p23 has evinced the highest fold decreases ( Fig 5B ) . Together , these results confirmed the validity of the gene expression profiles determined from RNA-seq analysis . Given the profound changes in gene expression profiles observed in KS tissues , the levels of transcriptional regulators for these genes may have been altered and could impact downstream biological processes . To investigate this issue , we tested the set of genes that were significantly differentially expressed in the lesions for enrichment of known key transcriptional regulators using the Upstream Regulator Analysis within the Ingenuity Pathway Analysis ( IPA ) . IPA Knowledgebase contains information about known regulators targets and can predict which transcriptional regulators are responsible for certain biological processes or it can suggest pathways and the potential impact regulators may have on those pathways . Our analysis identified 5 genes upregulated in lesion tissues with a significant number of their targets ( p<10−5 ) changed in a direction which indicates an activated regulator status on a protein level ( positive Z-score >2 ) ( Fig 6A ) . The top activated regulator was the tumor growth factor beta-1 ( TGFB1 ) , which potentially regulates the expression or function of 399 different genes and plays an important role in various immune responses such as regulation of B cells [54] . Another 8 potential regulators with decreased expression showed a negative Z-score , indicating an inhibited status ( Fig 6A ) . The most inhibited regulator was adiponectin ( ADIPOQ ) , which affects 45 different genes involved in various metabolic pathways . In order to find other regulators that may impact downstream pathways , we screened the data set with the IPA Regulator Effects analysis . The top scored regulator effect network was the interferon alpha network that drives activation of blood-cell related functions , such as cellular adhesion , extravasation and recruitment of leukocytes , suggesting tissues are producing signals to promote increased infiltration of immune cells into the lesions ( Fig 6B ) . To better understand which diseases or biological functions were the most correlated with the observed gene expression profiles , we focused our analysis on 884 genes that have at least 5-fold expression changes in the lesion relative to control tissue . As indicated by their p-values , the top 7 functions impacted by KS altered gene expression were all related to cancer . However , we cannot determine if these functions were increased or decreased due to lack of information in the IPA ( Fig 7A ) . More importantly , 4 functions that showed increased activities ( Z-Score > 2 ) were closely associated with cancer ( Fig 7A ) . In addition , another 115 genes associated with glucose metabolism disorder had increased activities ( p = 7x10-13 , Z-Score = 2 . 28 ) . Among the significantly enriched functions that showed decreased activities in the lesion tissues , 8 lipid metabolism functions and 7 functions related to small molecule biochemistry were revealed . These analysis suggested that fundamental glucose and lipid metabolic pathways are substantially altered in the KS lesions . To further illuminate the relationship of genes involved in the lipid and glucose metabolism in the lesion , the Ingenuity Knowledgebase was used to derive information for all considered functions . Significant downregulation of 49 of 67 genes ( 73% ) with overlap between the 6 lipids functions was noted ( Fig 7B , S3 Table ) . In contrast , the downregulation of genes involved in the glucose pathway associate with increased activity of glucose metabolism disorder , which in turn , could both increase or decrease the activity of the lipid metabolism . One of the roles for immune cells is to surveil against foreign antigens , however , such surveillance , if present , clearly did not prevent the development of KS lesions in these evaluated cases . To test the hypothesis that failure of immune cells to target and infiltrate tumor tissue contributed to KS , we interrogated the cellular gene expression dataset with the CIBERSORT software to estimate the proportions of immune cells in tissues . We found that the levels of B cells , macrophages and NK cells in lesions were predicted to be significantly higher than in control tissues ( p = 0 . 004 , p = 0 . 019 and p = 0 . 048 , respectively ) , and mast cells were lower ( p = 0 . 036 ) ( Fig 8 ) . T cell infiltration into the lesions was only predicted in 3 of 4 patients , and gene signatures associated with dendritic cell infiltration was only detected in lesions from patients p32 and p83 , but were not higher than in control tissues . These results are congruent with the regulator effect prediction above that suggested KS tumors were producing chemo-attractants to promote infiltration of immune cells . Since most of the available KS or KSHV infection gene expression data has been derived from in vitro experiments in cell lines , we investigated the overlap of our in vivo lesion transcriptomes with these dataset derived from model systems ( S4 Table ) . Gene expression data from 7 published studies were analyzed [41–47] . Only 3 of the 7 showed significant overlap with our data . Despite variability in the cell types employed in those 3 studies , our analysis indicated that 688 genes overlapped between our KS lesion dataset and these in vitro models ( Fig 9A ) , with 38 overlapping genes having similar expression pattern among all 4 studies ( Fig 9B ) . The KSHV infected TIME cells showed the highest similarity to lesions with 545 overlapped genes , followed by the LEC and HMVEC cells with 232 and 124 overlapped genes , respectively ( Fig 9A ) . Some of these overlapped genes included , for example , a disintegrin and metalloprotease domain 19 ( ADAM19 ) whose expression also correlated with KSHV RNA load in the lesion . In addition , the predicted upstream regulator erythroblast transformation-specific related gene ( ERG ) and one of the downstream affected genes C-X-C motif chemokine 11 ( CXCL11 ) were also upregulated in in vitro models . Moreover , some of the overlapping genes among the studies also shared similar enriched functions , especially in glucose metabolism disorder and lipid functions ( Fig 9C and S5 Table ) . However , overlapped genes represented only ~19% of the significantly differentially expressed genes in the lesion . The disparity in the gene expression profiles is likely due to the homogeneous nature of infected cell lines , whereas the lesion represented a much more complex , but realistic environment where multiple cell types co-exist and interact with direct and indirect effects of KSHV infection on cellular transcription .
This study is the first to profile the expression of KSHV and the cellular genes in the tumor microenvironment utilizing high-throughput RNA sequencing . Our data derived from lesion and contralateral control tissue biopsies of African epidemic KS patients showed a desultory pattern of KSHV gene expression in the tumor tissue between the patients . This is likely due to the multi-cellular lineage composition of the lesions , where diverse cell types in various stages of KSHV infection co-exist . However , we did find that the most dominantly expressed KSHV genes in all the lesions were involved in the establishment and maintenance of latency or in immune modulation . The high expression of viral immune modulation genes such as K2 ( viral interleukin-6 ) , K5 ( modulator of immune recognition ) , K7 ( viral inhibitor of apoptosis ) and ORF75 ( degradation of ND10 protein ) in the lesions , could have severely hindered the immune system from eliminating these KSHV infected cells and contributed to the development of cancer . The upregulation of K2 in our data is consistent with a previous KS lesion gene expression study using real time-PCR [55] . In contrast to a prior report suggesting the potential oncogenic properties of the viral genes K1 , ORF74 and vIRF1 , we detected very few reads for these genes ( ranged from 1 to 8 reads ) in tumor tissues , suggesting either a very limited role of these genes in KS neoplastic growth or that their function is only essential at a temporally earlier stage in tumor development [56] . In addition , lytic gene expression was robust in the rapidly expanding cancer cells for two patients , p22 and p23 but restricted the other two patients . This observation is reminiscent of the limited sporadic lytic replication in various cell lines experimentally infected with KSHV , suggesting that sustained growth of KSHV transformed cells does not require robust lytic replication [57] . Interestingly , we found that patients with longer exposure to ART tended to have lower KSHV viral gene expression in their lesion and vice versa . For example , the patient p32 received ART for 60 months and had the least number of reads aligned to KSHV . The low viral gene expression also had a moderate inverse-correlation ( r = −0 . 56 ) with higher lesion KSHV DNA copy number , as determined by real time-PCR . This suggests that prolonged ART treatment might suppress KSHV viral gene expression , but result in an increase of latent viral genome as shown by a higher KSHV DNA copy number in the lesions . A potential explanation is that the immune surveillance in these HIV-1 positive individuals improved after prolonged ART , effectively suppressing KSHV gene expression . However , a larger sample size will be needed to confirm these observed relationships . Our study also revealed that expression of KSHV genes had a tremendous impact on the cellular gene expression profile . This is illustrated by the significant differences in gene expression between the lesion and control tissues that demonstrate KSHV mediated global transcriptional reprogramming in the lesion . Despite the disparity in KSHV gene expression , the cellular gene expression in the lesions was relatively similar between patients . By analyzing these common cellular gene changes with the IPA package , several upstream regulator genes were predicted to be altered in the lesions . These regulators , for example , upregulation of the transforming growth factor-beta 1 ( TGFB1 ) , could potentially impact a series of proteins and pathways involved in cell proliferation , differentiation and growth . Moreover , our analysis predicted the activation in lesions of expression of several CXCR3 chemokine ligands , CXCL-9 , CXCL-10 and CXCL-11 that are known to play an important role in the recruitment of immune cells , and in particular , T cells . Interestingly , these 3 ligands were also reported to be upregulated in classic KS lesions , suggesting a potential universal phenomenon that is primarily driven by KSHV [58] . In agreement with the prediction of upregulation of CXCR3 chemokine ligands , CIBERSORT analysis implicated the lesion contained significant infiltration of various immune cells such as B cells , macrophages and natural killer cells in all patients . However , dendritic cells infiltration was predicted only in patients p32 and p83 , whereas patients p22 and p23 had over 4-fold decrease than control tissues in apparent dendritic cells infiltration . It is worth noting that patients p32 and p83 had been on ART for an extended period of time , suggesting that ART might have stimulated some recovery of innate and adaptive anti-KSHV or anti-tumor immune response , but still was insufficient to suppress the KS development . The CIBERSORT also detected no significant infiltration of T cells in tumors , despite the apparent expression of the CXCR3 T cell chemo-attractants . This brings into question the functionality of the infiltrating immune cells . It is possible that these immune cells were able to migrate into the lesions but were incapable of activating their effector functions due to suppression by cytokines , anergy or checkpoint regulation [59] . Unfortunately , we were not able to confirm or refute these concepts using the current RNA-seq analyses alone . Transcriptomic , proteomic and functional assays on separate lineages of tumor-infiltrating lymphocytes with additional samples will need to be analyzed . From the unique depth of the cellular transcriptomic data , we were also able to infer which biological functions in the lesions were most likely to have been affected by the alteration in cellular gene expression . Intriguingly , beyond a plethora of previously enumerated cancer-related functions , we discovered profound activation of glucose metabolism disorder , coupled with significant decreases in multiple lipid anabolic and catabolic pathways . The relationships between glucose and lipid metabolism in tumor is clearly complex , nevertheless , it is widely accepted that the Warburg effect , where multiple metabolic processes are altered in the tumor to sustain the rapid proliferation and expansion of cancerous cells , is common [60] . Our finding of an increased glucose metabolism disorder in KS lesions could be an indication of the Warburg effect and can be explained by the need for the cancer cells to replicate . Several studies have shown that KSHV is indeed capable to induce Warburg effect in cell-lines and resulted in up-regulation of metabolic pathways such as glycolysis [61–64] . Moreover , KSHV latency genes have been shown to affect glucose metabolism in vitro [65] . However , the decrease in lipid metabolism activities in these KS lesions is puzzling . Previous in vitro studies have suggested that KSHV requires and activates fatty acid synthesis for survival as latently infected endothelial cells . Additionally , multiple cellular genes ( SCP2 , PRDX5 , ACSL3 , MLYCD , AGPS , EHHADH , PEX19 , PEX12 , PEX5 and ABCD3 ) involved in lipid metabolism were up-regulated in these in vitro models [47 , 66] . Increased lipid metabolism was also observed in other forms of cancer [67 , 68] . Moreover , hypoxia conditions in the core of tumor should have stimulated lipid synthesis [69 , 70] . Surprisingly , we observed a down-regulation of lipid metabolism genes in the context of KS lesions . Among the lipid metabolism genes identified in the previous studies [47 , 66] , we observed only marginal increased expression for ACSL3 , AGPS and PEX12 , the remaining genes were all down-regulated in our KS lesions [47] . Why lipid metabolism is not up-regulated , as suggested by in vitro models , in the KS lesions is unclear . Since KS lesion is a mixture of both normal and tumor cells , it is plausible that we are seeing an averaging effect with the KSHV-infected tumor cells undergoing up-regulated lipid biosynthesis , whereas the non-tumor cells were down-regulated . Interestingly , it had been reported that both ART and HIV-1 are capable of inducing lipid dysregulation , which is therefore a plausible cause for the decreased lipid metabolism observed in these epidemic KS patients [71–73] . It will be important to analyze the transcriptomes of lesions from HIV-1 negative endemic KS patients in order to differentiate the impacts due to ART , HIV-1 and KSHV alone on these metabolic changes . Given the lack of suitable animal models for KSHV pathogenesis , in vitro cultures have been the only feasible mechanism to address the effects of KSHV on the infected cells . However , in vitro systems may not recapitulate the unique and complex microenvironment within the lesions . By comparing our lesion-based transcriptomic database with several in vitro studies , we showed that the TIME model reported recently is the most similar in its cellular gene expression profile to KS tumor biopsies . For example , the downregulation of peroxisome proliferator activated receptor gamma ( PPARG ) that regulates glucose metabolism and fatty acid storage was observed in both studies . Nevertheless , only 19% overlap between the lesion transcriptomes and in vitro derived expression repertoires was detected . More than 2901 genes were uniquely expressed in the lesions but not observed in the KSHV infected cell line models , suggesting that one needs to be cautious in extrapolating the gene expression data from cell lines to the human disease . Finally , although the ipsilateral /contralateral control tissue collected from each KS patient served as an ideal intra-patient comparator for the lesion , the transcriptomes from these control tissues may not represent the gene expression profile of a KSHV negative patient . It is possible that the gene expression of these control tissues was already indirectly or directly altered by KSHV or by chronic immune responses . Skin tissue from KSHV negative patients will need to be tested to address this issue . Despite this shortfall , our data provide unique initial insights into the close relationship between the cellular and KSHV genes in tumorigenesis , as well as the profound impact of KSHV on numerous cellular regulatory and metabolic pathways that could potentially be targeted for future treatment of KS patients .
|
Kaposi’s sarcoma-associated herpesvirus ( KSHV ) is endemic in sub-Saharan Africa and cause Kaposi’s sarcoma ( KS ) . KS is one of the most common cancer among HIV-1 patients in this region . Despite anti-retroviral treatment , prognosis for KS is poor with high mortality often due to presentation of late cancer stage . In order to identify biomarkers or therapeutic targets against KS , a better understanding of the viral and cellular genes expression/transcriptome in the tumor will be necessary . We used RNA-seq , a highly efficient method to sequence transcriptome , to characterize and compare the viral and cellular transcriptome in tumor and non-cancer tissues from KS patients . We found that viral genes involved in latency and immune modulation are most commonly expressed among KS patients . Additionally , cellular genes involved in lipid and glucose metabolism disorder pathways are significantly affected by the presence of KSHV . Despite the detection of immune cells in the tumor , it did not prevent the tumor progression , suggesting some level of immune cell dysfunctions in KS patients . Lastly , we found limited overlap of our data , derived from actual KS biopsy , with other cell culture models , suggesting that the complexity of tumor is difficult to be reflected in cell line models .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"carbohydrate",
"metabolism",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"cancers",
"and",
"neoplasms",
"microbiology",
"glucose",
"metabolism",
"retroviruses",
"oncology",
"immunodeficiency",
"viruses",
"viruses",
"regulator",
"genes",
"signs",
"and",
"symptoms",
"rna",
"viruses",
"genome",
"analysis",
"gene",
"types",
"lipid",
"metabolism",
"lipids",
"genomics",
"medical",
"microbiology",
"hiv",
"gene",
"expression",
"lesions",
"microbial",
"pathogens",
"kaposi",
"sarcoma",
"hiv-1",
"sarcomas",
"biochemistry",
"diagnostic",
"medicine",
"viral",
"pathogens",
"genetics",
"transcriptome",
"analysis",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"lentivirus",
"metabolism",
"organisms"
] |
2018
|
RNA-Seq of Kaposi’s sarcoma reveals alterations in glucose and lipid metabolism
|
Metronidazole , a 5-nitroimidazole drug , has been the gold standard for several decades in the treatment of infections with microaerophilic protist parasites , including Entamoeba histolytica . For activation , the drug must be chemically reduced , but little is known about the targets of the active metabolites . Applying two-dimensional gel electrophoresis and mass spectrometry , we searched for protein targets in E . histolytica . Of all proteins visualized , only five were found to form adducts with metronidazole metabolites: thioredoxin , thioredoxin reductase , superoxide dismutase , purine nucleoside phosphorylase , and a previously unknown protein . Recombinant thioredoxin reductase carrying the modification displayed reduced enzymatic activity . In treated cells , essential non-protein thiols such as free cysteine were also affected by covalent adduct formation , their levels being drastically reduced . Accordingly , addition of cysteine allowed E . histolytica to survive in the presence of otherwise lethal metronidazole concentrations and reduced protein adduct formation . Finally , we discovered that thioredoxin reductase reduces metronidazole and other nitro compounds , suggesting a new model of metronidazole activation in E . histolytica with a central role for thioredoxin reductase . By reducing metronidazole , the enzyme renders itself and associated thiol-containing proteins vulnerable to adduct formation . Because thioredoxin reductase is a ubiquitous enzyme , similar processes could occur in other eukaryotic or prokaryotic organisms .
Entamoeba histolytica is a microaerophilic protozoan parasite and the causative agent of amoebiasis , a disease that affects millions of people worldwide and claims up to 100 , 000 casualties per annum [1] . As is the case with other microaerophilic parasitic infections , such as giardiasis ( caused by Giardia intestinalis ) and trichomoniasis ( caused by Trichomonas vaginalis ) , the 5-nitroimidazole drug metronidazole has established itself as the most effective treatment of amoebiasis . Due to the high prevalence of these infections [2] and due to its role as a second-line defense against Helicobacter pylori infections [3] , metronidazole has been included in the “essential medicines” list by the World Health Organization [4] . Metronidazole , like other nitroimidazoles , requires reduction at the nitro group in order to be transformed into its cytotoxic form , the nitroradical anion [5] . The activated nitro group undergoes further reduction so that a nitrosoimidazole is generated [6] which can react with sulfhydryl groups [7] and with DNA [8] while being further reduced to an amine via a hydroxylamine intermediate . In the presence of oxygen , however , the nitroradical anion is suggested to be rapidly reoxidized to its respective parent drug before nitroso intermediates can be formed , i . e . , a redox cycling effect also termed “futile cycle” [9] . Despite the resulting oxidative stress , this futile cycle is believed to render metronidazole treatment safe in man . However , there are still concerns regarding its potential carcinogenicity [10] . Since reduction of the nitro group is essential for nitroimidazole toxicity , extensive research has been dedicated to enzymes that can act as metronidazole-activating nitroreductases . In rat liver extracts , the microsomal enzyme NADPH-cytochrome P450 reductase was found to be responsible for nitroimidazole activation [11] . The microaerophilic parasites G . intestinalis , T . vaginalis , and E . histolytica , however , lack mitochondria [12] but depend on substrate-level phosphorylation [13] . In these organisms , ferredoxin , which is being reduced by pyruvate:ferredoxin oxidoreductase ( PFOR ) , has been suggested to activate metronidazole [14] . Indeed , in T . vaginalis , metronidazole activation was found to take place in the hydrogenosome [15] , a hydrogen-producing organelle in which PFOR and ferredoxin are localized [16] . Moreover , purified ferredoxin was shown to be able to reduce various nitroimidazoles in vitro [17]; and in some highly metronidazole-resistant laboratory T . vaginalis strains , PFOR and ferredoxin were absent , stressing a direct relationship between ferredoxin and metronidazole activation in vivo [18] . Likewise , PFOR activity [19 , 20] and ferredoxin levels were reduced in metronidazole-resistant G . intestinalis strains [21] . In partially metronidazole-resistant E . histolytica , expression of ferredoxin 1 was sharply decreased [22] although PFOR levels remained unaltered [23] . In contrast to T . vaginalis [24] , metronidazole resistance in E . histolytica could be mainly attributed to the increased expression of the antioxidant enzymes peroxiredoxin [22] and superoxide dismutase [23] , rather than to loss of PFOR activity , as observed in the other two parasites . For several decades , great efforts have been undertaken to deepen the understanding of metronidazole activation in the parasitic cell , but the exact mode of action in vivo of this pivotal drug has remained rather understudied . DNA is suggested to be the major target of metronidazole [14 , 25] , as implied by several in vitro studies addressing metronidazole's mutagenicity and DNA-binding capability [8] . In addition , in vitro adduct formation of nitroimidazoles with proteins and thiols , e . g . , cysteine , was also demonstrated [11] , but specific targets in the treated parasites were never defined because nitroimidazole action was assumed to be indiscriminate . As a contribution to fill this gap , it was our goal to elucidate the processes that occur in the E . histolytica cell during metronidazole treatment at concentrations that are applied during the treatment of amoebiasis , and , if existent , identify specific targets of metronidazole . After completion of the E . histolytica genome project [26] , application of proteomic methods such as two-dimensional gel electrophoresis ( 2DE ) was greatly facilitated , permitting a comprehensive and rapid identification of proteins affected by metronidazole in treated E . histolytica cells . In this study , we show that , in E . histolytica , activated metronidazole does not bind to protein indiscriminately , but reproducibly forms covalent adducts with a small and defined number of proteins , including enzymes such as thioredoxin reductase , superoxide dismutase , and purine nucleoside phosphorylase , as well as the multiple-role reductant protein thioredoxin . When recombinantly expressed in Escherichia coli BL21 ( DE3 ) in the presence of metronidazole , the capability of thioredoxin reductase to reduce thioredoxin was significantly diminished . Moreover , levels of non-protein thiols , e . g . , cysteine , were found to be drastically lowered in metronidazole-treated E . histolytica cells due to adduct formation between activated metronidazole and accessible sulfhydryl groups . In accordance with this finding , addition of cysteine to the growth medium allowed the cells to survive otherwise lethal metronidazole concentrations and significantly reduced protein adduct formation . Finally , we propose an alternative mode of metronidazole activation by thioredoxin reductase , because it showed nitroreductase activity in enzymatic assays .
After having treated E . histolytica trophozoites for different time periods ( 1 h , 2 h , 3 h , and 6 h ) and with varying metronidazole concentrations ( 10 μM–1 mM ) , cell lysates were prepared for 2DE experiments . Metronidazole concentrations between 50 μM and 100 μM proved to be the most suitable because cells were viable for more than 5 h , a time span which is sufficient for the cell to react to stress by expression of mRNA and proteins . In addition , therapeutic levels lie within this range . Because higher metronidazole concentrations led to rapid disintegration of the cells , and incubation periods for more than 2 h with 50 μM metronidazole did not reveal any additional changes in the protein profile , we chose exposure to 50 μM metronidazole for 2 h as our standard condition when challenging cells with metronidazole . We reproducibly found seven new protein spots on the gels that were isolated and analyzed by mass spectrometric tryptic peptide fingerprinting in combination with additional verification of selected peptide sequences by tandem mass spectrometry ( MS/MS ) ( identified peptides and tandem mass spectra are listed in Figures S1–S5 ) . These seven spots corresponded to five proteins ( Figure 1A and 1B ) , identified as superoxide dismutase , purine nucleoside phosphorylase , thioredoxin , thioredoxin reductase , and a protein designated as “hypothetical protein XP_650662” in the E . histolytica protein database ( Table 1 ) . The last will further be referred to as “metronidazole target protein 1” ( Mtp1 ) , since its presence on 2D gels abolishes its hypothetical status . Surprisingly , the new spots did not correspond to newly synthesized protein , but appeared at the expense of other neighboring spots as shifted isoforms at a more basic isoelectric point ( pI ) ( Figure 1A and 1B ) . The widths of the shifts in pI differed with each of the proteins . Thus , it was hypothesized that metronidazole exposure leads to modification of these five proteins , and that the newly appearing spots correspond to isoforms of pre-existing protein in the cell . Concomitant treatment of the cells with 100 μM cycloheximide to block protein synthesis did not prevent the appearance of the shifts on the gel when cells were exposed to metronidazole ( unpublished data ) . This supported our notion that the observed spots do not correspond to newly synthesized protein . Moreover , we did not find any proteins significantly up-regulated or down-regulated in expression during metronidazole exposure; mRNA expression was also not found to be altered ( M . Tazreiter , unpublished data ) . Thus , presumably , E . histolytica does not react to short-term metronidazole treatment by mRNA synthesis or by synthesis of proteins that are involved in stress response or antioxidant defense . This indicated that the protein shifts might be due to metronidazole adduct formation with the five proteins rather than to a general stress response of the cell . Unfortunately , mass spectrometric analysis of tryptic fragments did not reveal any metronidazole-bound peptides . Thus , in order to confirm metronidazole binding to the proteins , we treated the cells with other nitroimidazoles , including the 2-nitroimidazole azomycin , and the two 5-nitroimidazoles tinidazole and ornidazole ( Figure 2A ) . Since cross-resistances were reported for most of the nitroimidazoles [2] , we expected all nitroimidazoles to give similar results as compared to metronidazole . Again , treatment regimens with 50 μM of the respective nitroimidazole for 2 h were chosen as the experimental conditions . Indeed , two-dimensional ( 2D ) gels revealed that the same proteins were affected , but the width of the pI shifts to the basic differed according to the varying pKa's of the nitroimidazoles , which can be attributed to the different side chains at the N1 position . Protein shifts upon ornidazole treatment were slightly narrower , whereas protein shifts with tinidazole were considerably narrower than those observed with metronidazole ( Figure 2B ) . Interestingly , the 2-nitroimidazole azomycin also shifted the same proteins . With regard to pI interval , the shifts by azomycin were wider than shifts by the other nitroimidazoles tested , but the amount of the respective proteins shifted was smaller . The modification of proteins seemed to be independent of the position of the nitro group in the ring , indicating a generalized pattern of nitroimidazole action in E . histolytica . Quantitative analysis with Melanie 2DE imaging software indicated that , even after prolonged incubation , only a defined fraction of each protein was shifted until a certain maximum was reached ( Table 1 ) . Thus , higher nitroimidazole concentrations only allowed this maximum to be more rapidly attained ( unpublished data ) . The proteins were also modified to the same extent in partially metronidazole-resistant E . histolytica cells that had been continuously cultured in the presence of 10 μM metronidazole ( unpublished data ) . Like others before [22 , 23] , we were not able to obtain highly metronidazole-resistant E . histolytica strains . Such would have been very helpful for an assessment of the impact of the modifications on metronidazole-mediated toxicity . Because nitroimidazoles have been found to form adducts with sulfhydryl group containing compounds [7] , we tested whether metronidazole treatment could reduce the levels of non-protein thiols in the cell , e . g . , cysteine , which constitutes the major reductant in E . histolytica [27 , 28] . Treatment with 50 μM metronidazole for 2 h decreased non-protein thiol levels to 49% ( Figure 3A ) of those in untreated cells ( 11 fmol/cell ) , whereas additional 50 mM cysteine ( 17 mM cysteine constitutes the standard concentration in TYI-S-33 medium ) , led to accumulation of cysteine in the cell and raised non-protein thiol levels by about 180% , to approximately 31 fmol/cell . When metronidazole and cysteine were used in combination , thiol levels were also sharply decreased ( 54% of the untreated state ) . Since cysteine could not accumulate intracellularly in the presence of metronidazole ( the observed 5% difference between total free-thiol levels of cells treated with metronidazole alone and those treated with metronidazole in combination with cysteine , is statistically not significant ) , these results indicate that cysteine levels in the cell are diminished by metronidazole . As the observed decrease in non-protein thiol levels could also be the consequence of oxidative stress , e . g . , oxidation of sulfhydryl groups by hydrogen peroxide , we attempted to verify covalent adduct formation of metronidazole with sulfhydryl groups by exposing the cells to 50 μM metronidazole under anaerobic ( Merck Anaerocult A , 0% O2 and 18% CO2 ) , microaerophilic ( Merck Anaerocult C , 5% O2 and 8% CO2 ) , and aerobic conditions ( aerated vials ) , respectively . We reasoned that , if the decrease in thiol levels were to be attributed to oxidative stress , exposure of the cells to metronidazole under aerobic conditions would result in even more strongly reduced thiol levels , whereas , under anaerobic conditions , the observed effect should be significantly mitigated or even absent . However , our experiment showed that the reduction of thiol levels , to 36% of the level of untreated cells , was the most strongly pronounced after metronidazole exposure under anaerobic conditions ( Figure 3B ) , followed by exposure to metronidazole under microaerophilic conditions , with 51% , and then under aerobic conditions , with 54% . Thus , oxygen did not promote , but conversely , counteracted the reduction of thiol levels upon metronidazole exposure , which strongly suggests that covalent adduct formation of activated metronidazole with thiol groups is the reason for the decrease in non-protein thiol levels . Moreover , these results are in line with the finding that oxygen can detoxify nitroradical anions and regenerate the parent drug by snatching the electron from the nitro group [9] . However , this effect was not as pronounced as anticipated , because reoxidation of the metronidazole radical anion by oxygen was incomplete , leading to almost halved non-protein thiol levels , even under aerobic conditions . Interestingly , adduct formation with the five proteins found was not affected in the presence of higher oxygen concentrations when cells were treated with 50 μM metronidazole for 2 h ( unpublished data ) . In order to evaluate the extent to which the observed decrease in non-protein thiol levels contributes to metronidazole toxicity , we treated cells with azomycin , which had been found to be by far less toxic to E . histolytica than metronidazole ( D . Leitsch , unpublished data ) . Indeed , non-protein thiol levels were found to be affected by azomycin , even if considerably less than by metronidazole , paralleling the observation that azomycin shifts the same proteins as metronidazole , albeit to a smaller extent . Under anaerobic conditions , a drop to 65% of the original level when treating cells with 50 μM azomycin for 2 h , as compared to a drop to 36% when treating the cells with metronidazole , was observed ( Figure 3B ) . In the presence of oxygen , non-protein thiol levels after exposure to azomycin were only slightly reduced: 84% ( microaerophilic ) and 82% ( aerobic ) , respectively , of the non-protein thiol levels were retained as compared to untreated cells . Interestingly , higher oxygen levels counteracted adduct formation of azomycin with non-protein thiol groups more strongly than was the case with metronidazole . As compared to anaerobic conditions , 49% less adduct formation with azomycin ( a drop of non-protein thiols of only 18%—to 82% of the untreated sample—instead of a drop of 35%—to 65% ) and 28% less adduct formation with metronidazole ( a drop of 46%—to 54% of the untreated sample—instead of a drop of 64%—to 36% ) were observed when treating cells under aerophilic conditions . Differences between microaerophilic and anaerobic conditions with regard to their impact on the decrease of non-protein thiol levels in the presence of metronidazole or azomycin were hardly significant , if existent . Although non-protein thiol levels were only insignificantly higher in cells that were treated with metronidazole and cysteine in combination , as compared to non-protein thiol levels in cells treated with metronidazole alone , we observed that the cells were not rounding off and disintegrating in the presence of higher cysteine levels . Therefore , we tested whether raised cysteine levels in the medium could protect E . histolytica during metronidazole treatment . Cells were treated with 30 μM or 50 μM metronidazole either in the presence or absence of additional 50 mM cysteine ( 67 mM in total ) ( Figure 3C ) . Addition of 50 mM cysteine slightly impaired viability; only 83% of the cells were still viable after 20 h of incubation . In the cultures treated with 30 μM metronidazole alone , only 23% of the cells were still viable after the same time span , whereas 50 μM of metronidazole was sufficient to kill almost all cells in the culture ( 2% viable cells ) . However , 86% of the cells in the culture treated with 30 μM metronidazole and 71% of the cells treated with 50 μM metronidazole were still viable after 20 h when 67 mM cysteine was present in the growth medium . These results clearly indicate that cysteine strongly counteracts metronidazole toxicity in E . histolytica . We speculated that the protective effect of cysteine might be due to less metronidazole adduct formation with proteins , because we expected free cysteine to compete with proteins for activated metronidazole . Thus , we assessed the influence of raised cysteine levels on adduct formation of metronidazole with the five proteins identified . Additional 50 mM cysteine markedly reduced metronidazole adduct formation with protein when cells were treated with 50 μM metronidazole for 2 h ( Figure 3D ) . The shifts in pI of superoxide dismutase , thioredoxin , and Mtp1 were clearly diminished in this case , whereas thioredoxin reductase was only subject to putatively one modification and purine nucleoside phosphorylase remained unmodified . The latter two observations were not always made when repeating the experiment , but a very distinct decrease of adduct formation with all five proteins was perfectly reproducible . These results suggest that non-protein cysteine competes with proteins in the formation of adducts with activated metronidazole , and that the observed protective effect of cysteine during metronidazole exposure might be due to fewer protein adducts formed . As a result of our observation that metronidazole adducts are also formed in Es . coli under microaerophilic conditions , albeit at much higher metronidazole concentrations than is the case in E . histolytica ( unpublished data ) , we decided to recombinantly express E . histolytica thioredoxin reductase and E . histolytica thioredoxin in Es . coli either in the presence or in the absence of metronidazole . We speculated that this strategy would allow us to obtain a large quantity of metronidazole-bound protein that could aid in the mass spectrometric identification of the observed modifications . In addition , the influence of metronidazole adduct formation on protein function could be assayed in vitro with the modified and unmodified recombinant proteins at hand . As thioredoxin and thioredoxin reductase are of central importance for the cell's physiology [29] , these two proteins were of particular interest to us . Moreover , human [30] and Arabidopsis thaliana [31] thioredoxin reductases were shown , in addition to their intrinsic disulfide reductase activity , to reduce nitro compounds such as tetryl or 1-chloro-2 , 4-dinitrobenzene ( CDNB ) . Therefore , we also wanted to determine in vitro whether E . histolytica thioredoxin reductase can , in addition to its role as a disulfide reductase , reduce nitroimidazoles . Recombinant E . histolytica thioredoxin reductase ( recEh TrxR ) and recombinant E . histolytica thioredoxin ( recEh Trx ) were produced in Es . coli BL21 ( DE3 ) cells and purified on Ni-NTA columns via their carboxy-terminal hexahistidine tags . RecEh TrxR had a deep yellowish color that can be attributed to its FAD or FMN cofactor [32] . When Es . coli BL21 ( DE3 ) cells were treated with 1 mM metronidazole during recEh TrxR and recEh Trx expression , the recombinant proteins were efficiently modified by metronidazole as verified by 2DE ( unpublished data ) . In contrast to our failed attempts to identify metronidazole on peptides that were directly isolated from 2D gels , we observed , by liquid chromatography electrospray ionization quadrupole time-of-flight tandem mass spectrometry ( LC-ESI-QTOF-MS ) of intact proteins , that a high proportion of recEh TrxR and recEh Trx displayed a shift in molecular mass ( Figure 4A ) . The shift in the deconvoluted spectra of both recEh Trx and recEh TrxR ( but here only illustrated with recEh Trx ) , corresponded to a mass gain of 141 Da . This is in good agreement with the in vitro reaction scheme for 5-nitroimidazoles as proposed by Wislocki and colleagues [33] ( Figure 4B ) . In this scheme , the activated 5-nitroimidazole is first reduced to an electrophilic nitrosoimidazole , which is subsequently attacked at its C4 atom by a sulfhydryl group . This is accompanied by further reduction to a hydroxylamine group , which is finally reduced to an amino group . Thus , the nitroimidazole loses two oxygen atoms from the nitro group and one proton from C4 , and gains two hydrogens , resulting in a decrease in mass of 31 Da . At physiological pH and under the conditions applied during LC-ESI-QTOF-MS , however , the amino group is protonated , leading to a total mass decrease of only 30 Da . Since metronidazole has a mass of 171 . 16 Da , a protein that binds metronidazole can be expected to increase in mass by about 140 Da because it gains 141 Da from the bound 5-aminoimidazole derivate of metronidazole but loses the proton of a sulfhydryl group . When allowing for the methodological restraints of the LC-ESI-QTOF-MS instrument , which has a mass deviation of ±4 Da when analyzing a protein of the size of recEH Trx , the calculated theoretical mass gain of 140 Da after modification by activated metronidazole corresponds well with the observed shift of 141 Da . In order to check whether the proposed model also applies for other 5-nitroimidazoles than metronidazole , we expressed recEh TrxR and recEh Trx in presence of tinidazole , which has a molecular mass of 247 . 3 Da . The result obtained with metronidazole was paralleled by that with tinidazole ( Figure 4A ) ; the shifts in the deconvoluted spectra amounted to 217 Da ( molecular weight of the nitroimidazole less 30 Da ) . Moreover , it is important to add that the proposed model for nitroimidazole binding is also supported by the fact that , on 2D gels , the pI values of all five proteins identified were shifted to the basic , probably due to the reduction of the nitro group of the nitroimidazoles to a basic amino group . Additional peaks that can be observed on the mass spectra of metronidazole- and tinidazole-bound recEh Trx correspond to mass increments of approximately 16 Da or 32 Da , and very likely can be attributed to oxidation ( one or two oxygen atoms , respectively ) . Not unexpectedly , reactive oxygen species that are generated during nitroimidazole treatment led to oxidation of proteins ( e . g . , recEh Trx ) and , probably , other cell constituents . Unfortunately , our attempts so far to pinpoint the modifications to specific tryptic peptide fragments from recEh Trx and recEh TrxR have been as unsuccessful as had been our attempts to identify the modifications on E . histolytica proteins directly isolated from 2D gels . Obviously , the nitroimidazole adducts were not stable under the experimental conditions applied , possibly because incubation periods with trypsin ( overnight at 37 °C ) were too long or the conditions during LC-ESI-QTOF-MS were too harsh . We will , therefore , intensify our efforts in the future and modify the standard protocols accordingly . Thioredoxin reductase activity of recEh TrxR ( applied at a concentration of 148 nM ) was verified ( Figure 5 ) by reduction of the disulfide 5 , 5′-dithiobis- ( 2-nitrobenzoic acid ) ( DTNB ) to 2-nitro-5-thiobenzoic acid ( TNB ) via recEh Trx ( applied at a concentration of 174 nM ) . Specific reduction of recEh Trx was determined by subtracting the ground-level reduction of DTNB ( 206 nmol min−1 mg−1 ) by recEh TrxR and was found to amount to 559 nmol min−1 mg −1 ( which equals a turnover of approximately 23 . 5 min−1 ) . RecEh TrxR and its substrate , recEh Trx , were used in roughly equimolar amounts because our 2D gels suggested that both proteins are about equally abundant in the cell . Because we wanted to stay as close to physiological conditions as possible , we did not apply such a high excess of the substrate to the enzyme as would be necessary for the exact determination of the kinetic constants of recEh TrxR . Nevertheless , the activity of recEh TrxR , determined by us , is in good accordance with the thioredoxin reducing activity of E . histolytica thioredoxin reductase as determined just recently by Arias and colleagues [34] . We used metronidazole-modified recEh TrxR and recEh Trx for an estimation of the influence of metronidazole on protein function . Thioredoxin reducing activity of metronidazole-modified recEh TrxR dropped by more than 50% to 265 nmol min−1 mg−1 ( Figure 5 ) , and when used in combination with metronidazole-modified recEh Trx , the efficiency of recEh Trx reduction by recEh TrxR was even further diminished ( 220 nmol min−1 mg−1 ) . When assaying metronidazole-modified recEh Trx alone , disulfide reduction lay also clearly below ( 427 nmol min−1 mg−1 ) the reduction rate as compared to using unmodified forms of both recEh TrxR and recEh Trx ( 559 nmol min−1 mg−1 ) . Using a recEH TrxR concentration of 118 nM ( equal to 4 μg/ml ) , CDNB reduction was determined by measuring NADPH consumption at 340 nm ( Table 2 ) . Nitroreductase activity amounted to 233 nmol min−1 mg−1 ( 7 . 8 reduction events per minute per molecule of recEh TrxR ) when using CDNB as the substrate at a concentration of 100 μM . Unfortunately , the assay was heavily disturbed by the absorbances of the nitroimidazoles . As an alternative method ( Table 2 ) , nitroreductase activity of recEh TrxR was indirectly determined by measuring reduction of cytochrome c by reduced nitro compounds [30] , or by superoxide radical anions generated by the transfer of an electron from nitroradical anions to oxygen , respectively . In both cases , reduction of cytochrome c could be attributed to the previous reduction of the assayed nitro compounds , i . e . , CDNB , the 2-nitroimidazole azomycin , and the 5-nitroimidazole metronidazole , by recEh TrxR ( 4 μg/ml ) . As expected , CDNB was readily reduced at a rate of 171 nmol−1 mg−1 at concentrations as low as 10 μM ( Table 2 ) . This rate lies in the range of the determined nitroreductase activity of recEh TrxR when measuring NADPH consumption . The 36% higher reduction of CDNB in the first assay is likely due to the 10-fold higher CDNB concentration used . The 2-nitroimidazole azomycin was also readily reduced at a concentration of 100 μM ( 63 nmol min−1 mg−1 ) . The reduction rate of metronidazole ( 1 mM ) amounted to 31 nmol min−1 mg−1 . All values given above have been corrected for the ground-level activity of recEh TrxR , i . e . , the reduction of molecular oxygen in the assay buffer , resulting in the formation of superoxide radical anions that , in turn , can reduce cytochrome c . In the absence of recEh TrxR , no reduction of cytochrome c was observed . In contrast to thioredoxin reductase activity , nitroreductase activity was not impaired with metronidazole-bound recEh Trx ( unpublished data ) . This is possibly due to the fact that the flavin cofactor rather than the enzymatic site of recEh TrxR is responsible for nitroreduction [31] . In order to distinguish between direct reduction of cytochrome c by nitroradical anions and between reduction of cytochrome c by superoxide radical anions that had previously been formed by the transfer of the electron of the nitroradical anion to molecular oxygen , superoxide dismutase was added to the reactions in about 20-fold excess ( 2 . 5 μM ) ( Table 2 ) . The addition of superoxide dismutase completely abolished ground-level reduction of cytochrome c by EhTrxR and decreased cytochrome c reduction in the presence of 10 μM CDNB to 47% of the original value . Cytochrome c reduction in the presence of 100 μM azomycin was diminished to 53% , whereas 75% of the original value was retained with 1 mM metronidazole . Higher concentrations of superoxide dismutase did not lead to further decreases in cytochrome c reduction . Thus , in the case of CDNB and azomycin , roughly half of the cytochrome c reduction is mediated by superoxide radical anions , whereas the metronidazole nitroradical anions directly transferred most of the electrons to cytochrome c . These results provide direct evidence for the formation of superoxide radical anions upon nitroimidazole treatment in general and within limits upon metronidazole treatment , and thereby for the generation of oxidative stress in the microaerophilic cell .
In this study , we show for the first time that metronidazole forms adducts with proteins and non-protein thiols in an in vivo model , i . e . , a parasite that is commonly treated with this drug . The shifts in masses , found on the mass spectra of metronidazole-bound recEh Trx and recEh TrxR , confirmed the in vitro model for 5-nitroimidazole adduct formation by Wislocki and colleagues [33] ( Figure 4B ) . In contrast to the in vitro data from the late 1970s and early 1980s , however , we found discrete changes in the protein profile of E . histolytica after metronidazole treatment , i . e . , modification of five proteins , rather than indiscriminate protein adduct formation . It is possible that there are some more proteins affected , because 2DE does not cover the whole proteome of a given organism . Very large proteins , low-abundance proteins such as transcription factors , and highly hydrophobic membrane proteins that could also potentially form adducts with nitroimidazoles cannot be identified by our approach . Nevertheless , even if allowing for these restraints , the number of affected proteins can be expected to remain small . Because we made similar observations with Entamoeba dispar ( the nonpathogenic relative of E . histolytica ) , G . intestinalis , T . vaginalis , and Es . coli ( unpublished data ) , we suggest that adduct formation with a defined subset of proteins can take place in any organism that is treated with nitroimidazoles ( unpublished data ) : the presumed reduction of the nitro group to an amino group during adduct formation with the proteins leads to easily discernable shifts on 2D gels to more basic pI values . It is , therefore , interesting to speculate that nitroimidazoles could be an invaluable tool in proteomics , because our data suggest that they allow identification of nitroreductases and associated proteins by shifting them to more basic pI values that can easily be detected on 2D gels . Apart from forming covalent adducts with proteins , metronidazole also diminishes non-protein thiol levels in the cell ( Figure 3A ) , including that of cysteine . The observed decrease in non-protein thiol levels is due to covalent adduct formation with metronidazole and not due to oxidative stress generated by the activated drug , because the drop in thiol levels was most pronounced in the total absence of oxygen and least pronounced under aerobic conditions ( Figure 3B ) . Since cysteine is a compound of essential importance to E . histolytica cell physiology [35] , and because it is assumed to function as the major reductant in the cell [27] , its depletion could contribute to metronidazole toxicity in E . histolytica . On the other hand , cysteine could also predominantly have a protective role because other essential thiols in the cell , such as coenzyme A , might also form adducts with metronidazole . Interestingly , after 2 h of incubation , non-protein thiol levels in cells treated with 50 μM metronidazole and 50 mM cysteine were almost diminished to the same extent as non-protein thiol levels in cells treated with metronidazole alone ( Figure 3A ) . Longer incubation periods with additional 50 mM cysteine in the growth medium , however , might lead to consumption and , consequently , detoxification of metronidazole . This is indicated by the observation that the toxic effect of metronidazole was drastically reduced after addition of 50 mM cysteine , because more than 70% of the cells in a culture survived 20-h exposure to ( otherwise lethal ) 50 μM metronidazole ( Figure 3C ) . In addition , raised cysteine levels also led to less protein adduct formation after metronidazole treatment ( Figure 3D ) , which indicates the interdependence of non-protein and protein sulfhydryl groups with regard to metronidazole toxicity and which could be an explanation for the protective effect of cysteine during metronidazole exposure . Treatment of E . histolytica cells with the clearly less toxic 2-nitroimidazole azomycin also led to the modification of the five proteins found ( Figure 2B ) and a decrease in non-protein thiol levels ( Figure 3B ) , albeit to a far smaller extent . Interestingly , the capability of azomycin to diminish non-protein thiol levels in the cell was also much more decreased by oxygen than was the case with metronidazole . Thus , the lower toxicity of azomycin could be based on its reduced tendency to form adducts with sulfhydryl groups and on its higher reactivity with oxygen ( Table 2 ) . A potentially detrimental effect of metronidazole binding to protein function is indicated by the significant decrease of the thioredoxin reductase activity of recEh TrxR after metronidazole treatment ( Figure 5 ) . However , it is also conceivable that the observed oxidation of several amino acids concomitant with nitroimidazole binding , as observed with recEh Trx ( Figure 4A ) , contributes to a diminished enzymatic function . E . histolytica thioredoxin reductase was found to be a nitroreductase that is able in vitro to reduce CDNB , as well as the nitroimidazoles azomycin and metronidazole ( Table 2 ) . In contrast to thioredoxin reductase activity of recEh TrxR , nitroreductase activity of metronidazole-bound recEh TrxR was not decreased ( unpublished data ) , suggesting that nitroreduction is directly exerted by the FAD or FMN cofactor [31 , 32] . Azomycin ( 2 . 4 min−1 at a concentration of 100 μM ) was more effectively reduced in the nitroreductase assay than metronidazole ( 1 . 2 min−1 at a concentration of 1 mM ) , possibly due to the higher redox potential of azomycin ( E17 = −418 mV ) as compared to metronidazole ( E17 = −486 mV ) [36] . Interestingly , CDNB , which was the compound tested to be most efficiently reduced ( 6 min−1 at a concentration of 10 μM ) by recEh TrxR , exerted only a mildly toxic effect on E . histolytica in our experimental setting—arguably because it does not form adducts with the same proteins as observed with nitroimidazoles and because it does not lead to a reduction of non-protein thiol levels in the cell ( D , Leitsch , unpublished data ) . It is also possible , however , that CDNB does not enter the cell as readily as metronidazole . When superoxide dismutase was added to the reactions , cytochrome c reduction rates in the presence of CDNB and azomycin were approximately halved ( 47% and 53% , respectively ) , whereas in the presence of metronidazole , 75% of the original rate was retained . These findings provide direct evidence for the generation of oxidative stress upon nitroimidazole treatment , because superoxide radical formation is evident . However , reoxidation of metronidazole by oxygen , with only 25% of the metronidazole nitroradical anions reoxidized , was by no means as complete and as rapid as anticipated [9] . Thus , it is doubtful whether the futile-cycle effect is really as influential on metronidazole toxicity as has been suggested . Although the nitroreductase activity of recEH TrxR is rather low , it is comparable to the nitroreductase activities determined for A . thaliana and mammalian thioredoxin reductases [30 , 31] . According to our quantitative evaluations of 2D gels from E . histolytica cell extracts , thioredoxin reductase amounts to approximately 0 . 2% of the total protein in the cell , equaling 10–20 million copies per cell . Our estimate of the concentration of non-protein sulfhydryl groups amounts to 11 fmol/cell , i . e . , only a 400-fold excess of non-protein sulfhydryl group levels over those of thioredoxin reductase . It is therefore conceivable that nitroimidazole reduction by thioredoxin reductase plays an important role in the decrease of non-protein thiol concentrations in the treated cell . In this context , it is interesting to note that studies in G . intestinalis have shown that the turnover of metronidazole reduction by purified ferredoxin was also not very pronounced ( 4 min−1 ) , when purified PFOR instead of cell extract was used as the electron donor for ferredoxin [20] . We suggest that , due to spatial proximity to the reactive nitroimidazole species generated , reduction of nitroimidazoles by thioredoxin reductase renders this enzyme vulnerable to nitroimidazole adduct formation . Thioredoxin , superoxide dismutase , and Mtp1 , in turn , can be expected to be localized in proximity to thioredoxin reductase or to interact with thioredoxin reductase . This could render these proteins prone to nitroimidazole modification as well . Proteins that do not interact with thioredoxin reductase are likely to be less affected , because activated nitroimidazoles react with non-protein thiols or other compounds before they can react with proteins that are more distant to the site of nitroimidazole reduction . Thioredoxin needs to be reduced by thioredoxin reductase in order to fulfill its multiple purpose as a reductant protein [29] , whereas superoxide dismutase could be required to remove superoxide anion radicals that are indirectly generated by the nitroreductase activity of thioredoxin reductase . Superoxide dismutase might minimize the damage caused by superoxide when being positioned near thioredoxin reductase . Metronidazole target protein 1 ( Mtp1 ) has no close homolog in any other organism whose genome has been sequenced so far , but it contains an O-glycosyl hydrolase domain and displays extended homology to an α-amylase in E . histolytica . Recent research suggests that thioredoxins are also of decisive importance for starch degradation in plants . Very strong evidence has been presented for thioredoxin-mediated regulation of α-amylases in barley grain [37] . Reduction of intramolecular disulfide bonds in amylases renders these enzymes more soluble , which is a prerequisite for amylase activity . Since Mtp1 , as a protein of about 14 kDa , contains as many as six cysteines , it is conceivable that it requires thioredoxin in order to be functional . In the case of purine nucleoside phosphorylase , the situation could be different because arsenate reductase activity has been observed with human purine nucleoside phosphorylase [38] . This gives reason to speculate about the potential nitroreductase activity of the corresponding enzyme in E . histolytica . Possibly , purine nucleoside phosphorylase binds to the imidazole ring of nitroimidazole compounds , because purines have an imidazole moiety . However , in the same study , the authors found arsenate reductase activity to strongly depend on reductants , especially DTT , which suggests that purine nucleoside phosphorylase could require reduction by thioredoxin as well . Very surprisingly , we did not find PFOR or ferredoxin among the proteins forming adducts with metronidazole , although PFOR can be readily found on 2D gels when analyzing E . histolytica cell extracts [39] . High percentage ( 20% ) acrylamide gels did not show any shifted proteins in the range of ferredoxin ( approximately 6 kDa ) . It has been reported that , in contrast to T . vaginalis or G . intestinalis , PFOR was not found to be down-regulated in metronidazole-resistant E . histolytica [22 , 23] , which supports the assumption that PFOR might not be involved in nitroimidazole activation in this organism . However , this is questioned by the fact that ferredoxin 1 levels were found to be decreased in metronidazole-resistant E . histolytica [22] . At the moment , we do not have a conclusive explanation for these contradictory results , but it is conceivable that down-regulation of ferredoxin 1 in metronidazole-resistant E . histolytica could also be an accompanying effect of down-regulation of thioredoxin and thioredoxin reductase . As a potential parallel , thioredoxin has been shown to regulate a large number of proteins in plants , including enzymes involved in glycolysis such as aldolase , enolase , glyceraldehyde 3-phosphate dehydrogenase , and triose phosphate isomerase [40] . Thus , loss of thioredoxin activity or diminished thioredoxin levels in the cell could also impair the cellular metabolism , consequently leading to a down-regulation of ferredoxin . Interestingly , a thioredoxin reductase originally with different annotation had been sequenced before the E . histolytica genome project had started . It was called disulphide oxidoreductase ( Eh34 ) [41] or later , flavin reductase , [22] , and was found to have reduced expression in metronidazole-resistant E . histolytica [22] . The very slight differences in the sequences of thioredoxin reductase from the genome project and Eh34 ( 2% on the DNA level ) and the absence of an exact copy of the Eh34 gene in the genome database have not been resolved conclusively ( I . Bruchhaus , personal communication ) . Eh34 was hypothesized by the authors to be involved in metronidazole activation because recombinant overexpression of Eh34 in E . histolytica rendered cells more vulnerable to metronidazole [22] . These unexpected findings of our colleagues strengthen our argument that thioredoxin reductase is involved in metronidazole activation in E . histolytica because down-regulation of thioredoxin reductase , as an enzyme known to be involved in oxidative stress response [42] , would otherwise be highly counterproductive during metronidazole exposure that leads to the formation of reactive oxygen species , at least under microaerophilic conditions . The data gathered prompted us to propose a model of metronidazole action in E . histolytica that implies , apart from generation of reactive oxygen species in the presence of oxygen , that toxicity of metronidazole could be attributed to covalent adduct formation with essential thiols and the proteins described , leading to impaired protein function ( Figure 6 ) . Arguably , the formation of covalent adducts and oxidative stress could even intertwine , because metronidazole toxicity was shown to be exacerbated in E . histolytica under microaerophilic conditions as compared to metronidazole treatment in the complete absence of oxygen [43] . At first glance , this is counterintuitive , because our results show that higher oxygen levels lead to less non-protein thiol depletion ( Figure 3A ) . Moreover , azomycin is by far less toxic to E . histolytica than metronidazole although it gives rise to more superoxide anion radicals ( Table 2 ) . However , azomycin forms fewer adducts with the five proteins identified , of which three , i . e . , superoxide dismutase , thioredoxin reductase , and thioredoxin , are known to be involved in antioxidant defense . We show here that metronidazole-modified recEh TrxR displays a considerably reduced thioredoxin reductase activity . Thus , it is conceivable that the higher sensitivity of E . histolytica to metronidazole in the presence of oxygen is not due to a increased toxicity of metronidazole itself , but due to a reduced tolerance to oxygen because the cell's capability to remove harmful oxidants is impaired . Thioredoxin , for example , requires prior reduction by thioredoxin reductase in order to reduce peroxiredoxin [44 , 45] , an enzyme that oxidizes hydrogen peroxide to water and oxygen . Since , in contrast to thioredoxin reductase activity , the nitroreductase activity of thioredoxin reductase is not diminished by metronidazole binding , superoxide radicals might be continuously generated after reduction of oxygen by activated nitroimidazoles . Superoxide dismutase breaks down superoxide radical anions to oxygen and hydrogen peroxide . The latter would then accumulate due to a reduced peroxiredoxin activity . In any case , even in the complete absence of oxygen , metronidazole is a potent drug , suggesting that adduct formation of nitroimidazoles with the proteins identified and with non-protein thiol compounds can be very effective in killing E . histolytica cells ( Figure 6 ) . In addition , because mRNA expression was unchanged during metronidazole-exposure with 50 μM metronidazole ( M . Tazreiter , unpublished data ) and higher doses ( e . g . , 1 mM ) of metronidazole rapidly led to the disintegration of cells , we do not believe that DNA damage plays a decisive role in short-term metronidazole-mediated toxicity . Finally , because thioredoxin reductase is a ubiquitous enzyme , we suggest that our proposed model of metronidazole action might also , at least partly , apply for other organisms . Our preliminary data from 2D gels of , for example , T . vaginalis extracts corroborate the findings presented and discussed in this study , because thioredoxin reductase was identified among the proteins modified in this parasite .
E . histolytica HM-1:IMSS cells were grown axenically at 36 . 5 °C in culture flasks that were completely filled with TYI-S-33 medium [46] and carefully sealed in order to ensure low oxygen tension . Culture medium was changed every 3 d . Es . coli BL21 ( DE3 ) was grown in LB medium with appropriate antibiotics . Agar plates contained 15 g/l of agar . When non-protein thiol levels were measured in E . histolytica while applying defined oxygen tensions , culture flasks were only half filled with growth medium , sealed with a vented plug , and then preincubated for 1 d either in tightly sealed jars containing Merck Anaerocult A ( Merck , http://www . merck . de/ ) for anaerobic conditions ( 0% O2 and 18% CO2 ) or Merck Anaerocult C for microaerophilic conditions ( 5% O2 and 8% CO2 ) or in a normal incubator in the presence of air ( 21% O2 and 0 . 4% CO2 ) for aerobic conditions . Cell cultures were then harvested , followed by the resuspension of the cell pellets in the respective preincubated media and the exposure to metronidazole or to azomycin as stated below . Cells were harvested by centrifugation at 500g at room temperature for 5 min and then washed twice with PBS to remove residual serum components . Cell lysates for 2DE were prepared as described previously [39] . Briefly , proteins were precipitated with trichloroacetic acid and acetone , and solubilized in a classical buffer containing urea , thiourea , CHAPS , and DTT . 2DE was performed as described previously [39] . Analytical gels were silver stained [47] , whereas preparative gels were stained with Coomassie Brilliant Blue R-250 . After staining , gels were scanned with an Epson 1680 Pro scanner ( Epson , http://www . epson . com/ ) and analyzed with the Melanie 2D gel analysis software ( GeneBio , http://www . genebio . com/ ) . The excised 2DE spots were destained , digested with trypsin , and analyzed by LC-ESI-QTOF-MS as described previously [48] . For protein identification , the MS/MS data were subjected to database search against the SwissProt database using the Mascot search engine ( http://www . matrixscience . com/ ) and Protein Global Server 2 . 1 ( Waters-Micromass , http://www . waters . com/ ) . A compilation of mass data is given in Figures S1–S5 . The mass of intact proteins with and without metronidazole or tinidazole treatment ( see above ) was determined using LC-ESI-QTOF-MS . An aliquot corresponding to approx 500–1 , 000 pmol of purified protein ( recombinantly produced in Es . coli ) was subjected to liquid chromatography ( LC ) using a BioBasic C4 column ( 30 × 0 . 32 mm; Thermo Electron Corporation , http://www . thermo . com/ ) using a CAP-LC ( Waters-Micromass ) . Proteins were loaded onto the column in solvent A ( water containing 0 . 1% formic acid ) and eluted using a gradient from 0%–70% solvent B ( acetonitrile containing 0 . 1% formic acid ) in 30 min . The flow rate was held at 5 μl/min throughout the analysis . Instrument calibration and tuning in the mass range from 500–4 , 000 Da was achieved using a 2 mg/ml solution of sodium iodide in 50% isopropanol . Spectra were deconvoluted using MaxEnt1 function of the MassLynx 4 . 0 SP4 software ( Waters-Micromass ) . The levels of non-protein thiols in metronidazole-treated , azomycin-treated , or in untreated cells were determined as described [49] . Briefly , cells were harvested and washed in 20 mM EDTA . Pellets were resuspended in 20 mM EDTA , and cells were disrupted by repeated freezing and thawing . Cell debris was removed by centrifugation at 20 , 000g for 15 min . After measuring protein levels with Bradford reagent , equal amounts of protein were precipitated in 5% TCA at room temperature , followed by centrifugation at 20 , 000g for 15 min . Supernatants containing the non-protein thiols were removed , and the double volume of 0 . 4 M Tris/HCl ( pH 8 . 9 ) was added . A total of 1 . 7 μl 100 mM 5 , 5′-dithiobis- ( 2-nitrobenzoic acid ) ( DTNB ) per ml reaction mixture was added . Reduction of DTNB was measured at λ = 412 nm in a Jenway 6505 UV/Vis spectrometer ( Δε412 = 13 . 6 mM−1 cm−1; http://www . jenway . com/en/index . php ) . Before addition of various amounts of metronidazole and/or cysteine , total and viable cell numbers were determined by trypan blue exclusion in a Bürker-Türk hemocytometer . After addition of reagents , cells were incubated for 20 h at 36 . 5 °C . After incubation , total and viable cell numbers were redetermined . The genes for thioredoxin reductase and thioredoxin were amplified from genomic E . histolytica HM-1:IMSS DNA . Primers were 5′-TAC GTA CGC ATA TGA GTA ATA TTC ATG ATG TTG TGA TTA TCG GC-3′ ( TrxR forward ) and 5′-TCA TCC AGC TCG AGT TAG TGG TGA TGG TGA TGA TGA GTT TGA AGC CAT TTT TCA CAG-3′ ( TrxR reverse ) for thioredoxin reductase , and 5′-TAC GTA CGC ATA TGG CTG TAC TTC ATA TTA ACG CTC TTG ATC AA-3′ ( Trx forward ) and 5′-TCA TCC AGC TCG AGT TAG TGA TGG TGA TGG TGA TGT CGT GTT TCA ACC ATT TGT TTT AAG GCA-3′ ( Trx reverse ) for thioredoxin . Forward primers include an NdeI restriction site , whereas reverse primers bear an XhoI restriction site and a hexahistidine tag for convenient protein isolation . PCR fragments were ligated into the pET 17b vector ( Novagen/VWR , http://www . emdbiosciences . com/html/NVG/home . html ) . The plasmid sequences were confirmed on both strands by using T7 and pET reverse primers ( GATC Biotech , http://www . gatc-biotech . com/en/index . php ) . The confirmed plasmids were transfected into Es . coli BL21 ( DE3 ) cells . Transformants were selected on 20 μg/ml ampicillin . Expression of recombinant proteins was induced by addition of 0 . 5 mM IPTG . If proteins were to be modified with metronidazole , 1 mM metronidazole was added to the LB medium , and cells were grown in completely filled tissue culture flasks under exclusion of air . Three hours after induction , cells were harvested and then disrupted by vigorous grinding in a mortar . Subsequently , recombinant proteins were purified via Ni-NTA spin columns ( Qiagen , http://www1 . qiagen . com/ ) . Recombinantly expressed E . histolytica thioredoxin and E . histolytica thioredoxin reductase are referred to as recEh Trx , and recEh TrxR , respectively . The assay was performed as described elsewhere [50] using recEh Trx and recEh TrxR in combination . The reaction buffer contained 100 mM potassium phosphate ( pH 7 . 5 ) , 1 mM EDTA , 1 mM DTNB , and 0 . 5 mM NADPH . All reactions were done with 2 μg/ml recEh Trx and 5 μg/ml recEh TrxR . RecEh TrxR oxidized NADPH to NADP in order to reduce the disulfide bond at the active site of recEh Trx , which , in turn , reduced DTNB . Ongoing reduction of DTNB was measured at λ = 412 nm ( Δε412 = 13 . 6 mM−1 cm−1 ) over a period of 5 min at 25 °C . Reduction of CDNB by Eh TrxR was measured by determining NADPH consumption at λ = 340 nm ( Δε340 = 6 . 2 mM−1 cm−1 ) . The reaction buffer contained 100 mM potassium phosphate ( pH 7 . 5 ) , 1 mM EDTA , and 0 . 1 mM NADPH . Reduction of CDNB was determined at a concentration of 100 μM . For reasons of practicability , because nitroimidazoles display high absorbances at λ = 340 nm and thereby heavily disturb the assay , reduction of nitroimidazoles was measured in a modified assay [30] via reduction of cytochrome c at λ = 550 nm ( Δε550 = 20 mM−1 cm−1 ) , either directly by reduced nitro compounds ( CDNB , azomycin , and metronidazole ) , or indirectly by superoxide radical anions that are generated when nitroradicals transfer an electron to molecular oxygen in the reaction buffer . In any case , nitro compounds had been previously reduced by recEh TrxR . Therefore , it was assumed that reduction of one nitro group by Eh TrxR subsequently resulted in one reduced cytochrome c molecule . Reaction mixtures contained 100 mM potassium phosphate ( pH 7 . 5 ) , 1 mM EDTA , 0 . 5 mM NADPH , 50 μM cytochrome c , 4 μg/ml recEh TrxR ( equal to 118 nM ) and different amounts of CDNB , azomycin , or metronidazole , respectively . Reduction of cytochrome c was measured over a time span of 5 min at 25 °C . In order to assess the proportion of electrons that are directly transferred from the reduced nitro compounds to cytochrome c , 2 . 5 μM bovine erythrocyte superoxide dismutase ( i . e . , an approximately 20-fold excess over Eh TrxR ) were added to the reactions to remove all generated superoxide radical anions . Bovine erythrocyte superoxide dismutase was purchased from Sigma ( http://www . sigmaaldrich . com/ ) .
The National Center for Biotechnology Information ( http://www . ncbi . nlm . nih . gov/ ) accession numbers for the proteins discussed in this paper are as follows: α-amylase ( XP_652601 ) , metronidazole target protein 1 ( XP_650662 ) , purine nucleoside phosphorylase ( XP_655398 ) , superoxide dismutase ( XP_648827 ) , thioredoxin ( XP_656726 ) , and thioredoxin reductase ( XP_655748 ) .
|
The protist parasites Entamoeba histolytica , Trichomonas vaginalis , and Giardia intestinalis grow in environments with low oxygen concentration . Infections with these parasites are commonly treated with metronidazole , a nitroimidazole drug that must be reduced for activation , resulting in several toxic metabolites . We examined the soluble proteome of metronidazole-treated E . histolytica cells for target proteins of these metabolites , applying two-dimensional gel electrophoresis and mass spectrometry . Of about 1 , 500 proteins visualized , only five formed covalent adducts with metronidazole metabolites , including thioredoxin , thioredoxin reductase , and superoxide dismutase . Metronidazole-bound thioredoxin reductase displayed diminished activity . In addition to these proteins , small thiol molecules , including cysteine , formed adducts with metronidazole . Supplementation with cysteine allowed the cells to survive otherwise lethal metronidazole concentrations . Finally , we discovered that one of the modified proteins , thioredoxin reductase , reduces metronidazole , suggesting a central role for this enzyme with regard to metronidazole toxicity . Taken together , our work reveals a new area of molecular interactions of activated metronidazole with cellular components . Because thioredoxin reductase is a ubiquitous enzyme , similar processes could also occur in other eukaryotic or prokaryotic organisms .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"biochemistry",
"infectious",
"diseases",
"pharmacology",
"microbiology",
"paramecium",
"etc)",
"ciliates",
"(tetrahymena"
] |
2007
|
Nitroimidazole Action in Entamoeba histolytica: A Central Role for Thioredoxin Reductase
|
Enterotoxigenic Escherichia coli ( ETEC ) encoding heat-stable enterotoxin ( ST ) alone or with heat-labile enterotoxin ( LT ) cause moderate-to-severe diarrhea ( MSD ) in developing country children . The Global Enteric Multicenter Study ( GEMS ) identified ETEC encoding ST among the top four enteropathogens . Since the GEMS objective was to provide evidence to guide development and implementation of enteric vaccines and other interventions to diminish diarrheal disease morbidity and mortality , we examined colonization factor ( CF ) prevalence among ETEC isolates from children age <5 years with MSD and from matched controls in four African and three Asian sites . We also assessed strength of association of specific CFs with MSD . MSD cases enrolled at healthcare facilities over three years and matched controls were tested in a standardized manner for many enteropathogens . To identify ETEC , three E . coli colonies per child were tested by polymerase chain reaction ( PCR ) to detect genes encoding LT , ST; confirmed ETEC were examined by PCR for major CFs ( Colonization Factor Antigen I [CFA/I] or Coli Surface [CS] antigens CS1-CS6 ) and minor CFs ( CS7 , CS12 , CS13 , CS14 , CS17 , CS18 , CS19 , CS20 , CS21 , CS30 ) . ETEC from 806 cases had a single toxin/CF profile in three tested strains per child . Major CFs , components of multiple ETEC vaccine candidates , were detected in 66 . 0% of LT/ST and ST-only cases and were associated with MSD versus matched controls by conditional logistic regression ( p≤0 . 006 ) ; major CFs detected in only 25 . 0% of LT-only cases weren’t associated with MSD . ETEC encoding exclusively CS14 , identified among 19 . 9% of 291 ST-only and 1 . 5% of 259 LT/ST strains , were associated with MSD ( p = 0 . 0011 ) . No other minor CF exhibited prevalence ≥5% and significant association with MSD . Major CF-based efficacious ETEC vaccines could potentially prevent up to 66% of pediatric MSD cases due to ST-encoding ETEC in developing countries; adding CS14 extends coverage to ~77% .
Enterotoxigenic Escherichia coli ( ETEC ) cause diarrheal disease in children <5 years of age in developing countries and travelers’ diarrhea among persons from industrialized countries who visit developing countries [1 , 2] . Human ETEC strains can produce a heat-labile enterotoxin ( LT ) that resembles cholera toxin and one or more heat-stable enterotoxins ( ST ) including human ST ( STh ) or porcine ST ( STp ) . Strains can produce both LT and ST ( LT/ST ) or be ST-only or LT-only . Most ETEC encode colonization factors ( CFs ) that allow the pathogen to attach to proximal small intestine enterocytes , the critical site of host-parasite interaction , before expressing enterotoxins that decrease villus tip cell absorption and evoke secretion of electrolytes and water by crypt cells [3] . Three main families of Colonization Factor Antigens ( CFAs ) are encoded by ETEC that cause diarrhea in humans including CFA/I , CFA/II and CFA/IV [3] . CFA/I is the sole member of the first family . CFA/II strains encode coli surface ( CS ) antigen 3 ( CS3 ) alone or in combination with CS1 or CS2 [3] , while CFA/IV strains encode CS6 alone or in conjunction with CS4 or CS5 [3] . CFA/I , CS1 , CS2 , CS4 and CS5 are rigid fimbriae ~6–7 nm in diameter , CS3 are thin flexible fibrillae 2–3 nm in diameter [4] , and CS6 morphology is nondescript . ETEC vaccines intending to stimulate anti-CF immunity , with or without accompanying antitoxin immunity , are in clinical development . These include purified fimbriae or tip adhesins [5] , inactivated fimbriated ETEC [6] , attenuated ETEC expressing CFs [7] , bacterial live vectors such as Shigella encoding ETEC CFs [8] , multiple epitope fusion antigens [9] , and ST toxoids [10] . Stimulating intestinal secretory IgA antibodies that bind CFs and prevent ETEC from attaching to human small intestine mucosa is generally considered to be fundamental to a successful ETEC vaccine , although some contend that parenteral vaccine-induced serum IgG antibodies that transude onto intestinal mucosa may also prevent diarrhea in humans caused by bacterial enteropathogens [11] . Most current ETEC vaccines contain a mix of antigens directed against CFA/I , CFA/II and CFA/IV antigens . Minor putative CFs also exist for which data supporting their role in pathogenesis in humans is less compelling or lacking , although they also mediate attachment to human cells in tissue culture . Possible exceptions are CS17 LT-only strains that evoked diarrhea in challenged volunteers [12] and LT-only isolates encoding CS7 that were incriminated in small cohort studies of pediatric diarrhea [13 , 14] . While minor CF antigens CS7 , CS12 , CS14 , CS17 , CS18 , CS19 , CS20 , CS21 and CS30 have received most attention , others have also been described including CS8 , CS10 , CS11 , CS13 , CS15 and CS23 [15–19] . DNA-based high-throughput diagnostics have enabled large epidemiologic studies to quantify the ETEC disease burden among young children in developing countries and to assess the prevalence of various CFs . The overall objective of the Global Enteric Multicenter Study ( GEMS ) was to estimate the population-based burden , microbiologic etiology and adverse clinical consequences of moderate-to-severe diarrhea among children 0–59 months of age in study sites in sub-Saharan Africa and South Asia to guide the development and implementation of vaccines and other interventions [1] . GEMS tested for a large number of diarrheal pathogens , including ETEC , among cases of moderate-to-severe diarrhea ( MSD ) and matched ( by age , gender , neighborhood and time of presentation ) control children without diarrhea in three age strata ( 0–11 , 12–23 and 24–59 months ) at four sites in sub-Saharan Africa and three in South Asia [1 , 20] , the geographic regions where 80% of global diarrheal deaths occur . The main underlying assumption of GEMS was that a limited number of etiologic agents may be responsible for a disproportionately large fraction of all MSD [21] . ST-producing ETEC , i . e . , LT/ST and ST-only strains , were significantly incriminated as pathogens and placed ETEC as one of the top four pathogens associated with MSD across all seven sites and age groups [1] . A secondary aim of GEMS was to elucidate the proportion of ETEC strains , by toxin genotype that encode the main CFs and selected minor CFs . Herein we present the proportion of GEMS ETEC isolates that encode the main CFs found in most ETEC vaccines under development , and the prevalence of ten other putative attachment factors ( CS7 , CS12 , CS13 , CS14 , CS17 , CS18 , CS19 , CS20 , CS21 , CS30 ) that have been proposed as potential antigens to broaden ETEC vaccine immunoprophylaxis . In addition , based on the GEMS case/control design , we utilized conditional logistic regression to assess the strength of association with MSD of ETEC of the different toxin genotypes encoding the major and minor CFs .
The rationale [20] , assumptions , clinical , epidemiological and microbiological methods of GEMS [1 , 22] , a three-year case-control study undertaken among children <5 years of age in Gambia ( Basse ) , Mali ( Bamako ) , Mozambique ( Manhiça ) and Kenya ( Siaya County ) in sub-Saharan Africa and India ( Kolkata ) , Bangladesh ( Mirzapur ) and Pakistan ( Karachi-Bin Qasim Town ) in South Asia , have been described . MSD was defined as an acute episode of diarrhea ( ≥3 loose stools during a 24-hour period ) that started within the previous seven days , was separated from another episode by ≥7 days , and was accompanied by either signs of dehydration ( sunken eyes , slow abdominal “skin pinch” recoil or administration of intravenous fluids ) , dysentery or admission to hospital based on clinical concern over diarrheal disease severity [1 , 23] . The current GEMS report includes a descriptive summary of the prevalence of CFs among ETEC isolates from cases and controls by toxin profile and country , followed by analyses that utilize the GEMS matched case-control design to test hypotheses that major or minor CFs might be significantly related to the risk of MSD . Collectively , this information can help guide ETEC vaccine developers . This research involved characterization of isolates of enterotoxigenic Escherichia coli obtained from participants in the Global Enteric Multicenter Study ( GEMS ) . The ethical review methods for this study were described in detail [23] , as well as summarized in the overall publication of the results of the clinical study "The clinical protocol was approved by ethics committees at the University of Maryland , Baltimore , MD , USA , and at every field site [1] . Written informed consent was obtained from the parent or primary caretaker of each participant before initiation of study activities [1] . ” The clinical protocol included the collection of stool specimens or rectal swabs that were tested for the presence of colonies of enterotoxigenic E . coli and for the presence of other enteric pathogens [1 , 22] . Stools/rectal swabs from cases and controls were cultured onto MacConkey and xylose/lysine/ deoxycholate agar and three E . coli colonies per subject were identified and pooled for extraction of DNA which was then tested by a multiplex PCR containing primers to amplify eltB ( LT B subunit ) and est ( ST ) [22] . ETEC strains were shipped to the University of Chile and confirmed by PCR to detect LT and ST variants ( STh and STp ) [24 , 25] . Confirmed ETEC isolates were further tested by monoplex or multiplex PCRs using primers that detect target genes encoding the major CFs ( CFA/I , CFA/II [CS1 , CS2 , CS3] , CFA/IV [CS4 , CS5 , CS6] ) [16 , 24 , 26 , 27] and various minor CFs ( CS7 , CS12 , CS13 , CS14 , CS17 , CS18 , CS19 , CS20 , CS21 , CS30 ) ( Table 1 ) [16–19 , 28] . The rationale for selecting some of the minor CFs for testing was because epidemiologic data incriminate them as being associated with pediatric diarrhea ( e . g . , LT-only strains expressing CS7 ) [13 , 14] . We tested for other minor CFs because volunteer challenges with well characterized strains encoding them showed that they can elicit diarrhea ( e . g . , CS17 and CS19 ) [29] . CS14 was studied because it has been common among ST-only and LT/ST ETEC in various reports [30 , 31] . CS18 and CS20 were studied because they share high homology . CS12 and CS21 ( “longus” ) were studied because of long-term interest of some GEMS investigators [32–34] , and their global prevalence [30] , and advocates contending that they are virulence attributes [35] . CS30 was studied because it is found in LT/STp isolates and has homology to CS18 and CS20 [19] . We also selected the cited minor CFs to be studied based on their genetic relationships within the usher genomic typing system [36–38] . The majority of ETEC CFs are synthesized and transported utilizing a chaperone-usher system that typically contains four genes encoding a periplasmic chaperone , a major fimbrial subunit , an outer membrane usher and a minor subunit tip adhesin . Since there is only a single usher gene among these ETEC CFs , they can be readily classified by their sequences [36–38] . All the major CFs except CS3 and CS6 are found within the α usher sequence group , including CFA/I , CS1 , CS2 , CS4 and CS5 . Minor CFs in this α group include CS7 , CS14 , CS17 and CS19; these homologies were another reason we tested for these CFs among the GEMS ETEC isolates . The γ2 usher family includes four minor CFs of interest , CS12 , CS18 , CS20 and CS30 , which is partly why we tested for them . CS13 belongs to the κ group [37] . CS3 and CS6 major CFs reside within the γ3 usher group . CS8 ( previously called CFA/III ) , which was not studied , and CS21 are not classifiable within the chaperone-usher system , since they are synthesized as type IVb pili . CS18 and CS20 were initially tested using previously described primers that amplify sequences within fotG ( which encodes the tip adhesin of CS18 ) [25] , and csnA ( which encodes the major subunit of CS20 ) [27] . With the recent report of CS30 , a new minor colonization factor ( CF ) [19] , and revelation of its similarity to CS18 and CS20 , new primers were designed to enhance specificity . The new primers to detect CS18 amplify a sequence within fotA ( that encodes the major fimbrial subunit ) rather than fotG . Alignments of major and minor structural subunit genes of CS18 , CS20 and CS30 are shown in Figs 1 & 2 . Reference strains served as positive controls [16] . All isolates from the 806 cases and 711 control participants whose cultures yielded ETEC isolates were also tested by polymerase chain reaction ( PCR ) for genes encoding CFA/I and CS1-CS6 , the major colonization factors ( CFs ) . In addition , these isolates were all tested by PCR for several minor CFs including CS7 , CS12 , CS14 , CS17 and CS21 , all of which had been proposed to be potential virulence attributes and potential antigens to be included in an ETEC vaccine intending to elicit anti-colonization immunity . ETEC isolates from cases ( N = 203 ) and controls ( N = 295 ) that were negative for the major CFs and for minor CFs CS7 , CS12 , CS14 , CS17 and CS21 were tested for genes encoding several additional minor CFs including CS13 , CS18 , CS19 and CS20; isolates from nine cases and eight controls could not be tested because they were not recoverable . After completion of testing for CS13 , CS18 , CS19 and CS20 , a new minor CF , CS30 , was reported as being found among a proportion of LT/ST isolates [19] . We thereupon re-tested for CS30 the 113 LT/ST isolates that were among the above-mentioned 203 case isolates; the 65 LT/ST isolates among the above-mentioned 295 control isolates were also tested . However , because of sequence homologies among CS30 , CS18 and CS20 , we also re-tested the 65 case and 113 control isolates for CS18 and CS20 using new primers that were designed to increase specificity ( vide supra ) ( Figs 1 & 2 ) . Crude bacterial lysate was obtained by boiling five pooled colonies of each ETEC isolate in 0·1% Triton X-100 for 10 min , followed by centrifugation at 8000×g for five minutes to separate template DNA in the supernatant from cellular debris . PCR was performed with total bacterial DNA in a 25-μL reaction , containing 10 mmol/L deoxyribonucleotide triphosphate mix , 30 mmol/L MgCl2 , 1× reaction buffer ( 10 mmol/L Tris–HCl , 50 mmol/L KCl ) , one Unit of Taq polymerase ( GoTaq; Promega , Madison , WI ) , and 1 μL of template DNA . Primers were used at concentrations shown in Table 1 . To prevent nonspecific amplification , we used the “hot start” technique , which includes preheating reaction mixtures to 94°C for five minutes before adding Taq DNA polymerase . Samples were amplified for 35 cycles , with each cycle comprising 90 seconds at 94°C for denaturation , 30 seconds at specific primers annealing temperatures , 60 seconds at 68°C for strand elongation , and a final extension at 72°C for five minutes . PCR products were electrophoresed in 2 . 0% agarose , stained with ethidium bromide , and amplicons identified based on expected size of the amplified product compared with amplicons of reference strains . A subset of ETEC isolates were sent to the WHO Enterotoxigenic Escherichia coli Reference Laboratory , University of Gothenburg , Sweden , where they were tested for STp , STh , major CFs and phenotypic expression of CFs using monoclonal antibodies [27] . Gothenburg primers for STp and CS5 were used in Chile in addition to local primers [16 , 24 , 25 , 27] . Presentation of the descriptive observational data and analyses were restricted to ETEC cases that had a single ETEC toxin/CF genotype pattern .
Among the 806 single toxin/CF profile cases and 711 controls , the percentages of children at each site who harbored ETEC isolates of the different enterotoxin genotypes are shown in Table 2 , revealing the relative frequency of LT-only , STh-only , STp-only , LT/STh , and LT/STp infections . Overall , 68 . 2% of isolates from cases ( N = 550 ) were either ST-only ( N = 291 , 36 . 1% ) or LT/ST ( N = 259 , 32 . 1% ) , the genotypes strongly associated with MSD in GEMS [1] . STh-only strains were isolated from 284 ( 35 . 2% ) of 806 cases . The remaining case isolates ( N = 256 , 31 . 8% ) were LT-only . The proportion of ETEC strains from MSD cases that carry major CF antigens including CFA/I and CS1-CS6 , by toxin genotype , are shown by country ( Table 3 ) and summarized by continent ( Fig 3 ) . Overall , 363 ( 66 . 0% ) of 550 ST-only and LT/ST strains encoded a major CF including 20 . 4% encoding CFA/I , 14 . 0% encoding CFA/II ( i . e . , CS3 alone or with CS1 or CS2 ) and 31 . 6% encoding CFA/IV ( i . e . , CS6 alone or with CS4 or CS5 ) . The only major CF commonly observed among LT-only isolates was CS6-only , recorded in 43 of 256 LT-only strains ( 16 . 8% ) . Only three of 256 LT-only strains ( 1 . 2% ) encoded CFA/I or CFA/II . The 975 putative ETEC strains from control subjects that arrived at the GEMS ETEC Reference Laboratory at the University of Chile were re-tested to detect LT , STh and STp genes , of which 748 were confirmed as positive . Table 4 summarizes the proportion of ETEC strains from controls that encoded the major CF antigens including CFA/I and CS1-CS6 , with data presented by country , continent and toxin genotype . Overall , 170 of 378 ST-only and LT/ST strains ( 45 . 0% ) encoded a major CF antigen including 6 . 9% encoding CFA/I , 15 . 1% encoding CFA/II and 23 . 0% encoding CFA/IV . Among 333 LT-only isolates , one ( 0 . 6% ) encoded CS3-only , 63 encoded CS6-only ( 18 . 9% ) , one encoded CS4+CS6 ( 0 . 3% ) and 29 had CS5+CS6 ( 8 . 7% ) . Recognizing that 34 . 0% of ST-only and LT/ST strains and 82 . 0% of LT-only strains do not encode a major CF , we investigated those isolates to detect ones that encode exclusively one of the following characterized minor CF antigens: CS7 , CS12 , CS13 , CS14 , CS17 , CS18 , CS19 , CS20 , CS21 or CS30 . We determined the proportion of ETEC MSD cases that had isolates encoding one of these minor CFs in the absence of a major CF and that accounted for at least 5 . 0% of the overall case isolates of that toxin genotype ( Table 5 ) . Among MSD cases with ST-only ETEC , only CS14 , identified in 58 ST-only cases ( 19 . 9% ) , reached a prevalence of ≥5% ( Table 5 ) ; four MSD cases with LT/ST isolates lacking major CFs also encoded solely CS14 ( 1 . 5% ) . Cases isolates having other minor CS antigens encoded as the sole CS were uncommon ( <5% ) among ST-only and LT/ST isolates . As an example , we cite recently described CS30 [19] . Among the 83 LT/ST cases whose isolates lacked major CFs , 16 , all LT/STp genotype , encoded CS30 but only five cases had CS30 as the sole CS . The 192 MSD cases with LT-only isolates lacking major CFs included strains encoding CS7 ( 7 . 8% ) or CS17 ( 6 . 6% ) as sole CS antigens , yielding a cumulative prevalence of 14 . 4% for LT-only strains encoding one of those two minor CFs ( Table 5 ) . The GEMS case/control design allowed us to assess the strength of association between the various major and minor CFs and MSD among cases versus their matched controls using conditional logistic regression models . To document the validity of this methodology , we first quantified the strength of association with MSD of the major CFs ( Table 6 ) , since they are widely regarded as true virulence attributes . ST-only and LT/ST strains encoding CFA/I , CFA/II and CFA/IV were all significantly associated with MSD ( p<0 . 0001 , p = 0 . 006 , p<0 . 0001 , respectively ) . In contrast , LT-only strains encoding only CS6 or CS5 and CS6 were not significantly associated with MSD ( p>0 . 05; Table 6 ) . Conditional logistic regression modeling was then performed to assess the association between LT/ST and ST-only ETEC expressing one of the 10 minor CFs alone ( CS7 , CS12 , CS13 , CS14 , CS17 , CS18 , CS19 , CS20 , CS21 or CS30 ) and MSD . Among ST-only and LT/ST ETEC strains encoding exclusively a single minor CF but no major CF , only CS14 was significantly associated with MSD ( Table 6 ) . When conditional logistic regression was performed for LT-only cases and ETEC strains encoding exclusively one of these ten minor CFs , only CS21 exhibited a significant association ( p = 0 . 028 ) . However , LT-only isolates expressing CS21 exclusively were uncommon among cases ( N = 4 ) and matched controls ( N = 0 ) . CS7 did not show a significant association ( p = 0 . 071 ) but the sample sizes of cases ( N = 20 ) and controls ( N = 12 ) were small . We did not use a Bonferroni adjustment for these individual conditional logistic regression analyses of the association of individual minor CFs with MSD , as in each instance an individual hypothesis was being tested [41–44] . ETEC isolates from 443 cases ( 338 encoding major CFs and 105 encoding a single minor CF ) were tested by dot blot immunoassay with specific anti-CF antibodies to determine the percent that phenotypically expressed on their bacterial surface the encoded major CF antigens . Of ETEC encoding CFA/I or CS1-CS5 , 73 . 8–95 . 1% of isolates tested were dot blot-positive ( Table 7 ) ; the exceptions were the 65 CS6-only isolates tested that showed only 38 . 5% positivity . Among ETEC case isolates encoding one of the four minor CFs tested , dot blot immunoassay positivity ranged from 67 . 2% for CS17 to 94 . 4% for CS7 .
The GEMS case/control study demonstrated that ST-only and LT/ST ETEC , the enterotoxin genotypes exhibited by circa two-thirds of ETEC isolates from patients , were strongly associated with MSD [1 , 49] . Field studies involving small pediatric cohorts prospectively followed under active household surveillance document that these toxin types are also incriminated as causing milder diarrhea [13 , 50 , 51] . Most LT-only strains are not associated with diarrhea [1 , 13 , 50 , 52 , 53] , as some descend from LT/ST strains through loss of genes encoding ST and a CF [3] . Nevertheless , evidence from diarrhea outbreaks in industrialized countries [54] , experimental challenges in U . S . volunteers [12 , 55] , and epidemiological studies in developing countries indicate that a subset of LT-only strains do appear to be bona fide diarrheal pathogens [13 , 14] , and it would be desirable to prevent diarrhea caused by that subset . The quandary , heretofore , has been how to identify accessory virulence attributes that distinguish the subset of LT-only ETEC that can cause diarrhea versus the non-pathogenic LT-only strains . Since individual E . coli colonies from each GEMS MSD case and their controls were tested for genes encoding LT and ST , it was possible to examine what major and minor CFs were encoded by the GEMS LT-only isolates as well as by isolates of the other toxin genotypes . Four cardinal findings emerged from examining the CF genotypes of the GEMS ETEC isolates . First , GEMS results confirm that ETEC vaccines based on stimulating immune responses to the major CFs ( CFA/I and CS1-6 ) , if highly efficacious in blocking CF-mediated attachment to enterocytes , could prevent diarrhea caused by up to 66% of the ST-only and LT/ST strains , the toxin genotypes strongly incriminated as pathogens ( Table 5 ) . The fact that ETEC encoding these CFs were observed in a very large study involving multiple representative sites in sub-Saharan Africa and South Asia validates that ETEC vaccine strategy for the geographic regions where 80% of young child diarrheal deaths occur worldwide . In contrast , major CFs were uncommon among LT-only isolates . Important avenues of ETEC vaccinology research have focused on identifying additional CFs among ST-only and LT/ST isolates that lack CFA/I , CFA/II and CFA/IV and to identify minor CFs that might be targets for protective immune responses directed against the subset of LT-only strains that are pathogenic . Thus , the second cardinal observation is identification of the proportion of strains in each toxin genotype that lacked a major CF but that exclusively expressed one minor CF , including either CS7 , CS12 , CS13 , CS14 , CS17 , CS18 , CS19 , CS20 , CS21 or CS30 . Collectively these minor CFs raised the percent of ST-only cases having a recognized CF from 64 . 3% ( 187/291 ) to 86 . 3% ( 251/291 ) and raised the percent of LT/ST cases having a recognized CF from 68 . 0% ( 176/259 ) to 79 . 9% ( 207/259 ) ( Table 5 ) . The percent of MSD cases with LT-only ETEC having a recognized CF similarly rose from 25 . 0% ( 64/256 ) to 54 . 7% ( 140/256 ) . Although minor CFs encoded by the different toxin genotypes of ETEC strains collectively raised the proportion of cases that had a CF target , it was not known if these minor CFs also identified these strains as being pathogenic , i . e . , significantly associated with MSD , as were the major CFs . Moreover , most individual minor CFs were uncommon , defined as <5% of strains of a toxin genotype that lacked a major CF . Thus , a third cardinal observation was assessment of the strength of association between MSD and ETEC encoding the various major and minor CFs within the different toxin genotypes . These novel analyses ( Table 6 ) revealed that the major CF families encoded by ST-only or LT/ST strains are significantly associated with MSD ( CFA/I and CFA/IV , p<0 . 0001; CFA/II , p = 0 . 006 ) . Thus , CFA/I , CFA/II and CFA/IV are not only surface-exposed targets for effector immune responses , but when expressed by LT/ST and ST-only ETEC they are markers indicating that these strains are strongly incriminated as diarrheal pathogens . In contrast , LT-only isolates encoding CS6 alone or CS5 and CS6 were not significantly associated with MSD ( Table 6 ) , nor was there a trend . A notable proportion ( 19 . 9% ) of ST-only isolates encoded CS14 alone , i . e . , with no other minor or major CFs , and these were significantly associated with MSD ( p = 0 . 001 ) ( Table 6 ) . No other individual minor CF was both common and significantly associated with MSD among the cases infected with ST-only and LT/ST isolates . Whereas LT-only strains encoding exclusively CS21 were significantly associated with MSD ( p = 0 . 021 ) , these strains were distinctly uncommon . LT-only strains encoding exclusively CS7 were prevalent ( 7 . 8% , Table 5 ) but they were not significantly associated with MSD ( p = 0 . 07 ) . However , the sample sizes of cases ( N = 20 ) and controls ( N = 12 ) with LT-only encoding exclusively CS7 were small , so further investigation of this CF should be encouraged to explore the potential role of CS7 for potential inclusion in an ETEC vaccine . Support for this notion comes from two small infant cohort studies in Guinea-Bissau ( N = 200 ) and Egypt ( N = 348 ) that assessed the association of ETEC encoding specific CFs with diarrhea using logistic regression models and reported significant associations of LT-only CS7 with infant diarrhea [13 , 14] . Since experimental challenge with an LT-only strain encoding CS17 fomented diarrhea in adult volunteers [12] , it was somewhat surprising that LT-only/CS17-only strains were not significantly associated with MSD in young children in GEMS . Nevertheless , because CS17 shares epitopes with CFA/I , CS1 and CS2 , and CS7 shares epitopes with CS5 , the immune responses to these major CFs within a vaccine that also contains a LT toxoid to stimulate anti-LT could collectively confer protection against pathogenic LT-only strains encoding CS17 and CS7 [56] . The fourth key observation is that inclusion of CS14 expands the breadth of vaccine coverage against ST-only pathogens , raising it from 64 . 3% to 84 . 2% ( Table 6 ) . Obviously , antigenic expansion , particularly if multiple minor CFs beyond CS14 ( e . g . , CS7 , CS17 , CS21 ) were to be added to a major CF-based ETEC vaccine , would increase the vaccine’s complexity and cost . Nevertheless , there is precedent for successfully addressing this problem with other bacterial vaccines . Pneumococcal conjugate vaccines were expanded from 7-valent to 13-valent to allow broader global coverage , while multivalent meningococcal conjugate vaccines currently include four separate serogroup conjugates . Some ETEC vaccine strategies , such as attenuated Shigella live vectors encoding two separate CFs per vector strain , can be adapted relatively easily to express additional CFs [57] . Another approach to broaden coverage of an ETEC vaccine is based on formulating a mix of fimbrial tip adhesin proteins [5] . Fimbrial CFs can be classified based on the amino acid sequence relatedness of their tip adhesin proteins , with several important ETEC CFs falling into Class 5 fimbriae assembled by the alternate chaperone pathway . Whereas the major fimbrial subunit proteins that create the stalks of these fimbriae differ substantially from one another antigenically , their tip adhesin proteins are highly conserved into three sub-classes [58 , 59] . Antibody against one adhesin of the subclass cross protects against attachment by other members . Thus , protection may also be broadened by this strategy . Selecting which tip adhesins to include in a multivalent vaccine requires knowing the frequency of the CFs among ETEC globally; so the GEMS data inform this vaccine strategy as well . Another strategy to broaden ETEC vaccine coverage would include non-fimbrial surface antigens , e . g . , EtpA ( a non-fimbrial adhesin ) and EatA ( a serine protease ) [60] . Among ETEC strains encoding a major CF other than CS6 alone or a minor CF , 67 . 2% - 95 . 1% of isolates reacted with the specific homologous anti-CF antibody by dot blot immunoassay , thereby documenting phenotypic expression . The exceptions were the CS6-only isolates of which only 25/65 ( 38 . 5% ) were dot blot-positive ( Table 7 ) whether they were LT-only isolates encoding CS6-only ( 10/32 , 31 . 3% ) or ST-encoding CS6-only strains ( 15/33 , 45 . 5% ) . The expression of CFs is highly regulated [61–63] , with temperature , bile , concentrations of glucose , glutamine and iron , and proximity to epithelial cells all influencing expression [64–67] . Thus , one explanation for lack of expression is that the in vitro growth conditions that we utilized did not induce the regulated biosynthesis of isolates encoding certain CFs . Transcriptional regulators such as CfaD ( also called CfaR ) and Rns that are members of the AraC family of transcriptional regulators modulate the expression of CFA/I , CS1 , CS2 , CS4 and CS5 [61 , 62 , 66 , 68] . In contrast , although certain growth conditions such as temperature modify CS6 expression [67] , no specific positive regulator has been identified for CS6 [68–70] . Alternatively , isolates that are PCR-positive but dot blot-negative may have single nucleotide polymorphisms , minor mutations in structural or chaperone genes or lower copy number plasmids that still allow amplification by PCR but may diminish or abrogate expression of the CF [69 , 71] . One theoretical limitation of our study is that the PCR primers designed to detect ST at the field sites were optimized for STh; thus some STp-only isolates may have been missed . LT-STp strains from cases were not under-estimated in GEMS because all LT-only strains were re-tested with PCRs individually optimized for STp and STh in the Chilean Reference Laboratory and upon re-testing only a limited number of LT-only isolates were found to be LT/STp . We believe that few cases and controls with STp-only were missed . A GEMS follow-on study detected STp and STh in genomic DNA extracted from whole stool specimens of a subset of 5304 case/matched control pairs using a TaqMan Card-based quantitative real-time PCR ( qPCR ) methodology and documented that the ST burden was overwhelmingly attributed to STh [49] . Optimized detection of STp by qPCR increased the overall ETEC disease burden estimate by only 15% versus what was recorded using the gel-based PCR methodology at the field sites [49] . This is similar to the overall difference based on presumed gene loss between primary isolation and results of re-testing strains following storage and transport to the Reference Laboratory . Other studies have found that STp-only isolates are uncommon compared to STh-only when methods sensitive for STp are used [13] . Analyzing the array of CFs among GEMS ETEC isolates has provided important information to guide ETEC vaccine development and future deployment . Since ST-only and LT/ST strains are strongly incriminated as the key ETEC pathogens , a fimbrial-based ETEC vaccine that included CFA/I , CS1-6 and CS14 , if highly efficacious , could theoretically confer protection against up to ~77% of such ETEC pathogens .
|
Enterotoxigenic Escherichia coli ( “ETEC” ) were found to be one of the four most consistently important agents that cause moderate-to-severe diarrhea among children <5 years of age in a large case-control study , the Global Enteric Multicenter Study , performed in four countries in sub-Saharan Africa and three in South Asia . ETEC attach to the lining of the human small intestine by means of protein colonization factors ( CFs ) , after which bacterial toxins stimulate intestinal secretion resulting in diarrhea . Moderate-to-severe diarrhea in young children in developing countries can lead to malnutrition and death . Vaccines are being developed to prevent ETEC diarrhea and its consequences . Several ETEC vaccines aim to stimulate antibodies ( protective proteins ) that will bind CFs and prevent the bacteria from attaching to intestinal cells , which should , in turn , prevent ETEC diarrhea . Different types of CFs exist . To guide the development of vaccines intending to provide broad protection against ETEC , one must know the frequency with which the different major CFs are produced by ETEC . This paper reports an extensive systematic survey of ETEC CFs and provides helpful information to guide the development of ETEC vaccines .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"bacteriology",
"medicine",
"and",
"health",
"sciences",
"toxins",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"enterotoxins",
"pili",
"and",
"fimbriae",
"toxic",
"agents",
"toxicology",
"vaccines",
"diarrhea",
"preventive",
"medicine",
"signs",
"and",
"symptoms",
"gastroenterology",
"and",
"hepatology",
"molecular",
"biology",
"techniques",
"infectious",
"disease",
"control",
"vaccination",
"and",
"immunization",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"adhesins",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"bacterial",
"toxins",
"microbial",
"physiology",
"artificial",
"gene",
"amplification",
"and",
"extension",
"molecular",
"biology",
"pathogen",
"motility",
"bacterial",
"physiology",
"diagnostic",
"medicine",
"cell",
"biology",
"virulence",
"factors",
"biology",
"and",
"life",
"sciences",
"vaccine",
"development",
"polymerase",
"chain",
"reaction"
] |
2019
|
Colonization factors among enterotoxigenic Escherichia coli isolates from children with moderate-to-severe diarrhea and from matched controls in the Global Enteric Multicenter Study (GEMS)
|
Besides being building blocks for protein synthesis , amino acids serve a wide variety of cellular functions , including acting as metabolic intermediates for ATP generation and for redox homeostasis . Upon amino acid deprivation , free uncharged tRNAs trigger GCN2-ATF4 to mediate the well-characterized transcriptional amino acid response ( AAR ) . However , it is not clear whether the deprivation of different individual amino acids triggers identical or distinct AARs . Here , we characterized the global transcriptional response upon deprivation of one amino acid at a time . With the exception of glycine , which was not required for the proliferation of MCF7 cells , we found that the deprivation of most amino acids triggered a shared transcriptional response that included the activation of ATF4 , p53 and TXNIP . However , there was also significant heterogeneity among different individual AARs . The most dramatic transcriptional response was triggered by methionine deprivation , which activated an extensive and unique response in different cell types . We uncovered that the specific methionine-deprived transcriptional response required creatine biosynthesis . This dependency on creatine biosynthesis was caused by the consumption of S-Adenosyl-L-methionine ( SAM ) during creatine biosynthesis that helps to deplete SAM under methionine deprivation and reduces histone methylations . As such , the simultaneous deprivation of methionine and sources of creatine biosynthesis ( either arginine or glycine ) abolished the reduction of histone methylation and the methionine-specific transcriptional response . Arginine-derived ornithine was also required for the complete induction of the methionine-deprived specific gene response . Collectively , our data identify a previously unknown set of heterogeneous amino acid responses and reveal a distinct methionine-deprived transcriptional response that results from the crosstalk of arginine , glycine and methionine metabolism via arginine/glycine-dependent creatine biosynthesis .
While amino acids are the building blocks of proteins , different amino acids also participate in a wide variety of biological processes . For example , amino acids supply carbon and nitrogen molecules for biosynthesis , feed substrates to maintain TCA cycle activity for ATP generation , and provide reducing equivalents to bolster anti-stress capacity for redox homeostasis . Therefore , all organisms have developed strategies to cope with metabolic stress and challenges posed by the deprivation of amino acids . In mammalian cells , there are at least two major adaptive mechanisms that sense and respond to fluctuations in amino acids levels . Mammalian target of rapamycin ( mTOR ) is a conserved Ser/Thr kinase that senses amino acid availability to regulate cell growth and autophagy . Another important sensor is the GCN2 ( general control nonderepressible 2 ) kinase that regulates protein translation initiation in amino acid–starved cells by detecting uncharged tRNAs . These two kinases are highly conserved from yeast to mammalian cells and play major roles in the control of protein translation , transcriptional programs , and regulation of adaptive responses during amino acid starvation . One of the downstream effects of amino acid deprivation is the phosphorylation of Ser51 on the α-subunit of eukaryotic translation initiation factor ( eIF ) 2α by GCN2 , which causes reduced rates of translation initiation and a general decline in protein synthesis . Besides GCN2 , three additional eIF2a upstream kinases , including heme regulated initiation factor 2α kinase ( HRI ) , protein kinase R ( PKR ) and protein kinase R like ER kinase ( PERK ) , guard translation initiation in response to distinct kinds of stress in mammals . All four kinases have highly similar downstream components as they all phosphorylate eIF2α on Serine 51 . While phosphorylated eIF2α generally suppresses protein synthesis , it also promotes the translation of select mRNA species that contain unique features in their 5’ untranslated regions , such as the activating transcription factor 4 ( ATF4 ) [1] . ATF4 triggers a general AAR by inducing expression of a large number of target genes , including activating transcription factor 3 ( ATF3 ) , CEBP homologous protein ( CHOP ) , and asparagine synthetase ( ASNS ) . An amino acid response element ( AARE ) in the promoters of these genes allows for the coordinated transcriptional regulation [2] . While the network of transcriptional changes of an AAR have been extensively investigated [3] , our understanding is still limited in several ways . First , are there additional , yet-uncharacterized pathways elicited as cells respond to amino acid deprivation ? Second , to what degree are similar or distinct responses triggered by deprivation of different individual amino acids ? Third , whether or how is the transcriptional response affected by the crosstalk of amino acid metabolism ? Among the twenty-two standard amino acids , nine are considered “essential” because the human body must obtain these from nutritional intake . For different types of cells , there also might be different dependencies based on their genetic makeup and metabolic flexibility [4] as well as the microenvironmental stresses of these cancer [5 , 6] . There has been much interest in identifying nutrient addictions of cancer cells with the hope for new therapeutic opportunities . For example , acute lymphocytic leukemia ( ALL ) cells are deficient in the asparagine pathway and require large amounts of exogenous asparagine . Therefore , asparaginase , through its ability to deplete extracellular asparagine , has become a cornerstone in the treatment of ALL . Glutamine addiction is found in basal-type breast cancer cells [7 , 8] and cancer cells with activated Myc and Ras [9 , 10] . Certain melanoma cells have a leucine addiction caused by defective adaptive autophagy [11] . An important implication of these studies is the significant heterogeneity in how cells sense and respond to the deprivation of each amino acid . Although the cellular response to general amino acid deprivation might be similar , the deprivation of individual amino acids may induce different phenotypic and transcriptional responses; each individual amino acid participates in distinct metabolic pathways and different cells might have different intermediate demands . However , few studies have extensively profiled the global transcriptional response to a large number of amino acids . To that end , we systematically depleted each one of , or all of , the amino acids at a time to observe cellular phenotypic and transcriptional responses . We found that the deprivation of most amino acids , except glycine , triggered a robust and mostly conserved AAR . Quite unexpectedly , methionine deprivation triggered the most dramatic and extensive gene expression changes . Methionine participates in multiple cellular metabolic pathways , including the salvage pathway , the SAM recycling pathway , the trans-sulfuration pathway for cysteine biosynthesis , polyamine synthesis , and creatine biosynthesis . Methionine , indirectly via SAM , also donates methyl groups for protein methylation , which can result in epigenetic changes when the proteins being methylated are histones . We dissected the contribution of each potential pathway to the cellular response of methionine deprivation and uncovered that creatine biosynthesis is essential for the methionine-deprived transcriptional response and histone modifications . Creatine biosynthesis depends on the availability of arginine and glycine; simultaneous deprivation of arginine or glycine abolished the methionine-deprivation transcriptional response and epigenetic changes . Together , our data reveal the heterogeneity of amino acid responses and a crosstalk among the metabolism of arginine , glycine and methionine in cells , in which arginine/glycine-dependent creatine biosynthesis is required for the methionine-deprivation response .
To determine the cellular responses to different amino acids , we performed amino acid deprivation in the breast cancer cell line MCF7 that is positive for the estrogen receptor ( ER ) and has wild type p53 [12] . MCF7 cells were usually cultured and propagated in DMEM media that contained 15 amino acids , including 11 essential amino acids and 4 non-essential amino acids . To systematically define the global transcriptional response to the deprivation of any one ( or all ) amino acids , we prepared custom-made DMEM medias with the removal of one ( or all ) amino acids and supplemented with dialyzed fetal bovine serum . Similar approaches of depriving single amino acids have been previously applied by different groups to analyze the signaling pathways [13] and AARE-driven gene expression [14] of human cell lines . First , we determined how the deprivation of each single amino acid affected the viability of MCF7 by crystal violet staining at different time points . During the experimental period of 72–120 hours , the viable cell number increased in the control media , while the deprivation of most individual amino acids dramatically reduced viable cell number ( Fig 1A ) . The only exception to this was in glycine-free media where MCF7 cells proliferated to a similar degree as in control media , suggesting that MCF7 cells do not require extracellular glycine for growth . The deprivation of most amino acids reduced cell number to approximately the same extent , except for methionine deprivation , which led to the most dramatic reduction in cell number ( Fig 1A ) . The deprivation of most amino acids led to moderate cell growth arrest within two days ( S1A Fig ) . When compared to the deprivation of representative amino acids ( leucine or glutamine ) , propidium iodide ( PI ) staining confirmed that methionine depletion caused the highest degree of cell death ( S1B Fig ) . Amino acid starvation generally triggers the transcriptional amino acid response ( AAR ) pathway and inhibits the nutrient sensing mTOR pathway . To examine the activities of these two pathways , we determined whether the phosphorylation of eIF2α ( GCN2 activation in the AAR pathway ) and S6K1 ( T398 , the mTOR sensing pathway ) were affected by individual amino acid deprivation . Consistent with the lack of effects on proliferation , glycine deprivation did not affect pho-S6K1 ( mTOR activity ) or pho-eIF2α ( GCN2 activity ) ( S1C Fig ) . In contrast , the deprivation of most other amino acids reduced phosphorylation of S6K ( pho-S6K1 ) and increased phosphorylation of eIF2α ( pho-eIF2a ) to varying degrees . These results indicate that the deprivation of most individual amino acids , except glycine , triggered the expected canonical amino acid response . To profile the transcriptional response to the deprivation of all or each of the 15 amino acids in DMEM , we exposed MCF7 to media with all ( +AA ) , no amino acids ( -AA ) or missing one of each of the 15 amino acids , in quadruplicate , for 24 hours . The 68 RNA samples were then interrogated by Affymetrix U133A2 genechips to obtain global gene expression profiles ( deposited into GEO with the accession number: GSE62673 ) . First , we normalized all intensities of the 68 arrays by RMA , then mean-centered and filtered by 3-fold variation in at least two arrays to select 1365 probesets . These samples were then grouped based on their similarity of gene expression via unsupervised hierarchical clustering ( S1D–S1F Fig ) . The control samples ( +AA ) clustered tightly with the glycine-deprived samples , consistent with the minimal effects on cell growth by glycine deprivation . This branch also contained His , Cys , Gln and Ser deprived samples ( S1D–S1F Fig ) . The second branch contained three sub-branches: Val and Lys sub-branch; Ile , Leu and Phe sub-branch; and Arg , Thr , Tyr and Trp sub-branch ( S1D–S1F Fig ) . The methionine ( -Met ) and all amino acid deprived samples ( -AA ) were in two distinct branches differentiated from all other amino acid deprived samples ( S1D–S1F Fig ) . These data were in agreement with the clustered pattern analyzed by cross-correlation of individual AAR ( S1G Fig ) . While not in perfect overlap , these clusters do have parallels with amino acid side chain characteristics: the nonpolar , branched amino acids ( Val , Ile , Leu ) were closely clustered . We also found that many features existed in one or only a subset of amino acid deprived samples ( S1F Fig ) . For example , deprivation of all , but not any one individual amino acid , induced the expression of calpain 9 ( CAPN9 ) , pre-B-cell leukemia transcriptional factor interacting protein 1 ( PBXIP1 ) and C-C chemokine receptor type 4 ( CCR4 ) . Only serine deprivation induced the expression of RAB26 ( a member of RAS oncogene family ) and vesicle-associated membrane protein 1 ( VAMP1 ) , both of which are involved in vesicle trafficking . Arginine deprivation induced a specific set of genes that included chemokines ( CCL1 and IL8 ) and kynureninase ( KYNU ) , which catalyze the degradation of kynurenine . Kynurenine was previously found to be elevated in ER- breast tumors [15] . We also found many genes in the interferon responses induced in the samples deprived of methionine , arginine and lysine , with particularly high levels in the methionine samples ( S1F Fig ) . Collectively , these results indicated that there is significant , previously-unappreciated heterogeneity in the transcriptional responses to the deprivation of different individual amino acids . To define a net transcriptional response for each amino acid deprivation , we used a zero transformation process [16] to compare the changes in levels for each gene in each of 16 conditions ( deprivation of one individual or all 15 amino acids ) to the average transcript levels of the control treatment . 3741 probe sets were selected by the mean changes with at least ±20 . 8 ( ~±1 . 74 ) fold change in more than three arrays and arranged using hierarchical clustering ( Fig 1B ) . Such analysis revealed that the deprivation of most individual amino acids triggered a similar transcriptional response with the exception of glycine and methionine . Glycine deprivation triggered little transcriptional changes ( Fig 1B ) , consistent with its lack of effects on proliferation ( Fig 1A ) , signaling ( S1C Fig ) and co-clustering with the control samples in the unsupervised analysis ( S1D–S1F Fig ) . In contrast , methionine deprivation triggered a distinct transcriptional response ( Fig 1B ) including extensive methionine-specific gene responses as well as aspects of the common amino acid deprivation response . Next , we sought to define the common amino acid response ( AAR ) signature . Using the criteria of probe sets with at least a 20 . 8 fold-change in more than three different individual amino acid deprivation samples , 778 probe sets were selected ( S1 Table; Fig 1C ) . The common AAR gene signature contained many genes known to comprise the canonical amino acid response , including ATF4 , ATF3 , ASNS , SARS and CHOP ( Fig 1C ) . We used real-time qPCR to validate the induction of ATF3 and CHOP mRNA when MCF7 was deprived of most individual amino acids , except glycine ( Fig 1D ) . In addition to these canonic AAR genes , we also noted and validated the consistent induction of TXNIP and ARRDC4 in response to the deprivation of most individual amino acids ( S1H Fig ) . Previous studies have suggested that glutamine deprivation and lactic acidosis affect glucose metabolism through the induction of TXNIP by the transcription complex MondoA/Mlx [17 , 18] . These results suggested that activation of TXNIP and ARDDC4 may also be part of the general amino acid response . The deprivation of most amino acids also induced p21 and MDM2 , two well-established p53 target genes . Their induction was validated by qPCR and we found that these inductions were strongly reduced in p53-silenced cells ( S1I Fig ) . Taken together , these data indicated that the deprivation of most individual amino acids activates the TXNIP and p53 pathways in addition to the canonical AAR . To determine the in vivo relevance of the common amino acid response , we projected our derived common AAR gene signature ( 778 probe sets ) to a breast tumor gene expression dataset with annotation of the p53 status [19] . We found that the common AAR gene signature was highly enriched in the ER and PR positive tumors ( Fig 1E ) . The tumors with high enrichment of the common AAR gene signature were mostly associated with wild type p53 and low Elston grade ( Fig 1E and S1J Fig ) . Furthermore , the tumors with high enrichment of the common AAR gene signature had better survival outcomes ( Fig 1F ) . Together , these results support an anti-growth capacity of the common amino acid response and the role of TXNIP and p53 in tumor suppression . Next , we identified individual amino acid-specific gene signatures based on the following criteria: at least ±20 . 8 fold change of a probeset induced by the target amino acid deprivation relative to the mean transcript levels of the control samples , while less than ±20 . 5 fold change of this probeset by all other amino acid deprivations ( S2A Fig ) . Among all of the amino acids tested , methionine deprivation triggered the most dramatic and distinct transcriptional response ( Fig 2A ) . Deprivation of most individual amino acids specifically affected very few probe sets ( S2A Fig; S2 Table ) , such as 3 probe sets for glutamine deprivation and 39 probe sets for lysine deprivation . In contrast , we identified 906 specific methionine-specific probe sets that included 568 induced and 338 repressed probe sets ( Fig 2A; S3 Table ) . When TRANSFAC was used to analyze the promoters of these induced methionine-specific genes , we noted an enrichment of the binding motifs of NRF2 ( nuclear respiratory factor2 ) , DEAF1 ( Deformed Epidermal Autoregulatory Factor 1 ) , GABP ( GA binding protein ) and E2F1 ( S4 Table ) . In addition , 196 induced genes also have the predicted binding sites of activating transcription factor ( ATF4 ) ( S4 Table ) . This suggests that ATF4 , that is triggered to mediate the canonical AAR ( Fig 1 ) , may also potentially contribute to the methionine-deprived specific response under methionine- deprived condition . The dramatic and distinct transcriptional response of methionine deprivation explained its distinct clustering pattern in the unsupervised analysis ( S1D–S1F Fig ) . To identify the biological pathways specifically affected by methionine deprivation , we performed Gene Set Enrichment Analysis ( GSEA ) by comparing methionine-deprived samples with samples deprived of several non-methionine amino acids , including arginine ( Arg ) , isoleucine ( Ile ) and tyrosine ( Tyr ) . When compared with arginine , the pathways enriched in the methionine-depleted samples included transcription ( Reactome transcription , Pol III transcription initiation , KEGG-RNA-polymerase ) , telomere ends and maintenance , meiotic synapsis , HoxA5 target genes and genes in response to Aplidin ( Fig 2B ) [20] . In contrast , the pathways depleted in the methionine-depleted samples included RB1 targeted cell growth genes , progesterone and estradiol response genes , genes down-regulated by glucocorticoid therapy , oxidative phosphorylation and house-keeping genes ( Fig 2C ) . Similar gene set or pathway enrichment or depletion patterns were seen when the methionine-deprived samples were compared with Ile or Tyr deprived samples . Additionally , we found that methionine deprivation suppressed ESR1 , the gene encoding the estrogen receptor ( S2B Fig ) . This observation was in agreement with depletion of estradiol response genes in the methionine-deprived samples ( Fig 2C ) . The depletion of the house-keeping genes and highly expressed ESR1 and estrogen-dependent pathways in the ER positive MCF7 suggest that exogenous methionine is required to maintain the expression of these genes . In addition , we projected the methionine-deprived specific gene signature ( MetDep-Sig , 906 probesets; S3 Table ) to a CCLE gene expression dataset from 1037 cell lines with different tumor types [21] . Interestingly , we found that the MetDep-Sig was highly enriched in the cell lines originated from haematopoietic and lymphoid tissue ( S2C Fig ) . The biological significance and the underlying mechanisms leading to such cell-type specific enrichment remain unclear . Together , methionine deprivation altered the expression of ER-dependent transcriptional program and other biological pathways in MCF7 . In addition , methionine deprivation gene expression program was a prominent feature of haematopoietic malignancy . To define the methionine concentrations at which methionine-specific responses were triggered , we exposed PC3 cells , a prostate cancer cell line , to different concentrations of methionine ( from 10% to 0% of the regular DMEM that contains 200 μM methionine ) for 24 or 48 hours . Using zero transformation analysis , we found that the methionine response could be triggered at 20 μM ( 10% ) of methionine ( S3A Fig ) , which are close to the levels found in human plasma [22] . Lower concentrations of methionine further enhanced the transcriptional response until the full methionine-deprivation response was triggered at 5μM ( S3A Fig ) . When we examined a time course of the methionine-deprivation response between 24 or 48 hours , we did not note significant differences in the expression patterns ( S3A Fig ) . To determine the cell-type specificity of the methionine-deprived responses , we compared the transcription response in MCF7 and PC3 cells . Since these two cell lines had different tissue origins , they had dramatically different basal gene expression profiles ( S3B Fig ) . Remarkably , they had an overall highly similar transcriptional response to methionine deprivation ( S3A Fig ) . Most of the 906 methionine-specific probe sets , defined in MCF7 cells , also showed concordant induction and repression in the PC3 cells ( Fig 3A ) . We used real-time RT-PCR to validate the induction of a few methionine-deprivation induced genes , including TEX14 ( Testis Expressed 14 ) , DAPK3 ( Death-Associated Protein Kinase 3 ) , ING2 ( Inhibitor of Growth Family , Member 2 ) , BAG5 ( Bcl-2-Associated Athanogene 5 ) and EGR1 ( Early Growth Response 1 ) . These genes were selected to represent the methionine-deprived specific response because their strong and reproducible induction under methionine deprivation . Also , these methionine-deprived genes encode proteins with a wide of variety of functions , including transcriptional regulation ( EGR1 [23] ) , epigenetic regulation ( ING2 [24] ) , spermatogenesis ( TEX14 [25] ) and apoptosis ( DAPK3 [26] , BAG5 [27] . While the functional implications of the induction of these genes by methionine deprivation remain unknown , we used these genes to dissect the regulation of methionine-specific gene response . In MCF7 cells , the reduction of methionine concentration to 10 μM ( from 200 μM ) started to trigger methionine-specific genes ( Fig 3B ) , which was lower than the 20 μM observed for PC3 cells . These results suggested that the level of intracellular methionine needed to drop below a certain threshold to trigger the methionine transcriptional response , and that this threshold varies in different cell types . In the PC3 cells , more cell death was seen as methionine concentration was decreased ( Fig 3C ) . Taken together , the methionine transcriptional response was triggered at 10–20 μM of exogenous methionine , and the induced transcriptional responses in MCF7 and PC3 were highly similar . We also evaluated these gene responses in non-proliferating cells when grown in media containing 0 . 5% FBS . Under these conditions , we found that methionine deprivation still triggered the AAR and the methionine-specific gene response , but to a lesser degree ( S3C Fig ) . In addition , when we examined how methionine deprivation ( in 10%FBS ) affects levels of the protein products of TEX14 and DAPK3 , two methionine deprivation induced mRNAs , we did not observe the corresponding increase in protein levels ( S3D Fig ) . These data suggested that methionine deprivation might also induce a profound inhibition of protein synthesis by the phosphorylation of eIF2α ( S1C Fig ) . However , the lack of the protein induction did not affect our mechanistic studies of the transcriptional induction of these methionine-responsive genes . Next , we sought to investigate how methionine metabolism affects the methionine-specific transcriptional response . The methionine recycling pathways consists of two branches: the S-adenosyl-methionine ( SAM ) cycle and the salvage cycle ( S4A Fig ) . In the SAM cycle , methionine is recycled from SAM via S-adenosyl-homocysteine ( SAH ) and Homocysteine ( HCY ) . In the salvage cycle , methionine is salvaged from 5'-methylthioadenosine ( MTA ) produced by polyamine biosynthesis that uses decarboxylated SAM ( dcSAM ) as substrates . In addition to methionine , methionine deprivation may affect the levels of other metabolites in the methionine recycling pathways . To evaluate this possibility , we used LC-MS/MS analysis and found methionine deprivation led to >90% depletion of SAM and MTA as well as a more modest depletion of SAH ( Fig 4A ) . S-adenosyl-methionine is a universal donor for the methylation reaction that modifies DNA , histones and other proteins . Therefore , it is possible that the methionine-specific gene expression may be due to changes in the epigenetic landscapes of the cells after methionine and SAM deprivation . Therefore , we used bisulfite pyrosequencing to determine whether a reduction of DNA methylation at the promoter regions of several methionine-deprived responsive genes might contribute to the induction of these genes . We tested seven genes , whose promoters are all located within CpG islands , and found no significant changes in DNA methylation at their promoter regions during methionine deprivation ( S4B and S4C Fig ) . In addition , we examined the changes of global DNA methylation by LINE1 assay [28 , 29] , and also found no significant changes on global DNA methylation during methionine deprivation ( S4D Fig ) . Furthermore , to globally evaluate the potential contribution of DNA vs . histone methylation to the methionine-specific transcriptional response , we compared our methionine-deprived specific genes with the published datasets on the transcriptional response to the inhibitors of DNA methylation ( 5-AZA ) and histone methylation ( DZNep ) in MCF7 cells ( GSE17589 ) [30] . The methionine-deprived specific response overlapped more transcriptional response induced by the inhibitors of histone methylation ( DZNep ) than DNA methylation ( S4E Fig ) . All these data suggested that alterations of DNA methylation did not play a major role for the methionine-deprived specific transcriptional response . Consistent with the potential role of histone methylation , we noted that methionine deprivation reduced methylation modifications of several histone residues , including H3K4me2 , H3K9me2 , and H3K27me3 ( Fig 4B ) . These results suggested a role for histone methylation in the transcriptional response to methionine deprivation . To determine this possibility in our own experimental system , we treated MCF7 cells with the EZH2 inhibitor 3-deazaneplanocin A ( DZNep ) that has been shown to be a global histone methylation inhibitor [31] . Microarray analysis indicated that DZNep treatment induced a robust gene response ( S4F Fig ) that had significant overlap with both the common AAR response and the methionine-specific transcriptional responses ( Fig 4C and S4F Fig ) . For example , YY1 , DICER1 , and SOD2 genes were induced by both methionine deprivation and DZNep treatment ( S4F Fig ) . However , there was also a portion of the methionine-deprivation induced signature , including Tex14 , DAPK3 , Egr1 and ING2 , that was not induced by DZNep ( Fig 4C and 4D ) . We also evaluated the levels of methylated histone in the promoter region of these genes during methionine deprivation . The level of H3K9me2 was high among the promoter regions of TEX14 , DAPK3 and EGR1 genes and significantly dropped during methionine deprivation ( S4G Fig , p<0 . 001 ) . The level of H3K27me3 also slightly dropped on the promoter regions of TEX14 and EGR1 genes during methionine deprivation ( S4G Fig , p<0 . 01 ) . Therefore , loss of histone methylation is associated and may partially contribute to the transcriptional response of methionine deprivation . To determine which branch ( es ) of the methionine recycling pathways might be responsible for the methionine-specific responses , we supplemented different metabolites ( SAM , SAH or MTA ) back to the methionine-depleted cells . SAM is a direct metabolic product of methionine and a substrate of both branches of the methionine recycling pathways . As expected , supplementation with SAM abolished the induction of all tested methionine-specific genes ( Fig 4E ) . The supplementation of MTA ( in the salvage cycle ) also abolished most of the methionine deprived specific gene responses . In contrast , the supplement of SAH in the SAM cycle had no impact on gene induction ( Fig 4E ) . Therefore , the depletion of SAM and MTA , and thus likely the salvage cycle , may play a more important role than SAH on the methionine-specific transcriptional responses in MCF7 cells . However , the result was tentative since we did not measure the cellular SAH level to determine the efficacy of SAH supplementation . In MCF7 , the MTAP locus is deleted and the methionine salvage pathway is disrupted [32] . Therefore , the supplemented 5'-Methylthioadenosine ( MTA ) cannot be readily salvaged back to replenish methionine during methionine deprivation . Therefore , MTA could not abolish the methionine-specific response by simply restoring methionine levels . MTA is a byproduct of the polyamine biosynthesis that combines the decarboxylated S-adenosyl-methionine ( dcSAM ) ( from SAM ) and putrescine ( from ornithine ) to synthesize spermidine and spermine ( S4A Fig ) . High levels of MTA inhibit polyamine biosynthesis [33] . Therefore , we reasoned that the supplementation of MTA might abolish the methionine-specific response by inhibiting the polyamine pathway . To test this possibility , we interrupted polyamine synthesis by inhibiting critical enzymes or removing its substrate arginine . We targeted the two key enzymes in the polyamine synthesis: ornithine cyclodeaminase ( ODC1 that catalyzes the synthesis of putrescine from ornithine ) and spermidine synthase ( SRM that catalyzes the synthesis of spermidine from putrescine ) to determine methionine-deprived specific responses . Surprisingly , we found that the inhibition of polyamine synthesis by genetically silencing ODC1 ( Fig 5A ) or SRM ( S5A Fig ) further enhanced the induction of TEX14 , DAPK3 and BAG5 during methionine deprivation . Similarly , the ODC1 inhibitors POB and DFMO also did not abolish the methionine-deprived specific gene responses ( S5B Fig ) . Polyamine synthesis requires exogenous arginine to generate the immediate substrate , ornithine ( S4A Fig ) . Therefore , we tested the relevance of exogenous arginine on methionine-specific gene response . Remarkably , simultaneous removal of both arginine and methionine almost completely abolished the methionine-specific gene response , including the induction of TEX14 , DAPK3 , EGR1 and ING2 in both MCF7 ( Fig 5B ) and PC3 cells ( Fig 5C ) . Furthermore , this abolishment of the methionine-deprived specific gene responses was specific to arginine deprivation , since the simultaneous removal of glutamine or cystine ( cysteine precursor ) with methionine did not have similar effects ( S5C Fig ) . Similar gene regulation patterns also occurred in the untransformed human primary fibroblast cells WI-38 and IMR90 ( Fig 5D and S5D Fig ) . Interestingly , this dependence on exogenous arginine for the methionine-deprived response was consistent with our initial array analysis , in which all amino acids deprived samples also lacked the methionine-deprived response ( S5E Fig ) . We then used microarrays to formally determine the effects of arginine deprivation on the global expression of MCF7 cells that were deprived of methionine , arginine or both methionine and arginine . We found that the induction of the 906 methionine-deprivation specific probesets was mostly abolished by the simultaneous removal of arginine ( Fig 5E and S5F Fig ) . Together , our data indicated that exogenous arginine was required for the methionine-deprived specific gene response . Since the blockage of polyamine biosynthesis pathway was unable to abolish the methionine-deprived specific gene response , we investigated other metabolic pathways by which exogenous arginine may regulate the methionine-deprived gene response . In addition to polyamine synthesis , arginine also participates in the urea cycle , nitric oxide production and creatine biosynthesis ( S4A Fig ) . We used inhibitors targeting metabolic enzymes in each of these arginine-dependent pathways . First , blocking the urea cycle by nor-NOHA ( an arginase inhibitor ) did not repress the inductions of methionine specific genes TEX14 , DAPK3 and EGR1 ( S6A Fig ) . Second , neither a NO scavenger ( c-PTIO ) nor inhibition of the nitric oxide synthesis ( NOS ) by L-NAME affected the methionine-specific gene responses ( S6B Fig ) . These data mostly ruled out the role of the urea cycle and nitric oxide production as mechanisms by which arginine regulates the methionine-specific response . However , this conclusion was tentative since we did not validate the intended inhibition of arginase and NOS . Finally , we examined the role of the arginine-dependent creatine biosynthesis pathway . Creatine biosynthesis consists of two steps ( S4A Fig ) : First , arginine and glycine are catalyzed by arginine:glycine aminotransferase ( AGAT ) to produce guanidinoacetate and ornithine . Second , guanidinoacetate N-methyltransferase ( GAMT ) transfers the methyl group from SAM to guanidinoacetate to yield creatine . Therefore , glycine is the other substrate required for the creatine biosynthesis . We considered whether the deprivation of glycine , similar to arginine , also affected creatine synthesis to abolish the methionine-specific response . However , glycine can readily be synthesized from serine by serine hydroxymethyltransferase ( SHMT ) , consistent with the lack of cellular response to glycine deprivation ( Fig 1 ) . Therefore , we removed both serine and glycine to deplete intracellular glycine . As expected , the depletion of either serine or glycine alone had no impact on the induction of these genes during methionine deprivation ( Fig 6A ) . However , depletion of serine and glycine completely abolished the induction of TEX14 , DAPK3 , EGR1 and ING2 upon methionine depletion in MCF7 cells ( Fig 6A ) . In PC3 cells , co-depletion of serine and glycine with methionine was unable to abolish the methionine-deprived transcriptional response ( S6C Fig ) , suggesting that there are alternative sources for the supply of glycine . In rodent and human , threonine can be synthesized to glycine [34 , 35] . Indeed , co-deprivation of threonine , serine and glycine completely abolished the induction of the methionine-deprived responsive genes ( S6D Fig ) . The use of SAM as a substrate for creatine biosynthesis has been considered a major SAM-consuming reaction [36 , 37] . Therefore , interrupting creatine biosynthesis by removing either arginine or glycine possibly slows down the SAM depletion and the resulting histone demethylation . Indeed , we found that the levels of several histone methylations , such as H3K4me2 , H3K4me3 , H3K9me2 , and H3K9me3 decreased during methionine deprivation , but were restored by the co-deprivation of methionine and arginine in both MCF7 and PC3 cells ( Fig 6B and 6C ) . In contrast , the reduction of H3K27me3 and H3K36me3 under methionine deprivation was modest , and sometimes inconsistent , in both MCF7 and PC3 cells . The reduction of H3K4me3 during methionine deprivation was also fully restored by the simultaneous deprivation of both serine and glycine in MCF7 and of all three threonine and serine and glycine in PC3 cells ( S6E Fig ) . We also evaluated the levels of related intracellular metabolites during deprivation of methionine , arginine , and both methionine and arginine . Consistently , methionine deprivation reduced the levels of intracellular methionine , SAM , SAH and MTA ( S6G Fig ) . Co-depletion of arginine and methionine maintained significantly higher levels of SAM and MTA than with methionine deprivation alone ( Fig 6D , S6F and S6G Fig; p < 0 . 01 ) . In addition , arginine depletion significantly reduced intracellular ornithine ( p < 0 . 001 ) , but not creatine ( S6G Fig ) . Genetic silencing of either AGAT or GAMT by shRNA to interrupt the creatine biosynthesis also significantly repressed the induction of methionine deprived specific genes ( Fig 6E and S6H Fig ) . Therefore , our data suggested that arginine- and glycine-dependent creatine biosynthesis consumes intracellular SAM to reduce histone methylation and cause a methionine-deprived specific gene response . Since intact creatine biosynthesis was required for reduction of histone methylations during methionine deprivation , we examined whether the inhibitors of histone methylation was able to restore the methionine-deprived specific gene response in the context of blocked creatine biosynthesis . Indeed , the general histone methylation inhibitor DZNep ( DZN ) fully restored the induction of TEX14 , ING2 and BAG5 genes when creatine biosynthesis was prevented by co-depleting serine , threonine and glycine ( the glycine branch of creatine synthesis ) ( Fig 7A ) . However , DZN only partially restored the induction of the methionine-deprived specific genes when the arginine branch of creatine biosynthesis was blocked ( Fig 7A ) . Besides being used for creatine biosynthesis , arginine can be synthesized to ornithine , as arginine deprivation reduced the level of intracellular ornithine ( S6G Fig ) . We hypothesized that ornithine-mediated signaling was also required for the full induction of methionine-deprive specific gene response . Indeed , addition of ornithine with DZNep rescued most of the methionine-deprived specific gene response when the creatine biosynthesis was abrogated by arginine depletion ( Fig 7B ) . We also examined UNC0638 ( UNC ) , a specific inhibitor of methyltransferases G9a and GLP , which are responsible for H3K9 di-methylation . We found that UNC had no significant rescuing effects on the induction of methionine-deprived specific genes ( Fig 7B ) , suggesting that demethylation of H3K9me2 alone may be not sufficient for the observed methionine-deprived transcriptional changes . The addition of ornithine enhanced the induction of some genes in the context of methionine deprivation alone , but ornithine alone was unable to rescue the induction of the methionine-deprived specific genes when the creatine biosynthesis was blocked by arginine depletion ( S7A Fig ) . Creatine had similar enhancement effects as ornithine on the methionine-deprived gene response ( S7B Fig ) . Taken together , our data suggested a model in which an intact arginine- and glycine-dependent creatine biosynthesis contributes to SAM depletion and resulting epigenetic changes . The combination of these epigenetic alterations and ornithine-mediated signaling are required to fully induce a specific methionine-deprived transcriptional response ( Fig 7C ) . Based on these data , we propose a model that the intact arginine- and glycine-dependent creatine biosynthesis contributes to two events in the methionine-deprived specific gene response: depletion of SAM for epigenetic changes and maintaining ornithine-mediated signaling ( Fig 7C ) .
In this study , we have comprehensively profiled the transcriptional responses to the deprivation of 15 individual amino acids . While the deprivation of most amino acids triggered the canonical amino acid response ( AAR ) , we further identified unexpected features and heterogeneities of the amino acid deprivation responses . Based on our observations , we propose a model of a shared amino acid response ( AAR ) , as well as a distinct methionine-specific response . The shared AAR was mediated by at least three different pathways , including the well-defined canonical AAR , p53 and TXNIP pathways . The unique and extensive transcriptional response to methionine deprivation was dependent on both epigenetic changes and the ornithine-mediated signaling events . We showed that methionine deprivation depleted cellular SAM and histone methylation . Importantly , we determined that arginine- and glycine-dependent creatine biosynthesis was required for the methionine-deprived specific response due to its role in SAM consumption . Collectively , our data revealed previously unappreciated heterogeneity among individual AARs and a distinct methionine-deprivation gene response that resulted from the crosstalk between arginine , glycine and methionine metabolism through creatine biosynthesis and ornithine-mediated signaling . In addition to the induction of a canonical AAR , we found that the deprivation of most amino acids also leads to a consistent activation of p53 and induction of TXNIP . Therefore , the activation of p53 and TXNIP should be considered additional branches of the common AAR , at least in cells with intact p53 . Activation of these programs by amino acid deprivation was reported by isolated previous studies . For example , serine deprivation can activate p53 and induce metabolic reprogramming [38] . Here , our results showed that p53 activation seems to be a general feature in the response to the deprivation of most amino acids and therefore suggests that the p53 status of tumor cells may affect their response to amino acid deprivation . One potential mechanism by which an AAR could activate p53 is through the ribosomal stress response [39] . Due to an excess of uncharged tRNAs , an imbalance of the ribosomal biogenesis results and the released ribosomal proteins bind and inhibit MDM2 to activate p53 and halt cell cycle progression [39] . Therefore , p53 activation , together with mTOR inhibition , may both contribute to the arrest of cellular proliferation as energy conservation mechanisms during amino acid deprivation . Under glutamine deprivation , the induction of MondoA-TXNIP activates glucose uptake to compensate for the limited availability of another carbon source [17] . The MondoA-Mlx transcription complex plays a pivotal role in glucose homoeostasis by activating TXNIP and limiting glycolysis in response to glucose 6-phosphate , adenine nucleotides and acidosis [18 , 40] . An ATF5-TXNIP axis was also suggested to switch from an adaptive UPR from ER stress to a terminal UPR and cell death [41] . It remains to be determined whether MondoA or ATF5 play a major role in the TXNIP induction associated with amino acid deprivation . Given the well-recognized role of TXNIP to repress glucose uptake and glycolysis [42] , we expect that amino acid deprivation may also affect glucose metabolism in cancer . Among different amino acids , our most striking result was that methionine deprivation induced a unique transcriptional response . Since SAM , a product of methionine catabolism , is a major donor of methyl groups for various epigenetic modifications in cells , we hypothesize that its depletion would result in epigenetic changes , and that these changes would play a role in the methionine-specific transcriptional responses . Through investigation of this hypothesis we discovered that histone methylation , rather than DNA methylation , played a significant role in the methionine-specific response . We noted a marked reduction in the levels of histone methylation , but very little difference in the levels of DNA methylation on relevant gene promoters and global DNA methylation . In addition , global inhibition of histone methylation , but not DNA methylation , induced a transcriptional response that had significant overlap with the methionine-deprived gene response . While this may reflect the different turnover mechanisms and dynamics of DNA and histone methylation , it also suggested an important role for histone methylation in the cellular response to methionine deprivation . Since the epigenetic landscape of a cell , defined in part by the DNA and histone methylation , contributes greatly to tissue-specific expression patterns , however , we observed that the methionine-deprived gene response did not exhibit much tissue-specific features . Our results of reduced histone methylation are consistent with a recent study that shows methionine deprivation of human ES cells reduced tri-methylation of histone H3 lysine-4 ( H3K4me3 ) , an active mark that is crucial for maintaining the stem cell fate [43] . In the mouse ES cells that depend on the threonine for SAM levels , threonine deprivation also reduced the level of H3K4m3 [35] . Very intriguing , H3K4me2 , 3 , as informative as activating methylation marks generally associated with active gene transcription[44] , are largely reduced during methionine deprivation . We speculate that induction of the methionine-deprived specific transcription response may be due to rearrangements of reduced both active and repressive histone marks , not solely relying on either one . Additionally , we observed a highly similar methionine-deprivation transcriptional response in MCF7 ( breast ) and PC3 ( prostate ) cells , regardless of their different tissues of origin and dramatically different basal levels of gene expression . These data suggested that shared signaling events , in addition to epigenetic alterations , play an important role in the methionine-specific responses . We determined that at least one “second” signal is mediated by ornithine . Methionine and its metabolic derivatives participate in several diverse metabolic pathways , including the biosynthesis of polyamines , glutathione , purines , and creatine . We identified that creatine biosynthesis was particularly crucial for the methionine-deprived specific gene response . Since both arginine and glycine are substrates for the creatine biosynthesis , the deprivation of either amino acid blocks creatine biosynthesis and abolishes the methionine-deprivation transcriptional response . We determined that the large consumption of SAM during creatine biosynthesis contributed significantly to the SAM depletion during methionine deprivation . Indeed , the co-deprivation of arginine and glycine , to block creatine biosynthesis and delay SAM depletion , allowed for the maintenance of histone methylation under methionine deprivation . In addition , we found that ornithine-mediated signaling was also required for the complete induction of the methionine-deprived transcriptional response . This signaling may account for the similar methionine-deprived transcription response that occurred in cells with different origins . We propose a model in which the intact arginine- and glycine-dependent creatine biosynthesis contributes to two events for the methionine-deprivation specific gene response: depletion of SAM for epigenetic changes and maintaining ornithine-mediated signaling ( Fig 7C ) . There may be other factors that contribute to the methionine-deprived specific transcriptional responses , such as ATF4 or other transcriptional factors . This model reveals the importance of the crosstalk between arginine , glycine and methionine metabolism for methionine-deprivation mediated biological consequences . These data are also highly relevant for developing chemical strategies to target both pathways to fully replicate the benefits of methionine deprivation . Our studies do have limitations . First , we focused on early time points within the first 1–2 days of methionine deprivation . At this time point , the main epigenetic changes occur at the histone methylation levels ( Figs 6B and S4D ) . It is possible that long-term methionine deprivation may lead to additional changes , such as DNA methylation . Second , it will be important to consider the range of methionine levels in plasma and tissues to determine the relevant range of responses under varying methionine levels during physiological and pathological adaptations . Importantly , we did observe significant gene expression changes triggered around the physiological levels of methionine in plasma ( ~20 μM ) [22] . Third , while we used several specific methionine-deprived genes to represent methionine-deprived specific responses to study their regulatory mechanisms , the results may be confined to these genes unless we have followed up with global array studies , such as the co-depletion of methionine and arginine ( Fig 5E ) . Several studies of methionine restriction in human and rodents have found that methionine deprivation can lower the plasma methionine levels and slow tumor growth in rodent models [45–47] . Therefore , methionine restriction may be an important strategy to treat the cancers that exhibit a dependence on methionine for survival and proliferation . We found that methionine deprivation activates the transcription of HoxA5 target genes . Compromised HOXA5 function can limit p53 expression in human breast tumors and overexpression of HOXA5 can induce tumor cell death and decrease invasive abilities of lung tumor cells [48] . Additionally , the methionine-deprivation signature was enriched for genes activated in response to Aplidin , a marine organism-derived compound with potent anti-myeloma effects that is currently in clinic trials [20] . The enrichment of HOXA5- and Aplidin-induced genes in the methionine-deprived samples may be consistent with the anti-tumor potential of methionine deprivation . Methionine dependence in cancer may also be due to one or a combination of deletions , polymorphisms or alterations of genes involved in the de novo and salvage pathways . Cancer cells with these defects in methionine metabolism are unable to regenerate methionine , and thus are addicted to exogenous methionine for survival or proliferation . Mostly interestingly , methionine restriction can increase longevity across species , including yeast , flies and rodents [49–51] . In addition to general mTOR inhibition by amino acid restriction , methionine restriction likely has unique features to extend lifespan , which are under active investigations [52 , 53] . Our data reveal that methionine restriction causes a unique methionine-deprived transcriptional response in the context of intact arginine/glycine dependent creatine biosynthesis . Supplements of arginine , or its derivatives ornithine and creatine , all of which are metabolites in the creatine biosynthesis pathway , have beneficial anti-aging effects , such as reduced risk of vascular and heart disease , reduced rate of erectile dysfunction , improvement in immune response and inhibition of gastric hyperacidity [54–56] . Our data indicates that methionine deprivation induced its unique gene response through a reduction of both histone methylation and ornithine-mediated signaling . It remains to be seen if the anti-aging effect of methionine restriction in vivo is associated with histone demethylation and/or ornithine-mediated signaling . It would be interesting to test in organisms whether additional arginine restriction abolishes the beneficial effects of methionine restriction on lifespan extension and examine whether the defined methionine-deprived gene signature is enriched in methionine-restricted animal tissues or in long-lived human individuals . Based on our current understanding of the consequences of methionine restriction , manipulating the level of methionine in our bodies may have beneficial roles for cancer control and anti-aging . Since the main source of methionine for humans is from food , one strategy to lower in vivo methionine levels would be to restrict or remove methionine from the diet . Vegan diets , which have lower levels of methionine , may therefore be a useful nutritional strategy to combat cancer growth and extend lifespans . Alternatively , plasma methionine levels could be reduced in vivo by methioninase , which degrades methionine [57] , similar to using asparaginase to deplete plasma asparagine levels in the treatment of ALL . Our studies present the mechanisms underlying a unique methionine-deprived transcriptional response , which may provide useful insights to understand the nature of methionine restriction for cancer control and lifespan extension .
MCF7 breast cancer cells and PC3 prostate cancer cells were cultured in Dulbecco's modified Eagle's medium ( DMEM; GIBCO-11995 ) supplemented with 10% fetal bovine serum and 1 × antibiotics ( penicillin , 10 , 000 UI/ml and streptomycin , 10 , 000 UI/ml ) . To prepare amino acid deficient media , Earle’s balanced salt solution was added with 4 . 5 g/L glucose ( Sigma-Aldrich ) , MEM vitamin solution ( Invitrogen ) , 0 . 37mM sodium bicarbonate ( Sigma-Aldrich ) , 24 . 8 μM ferric nitrates ( Sigma-Aldrich ) , with deficiency of one ( or all ) amino acid , and supplemented with 10% dialyzed FBS ( 10 , 000 MW cutoff , Sigma ( F0392 ) ) and 1X antibiotics . The cells were maintained in a humidified incubator at 37°C and 5% CO2 . Additional materials and methods were listed in S1 Text . Relative cell number was monitored by crystal violet staining . Triplicate samples of 104 MCF7 cells were seeded in 96 well plates . After 24 hours , the medium was removed and cells were washed twice with 1X PBS . The cells were then treated with individual amino acid deprivation media or the control full media . After the appropriate time period , the cells were fixed and stained with 1% crystal violet solution . With extensive washing , crystal violet was resolubilized in 10% acetic acid and quantified at 595 nm as a relative measure of cell number . Alternatively for propidium iodide staining method , the cells treated with the indicated amino acid deprivation were trypsinized and collected , fixed in ice cold 70% ethanol overnight . Cells were washed twice with 1XPBS buffer and resuspended in PBS buffer containing 50 μg/mL propidium iodide ( PI ) and 10 μg/ml RNase A . Flow cytometry was carried out using BD FACSCanto II flow cytometer . At least 10 , 000 cells were analyzed per sample . RNA from MCF7 cells exposed to the control , one ( or all ) amino acid deprivation conditions ( four replicates in each condition ) for 24 hours or RNA from PC3 cells exposed to different concentration of methionine ( three replicates in each condition ) for 24 hours or 48 hours were extracted by RNAeasy kits ( Qiagen ) and hybridized to Affymetrix human genome 133A 2 . 0 arrays with standard protocol . Similar methods were applied on PC3 RNA analysis . The data was deposited in NCBI GEO site ( GSE62673 ) . Affymetrix Probe intensities were normalized as log2 value by RMA and the expression data were subjected to unsupervised hierarchical average linkage clustering using Cluster 3 . 0 and displayed using TreeView . For supervised clustering analysis , the changes of gene expression upon one ( or all ) amino acid deprivation were derived by zero-transformation ( Δlog2 ) against the control condition . The probesets that varied from the baseline by 20 . 8 fold in at least 4 chips were selected for hierarchical clustering . To observe common amino acid deprivation gene response , 778 probes were selected by 20 . 8 fold varied from the control samples in at least 4 different individual amino acid deprivations . To identify the methionine deprived specific gene response , 906 probes were selected in methionine deprivation samples by 20 . 8 fold varied from the control samples , while no more than 20 . 5 fold changes in other individual amino acid deprivation . The common AAR gene signature “R-value” projection analysis was performed according to a previous report [58] . For example , the common AAR gene signature ( using ‘‘1” and ‘‘-1” , for up and down , respectively ) was listed in S1 Table . To score each tumor within a dataset for similarity to common AAR gene signature , we derived an R-value for each tumor in correlation to common AAR gene signature . The R-value was computed as the Pearson’s correlation between the pattern of common gene signature and the tumor’s expression dataset that were firstly normalized by the mean . In this way , tumors with high R-values would tend to have high similarity with common AAR gene signature . Data were analyzed using Gene Set Enrichment Analysis ( GSEA ) as described using indicated selection criteria . RNA was reverse-transcribed to cDNA with the SuperScript II reverse transcription kit using random hexamers . The gene expression level was measured by quantitative PCR ( qPCR ) with Power SYBR Green PCR Mix ( Applied Biosystems ) following the manufacturer's protocol . Genomic DNA was extracted from MCF7 cells using the QiaAmp DNA mini-prep kit according to the protocol provided by the manufacturer ( Qiagen ) . The genomic DNAs ( 800 ng ) were modified by treatment with sodium bisulfite using the Zymo EZ DNA Methylation kit ( Zymo Research , Irvine , CA ) . Bisulfite treatment of denatured DNA converts all unmethylated cytosines to uracils , leaving methylated cytosines unchanged , allowing for quantitative measurement of cytosine methylation status . Pyrosequencing was performed using a Pyromark Q96 MD pyrosequencer ( Qiagen ) . The bisulfite pyrosequencing assays were used to quantitatively measure the level of methylation at CpG sites contained . Assays were designed to query CpG islands using the Pyromark Assay Design Software ( Qiagen ) . PCR was performed using the following conditions: 95ns: 95following conditions: 95ions: 95 metet level of metl of metl of meteasure the level of mete level of metel of metevel of mett of cytosPyrosequencing was performed using the sequencing primer . PCR conditions were optimized to produce a single , robust amplification product . Defined mixtures of fully methylated and unmethylated control DNAs were used to show a linear increase in detection of methylation values as the level of input DNA methylation increased ( Pearson r > 0 . 98 for all regions ) . Once optimal conditions were defined , each assay was analyzed using the same amount of input DNA from each specimen ( 40 ng , assuming complete recovery after bisulfite modification ) . Percent methylation for each CpG cytosine was determined using Pyro Q-CpG Software ( Qiagen ) . Chromatin immunoprecipitations were performed using digested chromatin from MCF7 cells according to the protocol of SimpleChIP Enzymatic Chromatin IP Kit ( Cell signaling , #9002 ) . Purified DNA was analyzed by qPCR methods . All primers used in this study were listed in S5 Table . This section contains detailed materials and methods for metabolomics data shown in Fig 4A ( high resolution LC-MS ) , Figs 6D , S6F and S6G ( Quantitative LC-MS/MS ) . Experimental results were analyzed with a Student's t test and graphed using Prism ( GraphPad Software , Inc . ) . Data are expressed as mean ± SD with a p value <0 . 05 was considered statistically significant .
|
In order for mammalian cells to live and function , amino acids are required for protein synthesis and the generation of metabolic intermediates . An imbalance or deficiency of amino acids often triggers an “amino acid response” ( AAR ) to allow cells to adapt to their environment . However , it remains unclear whether the deprivation of any single amino acid leads to similar or different changes compared to the global AAR response or to other single amino acid deficiencies . To answer this question , we removed each or all of the 15 amino acids found in media from cells and comprehensively profiled the resulting changes in their RNA expression . Strikingly , we found a unique and dramatic gene expression program that occurred only when cells were deprived of methionine , but not any other amino acid . We also found that these methionine-specific changes depended on changes in histone modifications and an intact creatine biosynthesis pathway . Methionine deprivation reduced the degree to which histone proteins were indirectly modified by methionine ( histone methylation ) . Creatine biosynthesis consumed methionine’s derivate S-Adenosyl-L-methionine ( SAM ) , contributing to the reduction of histone methylation and an increase in ornithine-mediated signaling . Since methionine restriction may have anti-aging and other medical uses , our findings provide insights that will lead toward a better understanding of the underlying effects of methionine restriction and eventually improve human health .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Comprehensive Profiling of Amino Acid Response Uncovers Unique Methionine-Deprived Response Dependent on Intact Creatine Biosynthesis
|
Seasonal influenza surveillance is usually carried out by sentinel general practitioners ( GPs ) who compile weekly reports based on the number of influenza-like illness ( ILI ) clinical cases observed among visited patients . This traditional practice for surveillance generally presents several issues , such as a delay of one week or more in releasing reports , population biases in the health-seeking behaviour , and the lack of a common definition of ILI case . On the other hand , the availability of novel data streams has recently led to the emergence of non-traditional approaches for disease surveillance that can alleviate these issues . In Europe , a participatory web-based surveillance system called Influenzanet represents a powerful tool for monitoring seasonal influenza epidemics thanks to aid of self-selected volunteers from the general population who monitor and report their health status through Internet-based surveys , thus allowing a real-time estimate of the level of influenza circulating in the population . In this work , we propose an unsupervised probabilistic framework that combines time series analysis of self-reported symptoms collected by the Influenzanet platforms and performs an algorithmic detection of groups of symptoms , called syndromes . The aim of this study is to show that participatory web-based surveillance systems are capable of detecting the temporal trends of influenza-like illness even without relying on a specific case definition . The methodology was applied to data collected by Influenzanet platforms over the course of six influenza seasons , from 2011-2012 to 2016-2017 , with an average of 34 , 000 participants per season . Results show that our framework is capable of selecting temporal trends of syndromes that closely follow the ILI incidence rates reported by the traditional surveillance systems in the various countries ( Pearson correlations ranging from 0 . 69 for Italy to 0 . 88 for the Netherlands , with the sole exception of Ireland with a correlation of 0 . 38 ) . The proposed framework was able to forecast quite accurately the ILI trend of the forthcoming influenza season ( 2016-2017 ) based only on the available information of the previous years ( 2011-2016 ) . Furthermore , to broaden the scope of our approach , we applied it both in a forecasting fashion to predict the ILI trend of the 2016-2017 influenza season ( Pearson correlations ranging from 0 . 60 for Ireland and UK , and 0 . 85 for the Netherlands ) and also to detect gastrointestinal syndrome in France ( Pearson correlation of 0 . 66 ) . The final result is a near-real-time flexible surveillance framework not constrained by any specific case definition and capable of capturing the heterogeneity in symptoms circulation during influenza epidemics in the various European countries .
Seasonal influenza is an acute contagious respiratory illness caused by viruses that can be easily transmitted from person to person . Influenza viruses circulate worldwide causing annual epidemics with the highest activity during winter seasons in temperate regions and produce an estimated annual attack rate of 3 to 5 million cases of severe illness and about 250 to 500 thousand deaths around the world [1] . National surveillance systems monitor the influenza activity through a network of general practitioners ( GPs ) who report the weekly number of influenza-like illness ( ILI ) cases among the overall patients [2] . These traditional surveillance systems usually represent the primary source of information for healthcare officials and policymakers for monitoring influenza epidemics . However , due to the lack of specificity of influenza symptoms , they adopt quantitative indicators ( influenza-like illness ( ILI ) or acute respiratory illness ( ARI ) being the two most common ) which are defined at country level , while no defined standard exists at the international level [3–5] . One main reason might be that classification of ILI cases in GPs’ reports is usually based on common clinical symptoms observed among patients and , as with any syndromic-based disease surveillance , case definitions of “influenza-like illness” can vary [6–10] . They typically include fever , cough , sore throat , headache , muscle aches , nasal congestion , and weakness . Some previous works from hospital-based studies [11 , 12] , age-specific antiviral trials [7 , 13 , 14] and national surveillance activities [15] aimed at exploring suitable ILI case symptomatic descriptions but , so far , no unique definition has been widely adopted by the various national surveillance systems worldwide . For this reason , seasonal influenza surveillance in European countries remains rather fragmented . Only in recent years , some state members have adopted the case definition provided by the European Center for Disease Control and Prevention ( ECDC ) which defines an ILI case as the sudden onset of symptoms with one or more systemic symptoms ( fever or feverishness , malaise , headache , myalgia ) plus one or more respiratory symptoms ( cough , sore throat , shortness of breath ) [16] . Nevertheless , a significant fraction of European countries still adopts their own clinical case definition to compile seasonal influenza surveillance weekly reports . S2 Table highlights the existing issue in the heterogeneity of the ILI case definition in Europe [16–18] . In general , differences in seasonal influenza epidemics across European countries are characterised by heterogeneity in sentinel systems , climatic conditions , human mobility systems , as well as social contacts [19 , 20] . The result is a consequent heterogeneity in the prevalence of the disease among the population in the various countries which can present differences in severity during the same influenza season . This diversity makes it hard to have a unified , one-fits-all approach to influenza surveillance , let alone a unified ILI definition . Moreover , the ILI definition might change over time even for national sentinel systems [21] . For example , in Italy , the National Institute of Health ( Istituto Superiore di Sanità ) adopted the ECDC definition only in 2014 [22] . France is a peculiar example as it had double surveillance ( ILI and ARI ) up till 2014 ( Casalegno et al . [3] assessed the performance of various influenza case definitions in France between 2009-2014 ) . Mandl et al . [2] explicitly addressed the variation in the definition of ILI over time . In recent years the availability of novel digital data streams has given rise to a variety of non-traditional approaches for monitoring seasonal influenza epidemics [23–25] . Such new digital data sources can be exploited to capture additional surveillance signals that can be used to complement GPs surveillance data [26–29] . In this context , some so-called participatory surveillance systems have emerged in several countries around the world with the aim of monitoring influenza circulation through Internet reporting of self-selected participants [30–32] . One of these systems , the Influenzanet project [30] , has been established in Europe since 2011 and it is now present in ten European countries . In this study , we excluded from the analysis the country of Sweden , due to the fact that the Swedish cohort is solicited upon invitation when required and not on an annual basis [33] . The system relies on the voluntary participation of the general population through a dedicated national website in each country involved in the project . Data are obtained on a weekly basis through an online survey [34] where participants are invited to report whether they experienced or not any of the following symptoms since their last survey: fever , chills , runny or blocked nose , sneezing , sore throat , cough , shortness of breath , headache , muscle/joint pain , chest pain , feeling tired or exhausted , loss of appetite , coloured sputum/phlegm , watery/bloodshot eyes , nausea , vomiting , diarrhoea , stomach ache , or other symptoms . Differently , from most traditional surveillance systems , this participatory form of online surveillance allows the collection of symptoms in real-time and directly from the general population , including those individuals who do not seek health care assistance . The list of proposed symptoms has been chosen to include the various ILI definitions adopted by national surveillance systems in Europe and , at the same time , to get a comprehensive list of symptoms that could be clearly articulated and understood by participants and would allow the detection of various circulating flu-related illnesses . Even though participatory systems generally suffer from self-selection biases , causing the sample to be non-representative of the general population [35] , previous works have shown that the web-based surveillance data collected by Influenzanet can provide relevant information to estimate age-specific influenza attack rates [36 , 37] , influenza vaccine effectiveness [34 , 38 , 39] , risk factors for ILI [39–41] , and to assess health care seeking behaviour [39 , 42] . Moreover , it has been largely demonstrated that weekly ILI incidence rates computed from the web-based surveillance data by applying the ECDC case definition to the set of self-reported symptoms correlate well with the weekly ILI incidence reported by GPs surveillance [37 , 39 , 43] . An additional advantage of collecting symptoms directly from individuals among the general population in the various Influenzanet countries is that it is straightforward to compare the prevalence and the temporal dynamics of specific symptoms or groups of symptoms from one country to the other . In a previous work focused on France [44] , the authors proposed population-level indicators based on self-reported symptoms and analysed crowdsourced incidence estimates comparing them to official estimates provided by sentinel systems . In this work , we propose an approach that aims at addressing the heterogeneity of seasonal influenza epidemiological signals in the various European countries , focusing on the individual symptoms collected directly from the general population . The goal is to develop a mathematical framework able to extract , in an unsupervised fashion , the groups of symptoms that are in good correlation with the ILI incidence , as detected by traditional surveillance systems for each country without imposing an a priori a specific ILI case definition . By using the daily occurrence of symptoms in form of matrix , we employ an approach based on Non-negative Matrix Factorization ( NMF ) [45] , to extract latent1 features of the matrix that correspond to linear combinations of groups of symptoms . We assume that a specific combination of reported symptoms is the symptomatic expression of one or more illnesses experienced by the participants , i . e . of the syndromes affecting the individual . We can then select those groups of symptoms that better correlate with the sentinel-based ILI incidence , which will become our best approximation for the actual influenza-like illness signal for a specific country . The overall encouraging results suggest that such methodology can be employed as a near real-time flexible surveillance and prediction tool not constrained by any disease case definition . Thus , it can be employed to monitor a wide range of symptomatic infectious diseases or to nowcast the influenza trend , to help to devise public health communication campaigns .
This study was conducted in agreement with country-specific regulations on privacy and data collection and treatment . Informed consent was obtained from all participants enabling the collection , storage , and treatment of data , and their publication in anonymized , processed , and aggregated forms for scientific purposes . In addition , approvals by Ethical Review Boards or Committees were obtained , where needed according to country-specific regulations . In The United Kingdom , the Flusurvey study was approved by the London School of Hygiene and Tropical Medicine Ethics Committee ( Application number 5530 ) . In France , the Grippenet . fr study was approved by the Comité consultatif sur le traitement de l’information en matiére de recherche ( CCTIRS , Advisory committee on information processing for research , authorization 11 . 565 ) and by the Commission Nationale de l’Informatique et des Libertés ( CNIL , French Data Protection Authority , authorization DR-2012-024 ) . In Portugal , the Gripenet project was approved by the National Data Protection Committee and also by the Ethics Committee of the Instituto Gulbenkian de Ciência . In general , the inclusion criteria of participants in the data analysis vary depending on the specific aim of the study [35 , 39 , 49 , 50] . In our case , we included only the individuals registered on the Influenzanet national platforms who filled in at least one Symptoms Questionnaire ( hereafter referred to as “survey” ) per season . This was done to focus the analysis on participants for which we have some information . We had to necessarily exclude individuals who have registered on the platforms but who have not submitted any symptoms survey during any influenza season . This corresponds to the exclusion of 0 . 3% of the registered participants . Moreover , to reduce the noise due to low participation rates at the beginning of the data collection of each influenza season , we consider as starting point the first week for which the number of surveys corresponded at least to 5% of the total number of the surveys filled during the week with the highest participation for that season . This refers to the fact that at the beginning of the season , which is a period when the epidemic is still well below the epidemic threshold , the participation ( i . e . the number of symptoms surveys ) is rather low and therefore the signal to noise ratio can be very low too . Furthermore , we included only one survey per each week—the latest one—if more than one survey was submitted during the same week by the same participant . This exclusion corresponds to a small fraction of discarded surveys , approximately 5% of the total number of surveys; moreover , the distribution of the discarded symptoms and the submission time of the dropped surveys , are homogeneous2 . This exclusion criterion is essential to express the number of self-reported symptoms as probabilities in the final ILI syndrome emerging from our framework and to interpret the aggregation of symptoms as an “incidence” . S1 Table in the supporting information presents descriptive statistics for each country , namely: ( i ) the number of seasons analysed , ( ii ) the average number of participants per season , ( iii ) the average number of weekly surveys per season , ( iv ) the average percentage of surveys with at least one symptom , ( v ) the average number of surveys per participant per season and ( vi ) the average number of weeks within a single season . In this section , we describe the methodology employed to extract the latent features from the self-reported symptoms collected by the various Influenzanet platforms of the participating countries . Our approach relies on the assumption that a specific group of self-reported symptoms corresponds to the symptomatic expression of one or more illnesses , hereafter called syndromes , circulating among the population sample of Influenzanet . In our study we consider the 18 symptoms presented in the weekly Symptoms Questionnaire plus an additional symptoms-related variable , called “Sudden onset” , referring to the sudden appearance of symptoms , typically over the course of the previous 24 hours ( see Table 1 ) . This totalizes 19 symptom variables that we hereafter designate interchangeably as “symptoms” . The symptoms were treated as binary boolean variables having value 1 if the symptom is present and 0 if the symptom is absent . We then aggregated the reported symptoms across all participants to build a matrix X = [xij] , whose elements contain the occurrences of each symptom j ∈ {1 , ‥ , J} during each day i ∈ {1 , ‥ , I} . In other words , each element of the matrix corresponds to the number of times each symptom has been reported on each day of the influenza seasons under study . The result is a high-dimensional sparse matrix that can be linearly decomposed through a Non-negative Matrix Factorization ( NMF ) technique [45] . We opted for NMF since its non-negativity constraint offers the advantage of a straightforward interpretation of the results as positive quantities that can then be associated with the initial symptoms . This approach can be considered as a “blind source separation” problem [51] in which neither the sources nor the mixing procedure is known , but only the resulting mixed signals are measured . In our case , the time series corresponding to the daily symptoms counts are measured by the Influenzanet platforms and can be considered as the result of a linear mixing process driven by unknown sources , i . e . the latent syndromes . In the following we will use interchangeably the terms syndrome , source or component . According to this consideration , each element xij of the matrix X can be expressed as follows: x i j = ∑ k ∈ { 1 , ‥ , K } w i k h k j + e i j , ( 1 ) where the coefficients hkj describe the set of the unknown K sources , the factor wik represents the time-dependent mixing coefficients , and the terms eij correspond to the approximation error . The mixing equations Eq ( 1 ) can be equivalently expressed in matrix notation as: X = W H + E ( 2 ) where: W = [ w i k ] , H = [ h k j ] , E = [ e i j ] ( 3 ) It is worth stressing that in this representation the matrix X is known , while the matrices W and H are unknown and determined by the NMF algorithm . In particular , we used a variation of the NMF algorithm that minimizes the Kullback-Leibler divergence loss function [52] defined as follows: argmin W , H ∑ i , j x i j log ( x i j x ^ i j ) - x i j + x ^ i j , ( 4 ) where: x ^ i j = ∑ k w i k h k j . ( 5 ) To minimise the Kullback-Leibler divergence loss function , we adopted the multiplicative update rules described in [53] . Note that different initialisation of the matrices W and H might lead to different local minima , making the interpretation of the results not straightforward . To overcome this issue , we used an initialization technique called Non-negative Double Singular Value Decomposition [54] , that is based on a probabilistic approach equivalent to the probabilistic latent semantic analysis ( pLSA ) [55] , employed in the context of semantic analysis of text corpora . Since the two approaches of NMF and pLSA are equivalent ( see [56] for more details ) , the results of our matrix decomposition can be probabilistically interpreted as a mixture of conditionally independent multinomials , that we call p ( i , j ) . We can then write: π ( i , j ) ≈ p ( i , j ) = ∑ k p ( k ) p ( i , j | k ) = ∑ k p ( k ) p ( i | k ) p ( j | k ) , ( 6 ) where: π ( i , j ) = x i j / N , N = ∑ i , j x i j ( 7 ) and N is the total number of symptoms counts . According to Eq ( 6 ) , the total number of symptoms counts will be proportionally split among K latent sources according to p ( k ) , which is the probability to observe a specific component k; p ( i|k ) is the probability to observe a component k in a day i and p ( j|k ) is the probability to observe a specific symptom j in a component k , and they can be expressed as follows: p ( i | k ) = w i k / ∑ i w i k , ∑ i p ( i | k ) = 1 , p ( j | k ) = h k j / ∑ j h k j , ∑ j p ( j | k ) = 1 , p ( k ) = ∑ i w i k ∑ j h j k / N , ∑ k p ( k ) = 1 . ( 8 ) At this point , Eq ( 8 ) allows to determine the probability p ( i , k ) that , rescaled on the total number of symptoms counts N , yields the desired decomposition procedure , yik , which represents the contribution of a specific component k in a day i , given by the following expression: y i k = N p ( i , k ) = N p ( k ) p ( i | k ) ( 9 ) Thus , the final step in our approach is to determine the optimal number of components kmin to be used for the decomposition . A natural upper bound for k would be the total number of symptoms , i . e . 19 . We need to determine the number of components with the best trade-off between a model that best approximates the original matrix X and at the same time does not overfit the data . Each time we minimize the loss function Eq ( 4 ) for a specific number of components k , we obtain a candidate decomposition . To determine the best decomposition , we use an approximated model selection criterion , known as the Akaike Information Criterion ( AIC ) [57] . In particular , we employ the corrected version of the Akaike Information Criterion ( AICc ) proposed in [58] , valid for finite sample sizes . For each of the candidate decompositions generated by the various values of k , we estimate the value of AICc ( k ) , expressed as: A I C c ( k ) = - 2 L ( k ) + 2 P + 2 P ( P + 1 ) N - P - 1 , ( 10 ) where L ( k ) is the log-likelihood of the model with k components , defined in [56] as: L ( k ) = ∑ i , j x i j logp ( i , j ) . ( 11 ) P is the number of parameters of the model defined as: P = K ( I + J - 2 ) - 1 , ( 12 ) where K is the upper bound for the number of components , I is the total number of days and J is the total number of symptoms . The best candidate decomposition is the one that minimizes Eq ( 10 ) and we denote it as AICc ( kmin ) . The final result is a model , that we call y i k m i n , consisting of kmin components that best approximate the original matrix X . We applied the aforementioned framework to the data collected by the Influenzanet platforms in nine European countries throughout six influenza seasons ( from 2011-2012 to 2016-2017 ) . For each country , we applied the decomposition algorithm to the symptoms’ matrix X as represented in Eq ( 2 ) and , based on the AIC , we obtained the “optimum” number of components , kmin , for the decomposition . The daily counts of the emerged components are eventually aggregated weekly to allow the comparison with the weekly incidence reported by the traditional GPs surveillance . Among the kmin latent components , i . e . syndromes , extracted for each country , we identified the one that correlates better with the time series reported by the traditional GPs surveillance . In the following , we denote this component as IN_NMF . This component corresponds to the combination of symptoms that more closely represent the ILI time series recorded by the traditional surveillance , and hence , it can be used to build a data-driven , unsupervised ILI case definition , which is the ultimate goal of this study . To further evaluate the IN_NMF signal selected for each country , we also computed the Pearson correlation between: ( i ) the IN_NMF and the time series obtained by applying the ECDC case definition to the Influenzanet data ( hereafter called IN_ECDC ) ; ( ii ) the IN_NMF and the ILI incidence reported by the national surveillance systems per country ( hereafter called GP ) ; and ( iii ) the IN_ECDC and the GP . The reported correlations refer to the time series over the entire period analysed ( 2011-2017 ) . Additionally , we explored the predictive power of the proposed methodology in the following way: first , we trained the NMF decomposition framework with Influenzanet data only from 2011 to 2016 and then , we employed the resulting symptom weights to infer the weekly IN_NMF estimates during the 2016-2017 season . To assess the quality of this signal , we evaluated the Pearson correlation of the forecasted IN_NMF time series for 2016-2017 with both the GP time series and the IN_ECDC time series . Moreover , to broaden the scope of our framework in identifying syndromes not related to ILI ( e . g . gastrointestinal versus respiratory ) , we employed it to identify the syndrome related to gastrointestinal episodes by performing the Pearson correlation with data provided by the traditional official surveillance in France . We focused on the case of France due to the immediate data availability from the official surveillance . The Réseau Sentinelles in fact comprises a unique program of data collection about gastrointestinal illness episodes [59] . The identified component is denoted as IN_Gastro . For the entire analysis and simulations we used the Python programming language ( Python Software Foundation , version 2 . 7 , https://www . python . org/ ) .
S1 Fig in the supporting information depicts an exploration on the relative AIC values of a series of candidate models ( AICc ( k ) − AICc ( kmin ) , with k ∈ [1 , 6] ) , estimated according to Eq ( 10 ) . For the majority of the countries , the optimal decomposition consisted of kmin = 2 components , with the exceptions of the Netherlands and Belgium with kmin = 3 , and France with kmin = 4 . S2 , S3 , S4 , and S5 Figs in the supporting information depict for each country the respective time series of all the emerging kmin components and their symptoms composition . The component selected by our framework is highlighted by a blue square . These results show how our approach is capable of taking into account differences in ILI definition between countries since we can select the components that best correlate with the national ILI signal . In the left panel of Fig 1 , the IN_NMF component for each country is shown in comparison to the ILI signal as recorded by the traditional surveillance , GP . To allow for visual comparison , the IN_NMF time series has been rescaled on the GP time series with a fixed scaling factor . Specifically , the IN_NMF has been rescaled on the highest peak among all the GP time series for each country , hence the lower peak of the IN_NMF for the other peaks of the GP time series . Consequently , the performance of the selected ILI component cannot be evaluated in terms of amplitude and error with respect to the peak estimate . In the right panel of Fig 1 , the break-down of symptoms for each country’s IN_NMF component is expressed in terms of probabilistic contributions , denoted as p ( j|k ) , as described in Eq ( 6 ) . In terms of symptoms’ composition , IN_NMF appears to be stable across the various countries and consistent with the expected set of symptoms clinically associated with ILI . The top contributing symptoms are fever , chills and feeling tired , often reported in combination with a sudden onset of symptoms . Notably , each of these top three symptoms contributes for about 10% or more of the overall component composition . This is consistent across all the nine countries and it is the most important result of this study since it represents the basis towards the development of a common ILI definition . Small heterogeneities in the component composition across countries are most likely due to differences in the ILI case definitions used by sentinel doctors in each country which are reflected in the data that we use as ground truth . In principle , this issue might be overcome by using seroprevalence data as ground truth . For the sake of comparison , we have examined how our framework performs with respect to other similar approaches . For example , Goldstein et al . [60] have used two inference methods to estimate incidence curves from symptoms surveillance data . The first method essentially assumes that the distribution of symptoms is known . In our case , we have no such assumption; instead , we extract the symptoms and their probabilistic distribution from the observed data without making any a priori assumption on the distribution of symptoms . The second inference method proposed by Goldstein et al . [60] is closer to our framework and falls under the umbrella of the term “blind source separation” . The Non-negative Matrix Factorization can be formulated as an expectation-maximization problem [61] . The difference with our approach is that they assume as an initial condition that the expected weekly incidence is equal to 1 for each infection in their survey sample . Their approach is sensitive to the ratio of flu/non-flu distribution while NMF manages to overcome this problem . Table 2 reports all the Pearson correlations between the different time series as mentioned in the Data Analysis section . For all countries , the correlation between the IN_NMF component and the IN_ECDC is very high , ranging from 0 . 82 to 0 . 92 ( row ( i ) ) , thus showing that the IN_NMF signal captures symptoms highly compatible with those present in the ECDC ILI definition applied to the Influenzanet data . However , by carefully examining rows ( ii ) and ( iii ) , we note slight variations per country . For the Netherlands , Belgium , and Ireland the ILI incidence reported by the traditional surveillance ( GP ) was more strongly correlated with the IN_NMF component , than with the ILI incidence obtained by applying the ECDC ILI definition to the Influenzanet data ( IN_ECDC ) . For the United Kingdom , Spain , Denmark , and Portugal , the IN_NMF components perform equally well as the IN_ECDC . For Italy and France , the IN_NMF component had a slightly lower correlation ( about 11% and 7% less respectively ) with the traditional surveillance data ( GP ) than the IN_ECDC . Ireland is the only country for which we obtain a low correlation between the traditional surveillance data ( GP ) and both the IN_NMF and IN_ECDC , probably due to the limited number of participants in Influenzanet ( see S1 Table in the supporting information ) . Despite this , the IN_NMF performs much better than the IN_ECDC in capturing the ILI incidence trend in Ireland ( 0 . 38 vs 0 . 23 ) . This variation in performance is not an issue for the goal of this work since our focus is on paving the way towards a common cross-country ILI definition rather than finding the perfect signal that correlates best with the traditional national surveillance . Also , the loss in performance of IN_NMF vs GP with respect to IN_ECDC vs GP for Italy and France is only a small percentage . One might argue that , since it has been observed that people tend to go to the doctor if their symptoms are more severe or if the duration of the disease is longer [62] , the high correlation between the IN_NMF time series and the GP time series might be attributable to the fact that participatory surveillance only captures individuals with perceived severe symptoms , who did visit a doctor for their illness . Unfortunately , we cannot assess the severity of self-reported symptoms , but we can assess the fraction of participants who claimed they have visited a healthcare provider for their symptoms and , in line with previous studies , we found that the vast majority of participants did not seek medical consultation . Specifically , the percentages of participants who did seek medical consultation per country are: NL 12% , BE 22% , IT 23% , FR 26% , UK 14% , ES 17% , PT 17% , DK 11% , IE 16% . Moreover , to investigate the performance of our framework with respect to healthcare seeking behaviour , we employed two different approaches . First , we trained our framework only with the subset of self-reported symptoms from participants who consulted a medical doctor for their symptoms , obtaining the following Pearson correlations with the GP time series: NL 0 . 83 , BE 0 . 82 , IT 0 . 87 , FR 0 . 92 , UK 0 . 88 , ES 0 . 82 , PT 0 . 82 , DK 0 . 69 , IE 0 . 51 . Secondly , we trained our framework only with the subset of self-reported symptoms from participants who did not consult a medical doctor for their symptoms , obtaining the following Pearson correlations: NL 0 . 77 , BE 0 . 59 , IT 0 . 69 , FR 0 . 78 , UK 0 . 72 , ES 0 . 54 , PT 0 . 48 , DK 0 . 64 , IE 0 . 29 . We notice that since by default our framework selects as ILI component the one that best correlates with the official surveillance , the IN_NMF signal emerged represents better the data reported by the official surveillance systems . Unsurprisingly , the correlations are higher when we compare the same population of individuals who did seek medical consultation for their illness . On the other hand , it is of extreme importance that our framework is capable of extracting a relevant signal in the latter case since the population of individuals who do not seek healthcare is complementary to the one depicted by the official surveillance data . Finally , in order to assess the impact of the exclusion criterion for which we do not take into account duplicate reports from the same individual in a single week , we have determined the mean percentage of the symptoms discarded per country: NL 0 . 04% , BE 0 . 03% , IT 0 . 09% , FR 0 . 06% , UK 0 . 16% , ES 0 . 05% , PT 0 . 12% , DK 0 . 03% , IE 0 . 10% . Indeed , the duplicate report exclusion corresponds to a small number of symptoms discarded each week and the distribution of all discarded symptoms is homogeneous . The results of the prediction analysis described in the Data Analysis section are shown in S6 Fig . The fourth row of Table 2 ( iv ) reports the correlations of the forecasted IN_NMF time series and the national surveillance for the season 2016-2017 ( GP ) . The correlation between the two time series is good for all the countries , ranging from 0 . 60 to 0 . 85 . In supplementary information we depict the results of the prediction analysis described in the Data Analysis section . As already stated above , for the sake of visual comparison , the IN_NMF time series has been rescaled to the highest peak of the GP time series for each country , hence the lower peak for the other peaks . Consequently , the two time series cannot be evaluated in terms of amplitude and error . In Table 2 row ( v ) , we also report the correlation between the forecasted IN_NMF time series and the IN_ECDC time series emerged from applying the ECDC definition to the Influenzanet data for the season ( 2016-2017 ) . Also , in this case , the predicted trend of the ILI component have high correlations , ranging from 0 . 59 to 0 . 93 . Even if the focus of the paper is on the possibility of extracting a symptoms-based data-driven definition of ILI that is country specific , the forecasting capabilities of the framework represent an additional strengthening factor ( the forecasting potential of using participatory surveillance data , in combination with additional epidemiological signals has also been explored in a previous paper [29] ) . To further assess the robustness of the forecasts produced by the NMF framework , we have compared their accuracy with respect to a null model in two different ways . In the left panel of Fig 2 , we show the time series for the incidence of acute diarrhoea episodes ( GP_Gastro ) as detected by the official national surveillance in France , and the time series of the syndrome identified by our framework ( IN_Gastro ) . The Pearson correlation between the extracted syndrome and the official surveillance data is 0 . 66 . In the right panel of Fig 2 we depict the probabilistic contribution of each symptom to the IN_Gastro syndrome . Emerging symptoms , in this case , include also stomach ache , diarrhoea , and vomiting , which are in line with our expectations . Even if respiratory symptoms like runny nose or sneezing are also present , the contribution of fever and chills ( which were the main contributors to the IN_NMF signal ) is almost negligible . This suggests a rather good capability of our framework in discriminating between different syndromes . Despite limitations of the data availability , these preliminary findings indicate that the latent components of the decomposition , not related to ILI , may express syndromes related to allergies , common-cold or gastroenteritis . Understandably , additional adequate surveillance data are required to make a firm statement and reach a robust interpretation of the syndromes . Previous works have also focused on detecting gastrointestinal symptoms circulating among the general population through digital unstructured data [33 , 63–65] from participatory surveillance , big data , such as Twitter , as well as national pharmacy sales data . These examples show how crowdsourced digital health-related data , as well as passive digital traces generated on the web by individuals from the general population , can complement traditional and syndromic surveillance systems to estimate the circulation of gastrointestinal syndromes . This is particularly important because only a fraction ( about a third ) of individuals who reported gastrointestinal symptoms in France also declared that they visited a doctor . The NMF framework applied to the subset of data from participants who did not visit a doctor for their symptoms selected a component whose correlation with official surveillance data is 0 . 67 ( with respect to a correlation of 0 . 66 when using all the data ) . This shows that people tend to visit a doctor rarely and probably only if their symptoms are severe . The NMF framework is capable of providing robust results even if we focus the analysis only on those individuals who did not visit a doctor , for which we can safely assume that their symptoms were not severe . This approach has several limitations . As far as data are concerned , crowdsourced digital data are intrinsically biased due to the fact that the participants are self-selected and not representative of the general population , as extensively explored in a previous work [35] . However , such sample biases do not affect the robustness and accuracy of the epidemiological signal detected through participatory surveillance [37 , 39 , 43] . Previous works have shown that selecting groups with specific reporting patterns or combining data sources can improve the representativeness [28 , 66 , 67] . Extending this study , we will incorporate in our framework the user attributes to account for selection biases . Other issues could rise from the variable reporting behaviours along the season , individuals’ interpretation of the terms used for surveillance , and the correctness of their self-assessments . Some of these issues have been addressed partially in previous works [10 , 40 , 44 , 50 , 68] . In our approach , we assume that self-reported symptoms are consistent since Influenzanet data have been already proven to be accurate and reliable for ILI surveillance , even without providing any clinical confirmation . However , we are aware that one of the criticisms of online participatory surveillance is the lack of virological confirmation of influenza cases that would instead help to better assess the actual circulation of influenza in the population . To this respect , a pilot study has been developed in the United Kingdom by the national Influenzanet platform , called Flusurvey , which demonstrates that self-swabbing can be applied to an online cohort to conduct virological laboratory testing [69] . Moreover , in this work , we have not compared the performance of other machine learning algorithms besides NMF since this would go beyond the scope of this paper . Future work could explore the performance of other methods and clustering algorithms . Among the many algorithmic choices , LDA could be employed in a similar framework , since PLSA is simply a special case of LDA and Faleiros et al . [70] showed that indeed NMF with Kullback-Leibler divergence approximates the latent Dirichlet allocation ( LDA ) model under a uniform Dirichlet prior distribution . Finally , there are inherent socio-economic biases in influenza surveillance systems [71] due to the fact that in some countries traditional surveillance is based on primary healthcare which may be biased towards population with higher socioeconomic status . Even additional digital unstructured data sources are more representative of these population groups , thus even combining traditional and non-traditional data sources might fail in mitigating biases towards more at-risk groups .
The practice of seasonal influenza surveillance is affected by a lack of a common case definition for influenza-like illness across countries . Moreover , the seasonal influenza epidemics in the various European countries present a high degree of heterogeneity . To improve seasonal influenza surveillance beyond these issues , we propose an unsupervised probabilistic framework based on self-reported symptoms collected daily through a network of participatory web-based influenza surveillance platforms in Europe called Influenzanet . Our approach , which relies on a Non-negative Matrix Factorization of the daily symptoms matrix , is capable of producing an epidemiological signal that does not rely on a specific a priori case definition and that follows the temporal trend of influenza-like illness closely as detected by the traditional sentinel doctors surveillance in each country . The emerging signal successfully captures the ILI incidence trend estimated by the national surveillance data for all the nine countries included in this study . We also demonstrate that the proposed approach can be employed to forecast the forthcoming ILI incidence . Additionally , the proposed approach has the potential to be used to identify other illnesses , as shown here for gastrointestinal syndromes , although additional traditional surveillance data is needed to validate the generalisability of our framework . We can thus conclude that there is great potential in using symptoms directly collected from the general population to inform unsupervised algorithmic approaches aimed at detecting circulating bouts of illnesses without imposing an a priori case definition . The standardized technological and epidemiological framework and the ability to monitor symptoms from the general population , including individuals who do not seek medical assistance , provided by the Influenzanet participatory surveillance platforms , are what enables the application of unsupervised algorithmic approaches such as the one presented in this work . In the next future , we will include data from virologically confirmed influenza cases as ground truth to enhance the specificity of our framework . Regarding the forecasting capabilities of the framework , approaches from existing research on participatory flu surveillance suggest that the integration of real-time official data sources with the crowdsourced digital ones [72] [73] provide better forecasting performance . In our case , the weekly integration of sufficient traditional surveillance data in the framework could lead to a near-real-time selection of the component that better represents the symptoms in the ILI syndrome circulating among the general population . Finally , the flexibility provided by the participatory surveillance platforms in terms of symptoms that can be collected from the general population enables the possibility to extend the framework to other diseases , provided that traditional surveillance data are available to train the framework .
|
This study suggests how web-based surveillance data can provide an epidemiological signal capable of detecting the temporal trends of influenza-like illness without relying on a specific case definition . The proposed framework was able to forecast quite accurately the ILI trend of the forthcoming influenza season based only on the available information of the previous years . Moreover , to broaden the scope of our approach , we applied it to the detection of gastrointestinal syndromes . We evaluated the approach against the traditional surveillance data and despite the limited amount of data , the gastrointestinal trend was successfully detected . The result is a near-real-time flexible surveillance and prediction tool that is not constrained by any disease case definition .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"european",
"union",
"pathology",
"and",
"laboratory",
"medicine",
"statistics",
"influenza",
"geographical",
"locations",
"ireland",
"mathematics",
"forecasting",
"signs",
"and",
"symptoms",
"infectious",
"disease",
"control",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"epidemiology",
"mathematical",
"and",
"statistical",
"techniques",
"people",
"and",
"places",
"infectious",
"disease",
"surveillance",
"france",
"diagnostic",
"medicine",
"fevers",
"disease",
"surveillance",
"viral",
"diseases",
"physical",
"sciences",
"europe",
"statistical",
"methods"
] |
2019
|
Unsupervised extraction of epidemic syndromes from participatory influenza surveillance self-reported symptoms
|
Acquisition of iron is necessary for the replication of nearly all bacterial pathogens; however , iron of vertebrate hosts is mostly sequestered by heme and bound to hemoglobin within red blood cells . In Bacillus anthracis , the spore-forming agent of anthrax , the mechanisms of iron scavenging from hemoglobin are unknown . We report here that B . anthracis secretes IsdX1 and IsdX2 , two NEAT domain proteins , to remove heme from hemoglobin , thereby retrieving iron for bacterial growth . Unlike other Gram-positive bacteria , which rely on cell wall anchored Isd proteins for heme scavenging , B . anthracis seems to have also evolved NEAT domain proteins in the extracellular milieu and in the bacterial envelope to provide for the passage of heme .
Vegetative forms of Bacillus anthracis replicate in vertebrate tissues and form spores once their host has succumbed to anthrax infection [1] . Spore contamination of food sources for vertebrates ensures pathogen dissemination to new hosts and reiterative replication cycles [2] . A hallmark of anthrax is its low infectious dose ( 25–50 spores can kill an animal ) and explosive replication of vegetative forms that accumulate to 1010 colony forming units ( CFU ) per gram of host tissue [3] . Spores are taken up by phagocytes and germinate in the phagosome [4] , [5] . Upon phagosome lysis , vegetative forms first multiply in the cytoplasm , however , once released into body fluids , bacilli resist phagocytosis and replicate in extracellular spaces [6] . Several key features enable the invasion and replication strategies of B . anthracis . First , spores are metabolically inert and survive in the environment for long periods of time until taken up by a new host [7] . To escape phagocyte killing , bacilli secrete lethal toxin and edema toxin that subvert the host immune system and implement host killing [8] . Elaboration of the dense poly-D-glutamic acid ( PDGA ) capsule endows vegetative forms with the characteristic trait of resisting phagocytosis [9] . PDGA is attached to peptidoglycan [10] , which functions as an exoskeletal scaffold for immobilization of proteins , carbohydrates and the S-layer , a two-dimensional crystalline protein array that encases vegetative forms [11] , [12] . Heme scavenging has been studied in Staphylococcus aureus , a Gram-positive pathogen phylogenetically related to B . anthracis , albeit that the envelope structure of staphylococci is comprised entirely of cell wall peptidoglycan with associated protein , teichoic acid and carbohydrate polymers [13] , [14] . Staphylococci elaborate neither PDGA capsule nor S-layers and their ability to retrieve heme from hemoglobin/haptoglobin relies on Isd proteins that are anchored to cell wall peptidoglycan [15] , [16] . The S . aureus isd locus ( isdA-isdB-isdCDEF srtB isdG ) is comprised of genes that encode cell wall anchored surface proteins ( IsdA , IsdB , IsdC ) , membrane protein ( IsdD ) , ABC transporter for import of heme ( IsdEF ) as well as heme mono-oxygenase ( IsdG ) [17] , [18] . The NEAT domain ( near iron transporter ) of staphylococcal envelope proteins ( IsdA , IsdB , IsdC ) enables scavenging of heme and passage of the iron containing compound across the cell wall envelope [15] , [16] , [19] , [20] , [21] . Heme passage relies further on sortase A-mediated deposition of IsdA and IsdB at the bacterial surface as well as sortase B-mediated immobilization of IsdC within the cell wall envelope [15] , [22] , [23] . The B . anthracis isd locus ( isdC isdX1 isdX2 isdE1 isdE2 isdF-srtB-isdG ) is comprised of eight open-reading frames with three putative transcriptional units , each flanked by a Fur-box consensus sequence ( Fig . 1A ) [24] , [25] . The smallest gene , isdX1 , harbors a NEAT domain and is conserved in all members of the Bacillus cereus group but absent from staphylococci , listeria and clostridia ( Fig . 1B ) . The largest gene , isdX2 , is also conserved and contains five NEAT domains . Here we report the first identification of a secreted heme-scavenging protein , IsdX1 , from Gram-positive bacteria . Further , we demonstrate that IsdX1 and IsdX2 acquire heme directly from hemoglobin and that this activity enables bacilli to scavenge iron from host hemoglobin under iron-limiting conditions . These findings indicate that unlike staphylococci , which rely on cell wall anchored Isd proteins for heme scavenging , B . anthracis seems to have also evolved NEAT domain proteins in the extracellular milieu and in the bacterial envelope to provide for the passage of heme .
The presence of cleavable N-terminal signal peptides and the absence of membrane or cell wall anchoring signals suggested that IsdX1 and IsdX2 may be secreted . To test this , B . anthracis was grown in the presence or absence of iron and bacterial cultures were fractionated to separate proteins secreted into the medium ( S ) from those targeted to the cell wall envelope ( C ) or located in membrane and cytoplasm lysate ( L ) ( Fig . 2A ) . When analyzed by immunoblotting with rabbit antiserum raised against purified recombinant IsdX1 , 15 kDa and 100 kDa ( including some degradation products ) immunoreactive species were detected under iron-limiting conditions . Wild-type bacilli secreted both the 15 and 100 kDa proteins , which represent IsdX1 ( predicted molecular mass 14 , 579 ) and IsdX2 ( predicted molecular mass 99 , 610 ) , as ΔisdX1 and ΔisdX2 mutant strains failed to express the former or the latter species , respectively ( Fig . 2A ) . Cross-reactivity of IsdX1 was not observed for other NEAT domain proteins , suggesting that IsdX1 and IsdX2 may share unique structural and functional properties ( data not shown ) . A portion of IsdX2 , but not of IsdX1 , was found in the cell wall fraction [24% ( ±9 ) of the total] , suggesting that IsdX2 may be partially associated with the envelope of bacilli . As a control , immunoblotting with antibodies against cell wall anchored ( IsdC ) , membrane ( SrtB ) and cytoplasmic ( L6 ) proteins was used to ensure proper fractionation of B . anthracis cultures ( Fig . 2A ) . The amount of IsdX1 or IsdX2 secretion was similar when bacilli were grown at 30°C or 37°C ( Fig . S1 ) . Taken together , these data indicate that IsdX1 and IsdX2 are synthesized and secreted when bacilli are exposed to iron-limiting conditions , as occurs during infection of vertebrate hosts . isdX1 with a C-terminal hexahistidyl tag was cloned under control of the IPTG inducible Pspac promoter in pLM5 and recombinant plasmid was transformed into bacilli . Affinity blotting of fractionated cultures revealed that bacilli harboring pisdX1-H6 , but not bacteria harboring pLM5 vector control , secreted IsdX1H6 into the extracellular milieu ( Fig . 2B ) . To test whether B . anthracis synthesize IsdX1 and IsdX2 during infection , we analyzed the serum of guinea pigs that had survived anthrax infections . Following subcutaneous infection with spores of B . anthracis strain Ames , guinea pigs suffer lethal anthrax infections over seven days , even when animals are inoculated with low doses of spores [26] . To ensure survival of guinea pigs , animals were treated with ciprofloxacin five days following infection , at a time when spores had germinated and vegetative bacilli replicated throughout host tissues . Two weeks following infection , animals were bled and serum samples examined for the presence of antibodies against purified recombinant GST-IsdX1 , GST-IsdX2 or a GST control . Immune sera from infected animals reacted with GST-IsdX1 , and GST-IsdX2 , but not with the GST control ( Fig . 2C ) . These data suggest that B . anthracis secretes IsdX1 and IsdX2 during infection when vegetative forms encounter iron-restrictive conditions , thereby stimulating specific host immune responses against these proteins . Several NEAT-domain containing proteins have been shown to bind heme , including B . anthracis IsdC ( B-IsdC ) [15] , [20] , [21] , [25] . To determine whether IsdX1 and IsdX2 display a similar property , both genes were cloned as translational fusions to the 3′ end of glutathione-S-transferase ( gst ) and GST-IsdX1/-IsdX2 purified from E . coli lysate by affinity chromatography ( Fig . 3AC ) . Both GST-IsdX1 and GST-IsdX2 eluted with red-brown color , indicative of an association with endogenous iron-porphyrin from E . coli ( Fig . 3AC insets ) [25] . We estimate that about 10% of purified GST-IsdX1/-IsdX2 was bound to heme [27] . GST-IsdX1 was dialyzed to remove heme , cleaved with thrombin and IsdX1 purified ( Fig . 3A ) . Binding of added heme to IsdX1 , as analyzed by spectrophotometry ( Soret absorbance at 404 nm ) [28] , was dose-dependent and quantifiable ( Kd 5 . 40±0 . 85×10−6 M ) ( Fig . 3B ) . Heme binding was only marginally increased by an increase in temperature ( Fig . S2 ) . IsdX2 also bound heme in a dose-dependent manner and did so more efficiently than IsdX1 ( Fig . 3D ) . The heme binding curve of IsdX2 yielded multiple inflection points , suggesting IsdX2 contains multiple binding sites for heme , presumably provided by its five NEAT domains . The complexity of the associations between IsdX2 and heme did not allow us to calculate a dissociation constant ( Fig . 3D ) . Together these findings indicate that IsdX1 and IsdX2 bind heme and may be involved in iron scavenging during anthrax infections . Hemoglobin ( Hb ) is the most abundant hemoprotein of mammals and several bacterial pathogens target this molecule to obtain iron during infection [29] , [30] . To examine whether hemoglobin serves as a source of heme for the presumed iron-scavenging activity of IsdX1 , we developed a simple experimental protocol . Glutathione-sepharose loaded with GST-IsdX1 was incubated with hemoglobin . The resin was then sedimented by centrifugation , separated from supernatant containing hemoglobin , washed and GST-IsdX1 eluted ( Fig . 4A ) . As a control ( C ) , hemoglobin was incubated with glutathione-sepharose that had been charged with GST and compared with GST-IsdX1 treated samples ( T ) ( Fig . 4BC ) . Following incubation with GST-IsdX1 , the heme-specific absorbance of hemoglobin at 404 nm was diminished , indicating that GST-IsdX1 had removed heme from hemoglobin ( Fig . 4B ) . GST-IsdX1 mediated removal of heme could also be observed by inspection of hemoglobin: the red-brown color of hemoglobin is cleared in GST-IsdX1 treated , but not in GST control samples ( inset , Fig . 4B ) . When analyzed by spectrophotometry , GST-IsdX1 displayed an increase in absorbance at 404 nm following its incubation with hemoglobin ( Fig . 4C ) . Inspection of glutathione sepharose sediment revealed red-brown pigmented GST-IsdX1 , whereas GST control samples remained clear ( inset , Fig . 4C ) . When analyzed by spectrophotometry , GST-IsdX1 displayed an increase in absorbance at 404 nm following its incubation with hemoglobin ( Fig . 4C ) . The abundance of hemoglobin in the supernatant samples was unchanged in the treated versus control reactions , indicating that the observed color and spectral changes were caused by heme transfer to IsdX1 ( Coomassie stained SDS-PAGE , Fig . 4B ) . We sought to develop a second measure for GST-IsdX1 removal of heme from hemoprotein . Apo-hemoglobin ( hemoglobin lacking heme ) was loaded with [55Fe]heme and radiolabeled hemoglobin was purified . [55Fe]hemoglobin was incubated with GST or GST-IsdX1 bound to glutathione-sepharose . As before , glutathione sepharose was sedimented by centrifugation and transfer of [55Fe]heme from hemoglobin was measured by scintillation counting as an increase in [55Fe]ionization ( Fig . 4D ) . Addition of increasing amounts of GST-IsdX1 , but not of GST , to [55Fe]hemoglobin led to increased [55Fe]ionization in sediment samples , until eventually all [55Fe]heme had been removed from hemoglobin ( Fig . 4E ) and transferred to GST-IsdX1 ( Fig . 4D ) . Serratia marcescens HasA represents the best established paradigm of bacterial hemophores [31] . Following its secretion via the Serratia type I pathway , 19 kDa HasA binds heme ( Ka 5×1010 M−1 ) [32] , [33] . Due to its high affinity , HasA retrieves heme from hemoglobin and , in turn , transfers heme to the HasR outer membrane receptor for heme transport across the bacterial envelope and into the cytosol [34] . To validate our heme-transfer assay as a method to measure heme transfer between proteins , we compared the ability of IsdX1 to acquire heme from hemoglobin with that of HasA . We purified GST-HasA from lysates of recombinant E . coli by affinity chromatography . Glutathione-sepharose was charged with each GST-HasA , GST-IsdX1 or GST and then incubated with hemoglobin . Resin was sedimented by centrifugation , separated from supernatant containing hemoglobin , washed and bound proteins eluted ( Fig . 5 ) . Eluate was analyzed for heme binding by measuring the absorption spectrum of GST-HasA , GST-IsdX1 and GST for heme . GST-HasA and GST-IsdX1 displayed a similar ability to remove heme from hemoglobin . Thus , it seems plausible that IsdX1 functions as a hemophore for B . anthracis heme scavenging . When analyzed by spectrophotometry for absorption at 404 nm , IsdX1 bound heme with an affinity significantly lower than the affinity of apo-hemoglobin for heme ( Ka>1011 M−1 ) [35] . We therefore considered the possibility that IsdX1 may retrieve heme from hemoglobin by a mechanism that involves physical contact between both proteins [36] . Surface plasmon resonance ( SPR ) spectroscopy was used to measure the presumed physical association between IsdX1 and hemoglobin [37] , [38] . Infusion of IsdX1 over hemoglobin coated chips produced a large spike in the local light refraction index ( RU ) , indicative of a physical interaction between IsdX1 and hemoglobin . This association was saturated within ∼180 seconds and , when deprived of further IsdX1 infusion ( arrow ) , decayed to near baseline RU values ( Fig . 6 , + heme ) . Infusion of IsdX1 over chips coated with apo-hemoglobin failed to reveal a physical association between both proteins ( Fig . 6 , − heme ) . Following removal of heme from hemoglobin by IsdX1 , additional infusion of heme over apo-hemoglobin produced holo-hemoglobin ( data not shown ) , suggesting the inability of IsdX1 to associate with apo-hemoglobin is not caused by the unfolding of this polypeptide . Physical interaction between IsdX1 and hemoglobin occurred in a dose-dependent manner that could be saturated as the concentration of IsdX1 increased ( Fig . S3 ) . Dissociation constants for the interaction between IsdX1 and hemoglobin are 7 . 33×10−6 M ( holo-hemoglobin ) and 9 . 43×10−3 M ( apo-hemoglobin ) . Thus , IsdX1 appears to bind directly to hemoglobin and , upon transfer of heme , dissociates from apo-hemoglobin . To examine the specificity of IsdX1 and IsdX2 for host hemoproteins , GST-IsdX1/-IsdX2 were incubated with excess hemoglobin and myoglobin , a monomeric globin abundantly present in muscle tissue [39] . As compared to hemoglobin , GST-IsdX1/-IsdX2 displayed little hemophore activity towards human myoglobin ( Fig . 7 ) , and similar results were observed when bovine or equine myoglobin was examined ( data not shown ) . These data suggest that during B . anthracis infection IsdX1 and IsdX2 most likely prefer hemoglobin over myoglobin as a heme source . Almost the entire IsdX1 polypeptide is comprised of its NEAT domain ( Fig . 3A ) . To test whether other NEAT domain proteins also display hemophore activity , GST fusions to S . aureus IsdC and B . anthracis IsdC were purified and compared to GST-IsdX1/-X2 ( Fig . 8 ) . All four hybrids were able to remove heme from hemoglobin . IsdX2 , which contains 5 NEAT domains , was 3 . 4 fold more efficient than IsdX1 and 7 . 25 or 12 . 6 fold more effective than B . anthracis IsdC or S . aureus IsdC . Also , hemoglobin was not sedimented in any of the reactions , suggesting a transient association similar to that observed for IsdX1 ( Fig . 8 , inset ) . Thus , the direct acquisition of heme from hemoglobin appears to be a general property of some NEAT domain proteins , albeit that IsdX1 and IsdX2 , when compared to IsdC , clearly display superior activity . This finding is compatible with their localization to the extracellular milieu , a site expected to optimize their interaction with hemoglobin . To examine whether the in vitro biochemical activity ascribed to IsdX1 and IsdX2 correlated with in vivo biological function , wild-type , ΔisdX1 , ΔisdX2 , and ΔisdX1/ΔisdX2 mutant B . anthracis strains were analyzed for growth in iron defined media ( IDM ) with hemoglobin as the only source of iron [40] . In the absence of added hemoglobin , all strains grew very poorly in IDM ( Fig . 9 ) . The addition of increasing amounts of hemoglobin allowed wild-type B . anthracis to grow with increasing rates ( Fig . 9 ) , indicating that bacilli can utilize hemoglobin as a source of iron . All three mutant strains ( ΔisdX1 , ΔisdX2 , and ΔisdX1/ΔisdX2 ) displayed a growth defect under iron-depleted conditions with hemoglobin as the sole iron source ( Fig . 9 ) . Whereas deletion of individual genes , isdX1 or isdX2 , caused a reduction in growth , these defects were exacerbated for the double mutant strain , which is unable to secrete IsdX1 or IsdX2 ( Fig . 9 and Fig . S4 ) . These data suggest isdX1 and isdX2 perform partially overlapping functions in the heme scavenging pathway of bacilli . Growth defects of ΔisdX1 and ΔisdX2 mutants were restored when bacilli were transformed with plasmids providing for IPTG inducible expression of each respective gene . Finally , all strains examined grew equally well in iron-replete media ( Fig . 9 , far right columns ) . Collectively , these experiments suggest that B . anthracis IsdX1 and IsdX2 function as secreted hemophores for heme-scavenging from hemoglobin .
The ability of mammalian organisms to sequester iron and limit its availability serves as a defense against microbial infection [41] . Iron is stored intracellularly , where ferric iron is complexed by ferritin or incorporated by ferrochelatase into porphyrin . The resulting product , heme , is bound by hemoproteins , e . g . hemoglobin or myoglobin [42] . Dedicated traffic systems for ferric iron ( transferrin ) or heme ( hemopexin ) transport iron in body fluids between tissues . A key feature that enables bacteria to replicate within their hosts is the production of siderophores , iron-sequestering compounds that scavenge iron from transferrin , and synthesis of cognate siderophore transport systems for the bacterial envelope [43] . Vertebrates , in turn , evolved defense mechanisms that exploit the bacterial requirement for iron by producing lipocalin , siderocalin or related proteins which sequester iron [44] . B . anthracis employs two siderophores to retrieve ferric-iron during infection , bacillibactin and petrobactin ( anthrachelin ) [45] , [46] . Petrobactin , enzymatically derived from 3 , 4-dihydroxybenzoate , spermidine and citrate via products of the asbA-F locus , is essential for B . anthracis growth , as mutations in asbA-F cause significant defects in the pathogenesis of anthrax [40] , [47] , [48] . Interestingly , this siderophore is resistant to sequestration by siderocalin , an immune protein which binds siderophores as a bacterial defense strategy [49] . B . anthracis has also evolved a scavenging pathway for heme that is encoded by the isd locus ( isdC-isdX1-isdX2-isdE-isdE2-isdF-srtB-isdG ) [25] . IsdC , a NEAT domain protein with C-terminal sorting signal , is anchored to cell wall peptidoglycan by sortase B ( SrtB ) [25] . IsdE-IsdE2-IsdF membrane transporter is thought to import heme into bacterial cells , while IsdG , a cytoplasmic monooxygenase , cleaves the tetrapyrrol of heme , thereby liberating iron [24] . Heme scavenging strategies of B . anthracis must take into account the unique envelope attributes of this pathogen . Bacilli evolved a thick murein sacculus comprised of peptidoglycan with attached envelope polymers: poly-D-glutamic acid ( PDGA ) capsule , carbohydrate polysaccharide , teichoic acid and proteins [12] . Further , bacilli elaborate S-layers , two-dimensional crystalline arrays of proteins bearing SLH domains that are immobilized by interaction with pyruvylated cell wall polysaccharide [11] , [50] . It is not certain that bacilli elaborate all envelope components at each stage of infection [51] . Nevertheless , explosive growth of B . anthracis and the accompanying need for nutrients likely demand that heme scavenging pathways must engage all structural components of the bacterial envelope . Here we report that B . anthracis secretes two polypeptides , IsdX1 and IsdX2 , into the extracellular milieu . The absence of a canonical sortase recognition motif in the C terminus of IsdX2 suggests it is not anchored to the cell wall by a sortase . Both proteins remove heme from hemoglobin , thereby enabling B . anthracis growth under conditions when hemoglobin is the sole source of iron . These findings , along with the data presented in Figures 4–8 , suggest one of the functions of the NEAT domain is the direct acquisition of heme from hemoglobin . How IsdX1 and IsdX2 bind heme is currently unknown; however , studies from other NEAT proteins suggest that heme-iron is ligated by a conserved tyrosine with high spin , five-coordinate geometry [20] , [52] , [53] . It seems unlikely that IsdX1 or IsdX2 deliver heme directly to the bacterial membrane , as the cell wall envelope cannot be penetrated by proteins . Instead , IsdX1 and IsdX2 probably transfer heme to other NEAT domain proteins at strategic positions throughout the bacterial envelope , a hypothesis consistent with their secretion into the surrounding milieu . In agreement with this conjecture , in silico analysis of the B . anthracis genome identified several genes encoding NEAT domain proteins with variable envelope locations: peptidoglycan linked IsdC [25] , [54] , BasJ positioned in the plasma membrane [55] , and BslK , an S-layer protein [56] . In contrast to the complex features of the envelope in bacilli , staphylococci , listeria and clostridia are much simpler and cannot elaborate a large capsule or S-layer [18] . Not surprisingly , these microbes are capable of scavenging heme with NEAT domain proteins that are exclusively immobilized in cell wall peptidoglycan . Heme scavenging pathways in Gram-negative bacteria have been studied in great detail . S . marcescens employs a type I secretion machine ( HasDEF ) and recognition of a C-terminal secretion signal to transport HasA across the bacterial double membrane envelope [32] , [57] , [58] . By virtue of its unique structure and affinity for ligand ( Ka 5×1010 M−1 ) , HasA retrieves heme from hemoglobin , myoglobin or hemopexin [30] , [59] , [60] , [61] and delivers the compound to HasR , the outer membrane receptor . Although HasR has much lower affinity for heme ( Ka 5×106 M−1 ) , the outer membrane receptor receives heme from HasA by a mechanism involving physical interactions between both proteins [60] , [61] . TonB ( HasB ) -ExbB-ExbD dependent relay then transfers heme from HasR across the periplasm , initiating subsequent import into the cytoplasm [62] . HasA production and secretion are regulated by an ECF type sigma factor ( HasI ) and its cognate anti-sigma factor ( HasS ) [63] . Biological activities of HasI/HasS are informed by reciprocal associations between HasA , HasR and heme [64] . Hemophore systems with similar design exist in Haemophilus influenzae [65] , Yersinia enterocolitica [66] , and Pseudomonas aeruginosa [67] , [68] , [69] . Pathogenic Neisseria spp . , on the other hand , elaborate outer membrane proteins that not only bind hemoproteins but also remove heme . IsdX1 represents the first secreted hemophore in Gram-positive bacteria , a finding that invites a functional comparison with HasA , the secreted hemophore of Gram-negative microbes [31] . Unlike HasA , which acquires heme from diverse hemoproteins such as myoglobin , IsdX1 appears to be specific for hemoglobin [30] . Further , whereas HasA seems to acquire heme from hemoglobin by virtue of its higher affinity for heme [31] , [60] , IsdX1 directly associates with hemoglobin for extraction of the heme . Finally , the structure of HasA is quite distinct from that of other NEAT-domain proteins [20] , [53] , [70] , [71] . These findings suggest that the molecular mechanism whereby IsdX1 acquires heme from hemoglobin must be distinct from that of HasA . While HasA delivers heme to outer-membrane receptors [60] , [61] , secreted components of the isd locus encoding NEAT domain proteins , such as IsdX1 , provide a versatile strategy for stealing heme that can be adapted to unique microbial envelope structures of Gram-positive pathogens . Whether these specific adaptations are important during infections caused by Gram-positive pathogens , e . g . B . anthracis , is a topic currently being explored in our laboratory .
B . anthracis strain Sterne 34F2 [72] and E . coli strains ( DH5α , XL1-Blue or K1077 ) were grown in Luria-broth ( LB ) or brain-heart infusion ( BHI ) ( Table S1 ) . Antibiotics were used for plasmid selection ( ampicillin 50 µg/ml , kanamycin 20 µg/ml ) . All reagents were purchased from Sigma unless otherwise noted . B . anthracis chromosomal DNA was extracted using the Wizard Genomic DNA Purification Kit ( Promega ) . The isdX1 gene ( BAS4443 ) of B . anthracis Sterne was deleted by allelic replacement with the temperature- sensitive pLM4 [26] . Briefly , 1 , 000 bp of 5′ and 3′ isdX1-flanking sequences were PCR amplified with primer pairs isdX1-EcoRI ( 5′-gatcgatcgaattgattttcattgagaatgataatc-3′ ) and isdX1-SacI ( 5′-gatcgatcgagctcttgtttaaacatatattcatcacc-3′ ) as well as isdX1-SacI ( 5′-gatcgatcgagctcgggaacagtattaaataattttc-3′ ) and isdX1-KpnI ( 5′-gatcgatcggtacccctctggttgtttctcttc-3′ ) . Following ligation , the 2-kb inset was cloned between the EcoRI/KpnI sites of pLM4 to create pLM4-ΔisdX1 . After transformation into BAS7 , bacilli were grown first at 30°C ( permissive temperature ) on LB/Km and then shifted to 43°C ( restrictive temperature ) , followed by growth at 30°C to induce plasmid loss , thereby generating BAS8 . DNA was analyzed for the presence of isdX1 by PCR and deletions confirmed by DNA sequencing . Deletion of isdX2 ( BAS4442 ) was achieved as previously reported [25] . The deletion of both isdX1 and isdX2 in the same strain was achieved via the procedure described above for ΔisdX1 using the ΔisdX1 5′ flank primers and the following 3′ flank primers: isdX2-SacI ( 5′-gatc gatcgagctcctagttcgtaaatatagagcagg-3′ ) and isdX2-KpnI ( 5′-gatcgatcggtaccccttgtacaagttc aacaatacc-3′ ) . Plasmid DNA was amplified in dam mutant E . coli strain K1077 prior to electroporation of bacilli [73] . Signal peptides of IsdX1 , IsdX2 , B-IsdC , and Sa-IsdC were replaced with glutathione S-transferase ( GST ) and recombinant proteins were purified by GST-affinity chromatography ( see Protocol S1 ) . Overnight B . anthracis cultures were incoculated into 2 ml of BHI ( + Fe ) or chelex-treated BHI ( − Fe ) supplemented with Ca2+ , Mg2+ , Mn2+ , and Zn2+ and incubated at 37°C for further growth [25] . Bacilli were sedimented by centrifugation at 10 , 000×g , washed twice with 1 ml of PBS ( pH 7 . 4 ) and fractionated as previously reported [25] . Samples were analyzed by immunoblot with αL6 , αSrtB , αIsdC , or αIsdX1 specific rabbit antisera ( 1∶1 , 000 ) , followed by mouse anti-rabbit HRP-linked antibody ( 1∶10 , 000 ) and ECL ( enhanced chemiluminescence , Pierce , Rockford , IL ) . By comparing the amount of secreted IsdX1 and IsdX2 to a known amount of recombinant purified IsdX1/X2 via immunoblot , we estimate that a 3 mL culture of B . anthracis containing an optical density of 1 . 0 will secrete 0 . 52±0 . 25 µg of total IsdX1 in 12 hours . This compares to 0 . 55±0 . 07 µg of total IsdX2 secreted under the same conditions . IsdX1 ( 20 µM ) or IsdX2 ( 1 µM ) were incubated in 50 mM Tris-HCl , pH 8 . 0 with or without hemin chloride ( 0 . 01–40 . 0 µM in 0 . 1 M NH4OH ) for 5 minutes at 25°C , followed by spectrophotometry ( 300–700 nm ) in a Varian Cary 50BIO instrument . Peak absorbance at 404 nm , characteristic of heme binding , was monitored following subtraction of a hemin-only reference cuvette value at each concentration . GST-IsdX1 ( 60 µM ) or PBS ( control ) was incubated with 50 µL of glutathione-sepharose ( Amersham ) for 30 min at 25°C , followed by 3 washes of 200 µL with PBS . Bovine hemoglobin ( Sigma H2500 ) was added to 60 µM ( monomer ) and the X1/Hb mixture was incubated for 30 min at 25°C . Reactions were centrifuged at 13 , 000×g to sediment glutathione-sepharose/GST-IsdX1 complexes , reactions washed three times with 200 µL of PBS and GST-IsdX1 eluted in 50 µL of 600 mM reduced glutathione ( pH 8 . 0 ) . Sediment ( GST-IsdX1 ) and supernatant ( hemoglobin ) were analyzed by spectroscopy and heme binding quantified by measuring absorbance at 404 nm . For [55Fe]heme transfer , reactions were prepared as indicated above except that the amount of GST-IsdX1 added varied from 0 . 1–140 µM ( see Protocol S1 ) . The amount of [55Fe]heme in the sediment ( GST-IsdX1/resin ) or supernatant ( hemoglobin ) was quantified in a Beckman LS-6000IC instrument ( Beckman-Coulter , Fullerton , CA ) . Percent amount of heme was calculated by dividing the counts in the sediment or supernatant by the total number of counts in each reaction multiplied by 100 . For the experiment presented in Fig . 7 , the heme-transfer assay was utilized with the concentrations of hemoglobin and myoglobin ( Sigma M0630 ) at 800 µM . Heme acquisition was calculated as follows: [ ( GST-IsdX1Abs . 404nm ) minus ( glutathione-sepharose ( background ) Abs . 404nm ) divided by [total inputAbs . 404nm] times 100 . IsdX1-hemoglobin interactions were measured with a BIAcore 3000 biosensor ( GE Healthcare ) via surface plasmon resonance ( SPR ) [37] , [38] . Hemoglobin , 180 pmol in HBS ( 10 mM HEPES , pH 7 . 4 , 0 . 15 M NaCl , 50 mM EDTA , 0 . 05% Tween 20 ) , was amine coupled to CM5 sensor chip at 25°C at a flow rate of 5 µL/min [74] . Hemoglobin injection was stopped once response was saturated at 2 , 100 RU and 50 µM IsdX1 in HBS was infused at 20 µL/min with a dissociation time of 300 sec at 25°C . Data were fit to a model of equimolar IsdX1-hemoglobin association with BIAevaluation version 4 . 1 . A dose-dependent response was observed over an IsdX1 concentration range of 3–50 µM . Bacilli from overnight cultures in 2 ml of LB+Km at 30°C were inoculated into IDM+Km [40] , grown for 12 hours at 30°C , bacteria harvested by sedimentation at 10 , 000×g , washed twice and then suspended in 1 ml IDM ( O . D . 4 . 0 ) . Aliquots ( 5 µL ) were inoculated into 150 µL IDM , with or without Hb ( 20 , 100 , or 500 µM ) using 96-well U-bottom plates ( Corning , Corning , NY ) . After 16 hours of incubation at 30°C , growth was assayed by plating 5 µL of a 1∶400 dilution of bacterial culture onto LB/ Km agar plates and colony forming units per mL ( CFUs/mL ) determined . For plasmid complementation , 1 . 5 mM IPTG ( final concentration ) was added to culture media . A list of accession numbers ( NCBI ) for genes in this study are as follows: isdX1 = YP_030690 , isdX2 = YP_030689 , b-isdC = YP_030691 , isdC = YP_001332076 .
|
Iron is an essential nutrient used by almost all organisms . Bacterial pathogens must acquire iron in order to grow inside mammalian hosts . The host , however , limits the availability of free iron , thereby providing an effective defense strategy against infection . In response , bacteria have evolved clever ways to subvert host sequestration of iron . In this work , we report that the causative agent of anthrax disease , Bacillus anthracis , produces two proteins ( IsdX1 and IsdX2 ) , which act to acquire iron complexed to heme , a co-factor of host hemoproteins such as hemoglobin . This activity is dependent on a conserved protein domain found in many Gram-positive bacterial pathogens and is necessary for growth of B . anthracis in low-iron environments . Our results yield a greater understanding of the mechanisms used by bacterial pathogens to subvert host defenses and provide an avenue for the development of antiinfectives that aim to block these strategies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"infectious",
"diseases",
"microbiology"
] |
2008
|
Bacillus anthracis Secretes Proteins That Mediate Heme Acquisition from Hemoglobin
|
The repeated rDNA array gives rise to the nucleolus , an organelle that is central to cellular processes as varied as stress response , cell cycle regulation , RNA modification , cell metabolism , and genome stability . The rDNA array is also responsible for the production of more than 70% of all cellular RNAs ( the ribosomal RNAs ) . The rRNAs are produced from two sets of loci: the 5S rDNA array resides exclusively on human chromosome 1 while the 45S rDNA arrays reside on the short arm of five human acrocentric chromosomes . These critical genome elements have remained unassembled and have been excluded from all Hi-C analyses to date . Here we built the first high resolution map of 5S and 45S rDNA array contacts with the rest of the genome combining over 15 billion Hi-C reads from several experiments . The data enabled sufficiently high coverage to map rDNA-genome interactions with 1MB resolution and identify rDNA-gene contacts . The map showed that the 5S and 45S arrays display preferential contact at common sites along the genome but are not themselves sufficiently close to yield 5S-45S Hi-C contacts . Ribosomal DNA contacts are enriched in segments of closed , repressed , and late replicating chromatin , as well as CTCF binding sites . Finally , we identified functional categories whose dispersed genes coalesced in proximity to the rDNA arrays or instead avoided proximity with the rDNA arrays . The observations further our understanding of the spatial localization of rDNA arrays and their contribution to the architecture of the cell nucleus .
Ribosomal RNAs ( rRNAs ) are essential components of the cell , and are encoded in the 5S and 45S ribosomal DNA ( rDNA ) arrays of higher eukaryotes [1–4] . The 5S rDNA array resides on chromosome 1 and encodes the 5S rRNA , whereas the 45S rDNA array resides on five human acrocentric chromosomes and encodes the 18S , 5 . 8S , and 28S rRNA components of the ribosome [5–7] . The nucleolus , the first recognized nuclear organelle , is the site of 45S rRNA transcription [1 , 2 , 4 , 8] . The lack of homology between the 5S rDNA and the subunits of the 45S rDNAs arrays reflect deep evolutionary separation . For instance , RNA polymerase I is exclusively dedicated to the transcription of the 45S rRNA , while RNA polymerase III transcribes the 5S rRNAs and tRNAs . The distinct RNA polymerase machineries required for transcription of 5S and 45S subunits are a conserved feature of yeasts , plants , fruit flies , and humans . Furthermore , distance to the nucleolus is thought to be relevant for global gene expression . For instance , proximity to the nucleolus can in some cases promote inactivation of certain RNA polymerase II transcribed genes [9] , although the observation has not been systematically tested across the genome . Finally , localization of the 5S array has been documented at the periphery of the nucleolus [9 , 10] , but also away from the organelle [11] , with a substantial fraction of cells showing 5S arrays that are localized elsewhere in the nucleus [10] . Uncovering physical contacts between the rDNA arrays and the rest of the genome can expand our understanding of nuclear architecture , nucleolar structure and function , and the mechanism of concerted copy number variation between 5S and 45S rDNA arrays . However , studies of nuclear architecture have largely excluded analyses of spatial interactions with the 5S and 45S rDNA arrays . Ligation-capture Hi-C sequencing technology [12–14] enabled a revolution in our understanding of nuclear organization with the identification of hundreds of topologically associated domains ( TADs ) . Human TADs span an average 900 KB each and display remarkable conservation with TADs identified in mice . TADs display , moreover , remarkable structural stability through development and when cells are perturbed in gene knockdown experiments [15 , 16] . On the other hand , deep sequencing of nucleoli led to the documentation of nucleoli associated DNA ( naDNA ) and the identification of nucleolus associated domains ( NADs ) [17–19] . While NADs display size variation spanning multiple orders of magnitude , they are generally large . NADs covering less than 0 . 1 MB are relatively rare with most NADs around 1 MB or larger . The domains encompass about 5% of the human genome , are represented in all chromosomes , and are now recognized to be stably associated with nucleoli . Analysis of rDNA interactions with Hi-C might provide a complementary approach to localize the rDNA in the nuclear space possibly informing nucleolar interactions with the genome at a different scale than those afforded by analysis of naDNA . Here we addressed the landscape of long-range rDNA interactions with 16 , 482 , 743 reads identified from a total of >15 billion ( 15 , 165 , 355 , 427 ) Hi-C reads in five cell types and two cell lines . The data enabled a map of long-range rDNA interactions at 1MB resolution , and the identification of segments displaying statistically significant differential contact density between cells . The map yielded a number of observations and suggest that the 5S and 45S arrays are not as spatially close as typically expected , yet share significant overlap with common contacts elsewhere . Finally , the data uncovered functionally coherent categories whose dispersed genes either coalesce in proximity to the rDNA arrays or avoid proximity with the rDNA arrays .
We investigated human Hi-C data for two cell lines and five cell types; the two cell lines represent the most replicated human Hi-C datasets to date , yet yielded a relatively small number of rDNA informative reads . For instance , we mined 5 , 356 , 990 , 189 high quality Hi-C reads in LCL to identify 13 , 528 , 436 reads with at least one end mapped to the 45S rDNA and 105 , 147 reads with at least one end mapped to the 5S rDNA ( S1 and S2 Tables ) . Similarly , for K562 cells , we mined 903 , 837 , 936 high quality Hi-C reads to identify 1 , 698 , 063 reads with at least one end mapped to the 45S rDNA and 47 , 691 reads with at least one end mapped to the 5S rDNA . This represents a 0 . 25% and 0 . 19% recovery rate of 45S rDNA reads in shotgun Hi-C in LCL and K562 , respectively . These numbers were substantially larger than the meager 0 . 002% and 0 . 005% recovery rate for 5S rDNA reads in LCL and K562 , respectively . Similar recovery rates were obtained with the other five cell types studied ( Table 1 ) . Overall , we uncovered 16 , 322 , 538 reads with at least one end mapped to the 45S rDNA and 160 , 205 reads with at least one end mapped to the 5S rDNA ( Table 1 ) . The mining effort illustrates the challenge in recovering rDNA information in shotgun Hi-C experiments . Nevertheless , the data revealed that rDNA contacts are dispersed across the entire genome , with segments differing in the density of rDNA interaction . The maps also revealed that naDNA and rDNA-contacts are not overlapping domains and likely reflect different attributes of the nucleolus/rDNA ( S1 Fig ) . Here we partitioned human autosomes ( Chr 1 to 22 ) into 2897 segments of 1MB , 2465 and 2658 of which had no evidence of containing a 5S or 45S pseudogene , respectively . Segments containing an rDNA pseudogene were disproportionately found adjacent to centromeric and telomeric regions , and were excluded from all further analyzes . Unsurprisingly , all 1MB segments across all chromosomes displayed evidence of rDNA contact ( S1 and S2 Figs ) . Moreover , at the 1MB scale , we observed good reproducibility between replicates of a cell line using the same restriction enzyme as well as different restriction enzymes , with consistent results across biological replicates and across cell lines/cell types ( S3 , S4 , S5 , S6 , S7 , S8 and S9 Figs ) . Fig 1 illustrates the distribution of rDNA contact density for 1MB segments before normalization by sequencing effort . The data shows a 5-10-fold variation in the logarithm of the contact density across segments within a cell type . The mean difference in the average contact density among cells reflects variation in the amount of Hi-C data in each cell type . For 45S rDNA contacts all 1MB segments contained appreciable density of contacts in LCL and K562 . However , the ESC and ESC-derived cell types ( ESC set ) displayed a truncated distribution with many segments that contained very few rDNA contacts ( Fig 1A ) . This was due to the lower number of Hi-C reads for those cells ( Table 1 ) . The resolution was much worse for the 5S rDNA arrays ( Fig 1C ) . Therefore , the following analyses focused primarily in the data for LCL and K562 cell lines , with the ESC or ESC-derived cells mostly used for comparisons . Here we first addressed variation in rDNA contact density across cell lines ( LCL vs K562 data collected with the same enzyme and protocol ) . We found 808 segments of 1MB with significantly different Density of Interactions ( DI ) of 45S rDNA contacts between LCL and K562 ( Fig 2; Fig 3 , FDR < 0 . 05; S3 Table ) , whereas none is identified among biological replicates of LCLs processed with different enzymes ( Fig 3 ) . We observed that 350 DI segments displayed increased density in LCL and 458 segments displayed increased density in K562 . Among those 808 DI segments , 302 of them displayed a greater than 2-fold difference in contact density between LCL vs K562 . Similarly , nearly half of the 224 segments of 1MB in chromosome 1 showed evidence of DI density between LCL and K562 ( Fig 3; chromosome 1: 106 segments significantly different , and 118 non-significant bins ) , with 97 segments displaying greater contact density in LCL and 9 segments containing greater contact density in K562 . Chromosome 1 had the largest number of significantly different DI , followed by Chr 13 ( 89 ) , Chr 9 ( 63 ) , and Chr 6 ( 61 ) . Among the five cell types ( ESC related ) , there were 193 segments of 1MB with significantly different DI with the rDNA ( FDR<0 . 05; S4 Table ) . Finally , we detected a meager 15 segments with evidence of differential DI between LCL and K562 for the 5S rDNA ( FDR<0 . 05 ) ; the small number of differential DI likely reflects the many fewer 5S rDNA reads and thus the much-lowered statistical power of this analysis . Similarly , there was not enough Hi-C data to enable statistical analysis of DI with the 5S rDNA among the five ESC related cell types . We identified 9 , 595 and 9 , 864 genes without evidence of a 5S or 45S rDNA pseudogene , respectively . The remaining genes were excluded from all further analyzes . The data showed a continuous distribution of rDNA-gene contact density for the 45S and 5S rDNA ( Fig 1B ) , with much better resolution for the 45S rDNA than for the 5S rDNA ( Table 2 , Table 3 ) . As expected , the rDNA-gene contact density was correlated with gene length . We have thus calculated the 45S contact density per gene per nucleotide ( “Contacts per gene per nucleotide , CPGN” ) . This removed the correlations between gene length and 45S rDNA contacts and revealed that CPGN for the 45S rDNA arrays was strongly correlated between LCL and K562 ( rho = 0 . 65; P < 0 . 001 ) . This correlation was stronger than those between LCL and ESC ( rho = 0 . 27; P < 0 . 001 ) or between K562 and ESC ( rho = 0 . 34; P < 0 . 001 ) . The lower correlations with ESC might partially reflect the lower resolution of the ESC contact map with a substantial fraction of genes showing less than 10 reads with rDNA contacts ( Table 2 ) . Indeed , although the overall amount of HI-C data was large , the resolution to ascertain 45S rDNA-gene contacts was only sufficient for LCL and K562 , the two biological sources with the largest number of Hi-C reads to date . The issue of low rDNA-gene resolution was particularly evident for the 5S rDNA . Out of 9595 genes analyzed for 45S rDNA arrays , there were 67 and 612 genes with zero 5S contacts in LCL and K562 , respectively . For the ESC set , however , there were 1745 genes with zero contacts with the 45S rDNA arrays . Out of 9864 genes analyzed for 5S rDNA arrays , there were 5916 and 7494 genes with zero 5S contacts in LCL and K562 , respectively . For the ESC set , we observed that greater than 95% of the genes had zero 5S contacts . The density of 5S rDNA-gene contacts was most strongly correlated with gene length ( rho > 0 . 3 , P < 0 . 001 ) , but calculating the 5S contact density per gene per nucleotide ( “Contacts per gene , CPGN” ) removed the positive association . Among genes with at least one read showing 5S-rDNA contact in both LCL and K562 we found that CPGN is strongly correlated between LCL and K562 ( rho = 0 . 64 , P < 0 . 001 ) . Evidence for a positive association between the density of 45S rDNA contacts and the density of 5S rDNA contacts is also observed in other partitions of the data , and across genes and 1MB segments in both LCL and K562 ( Table 4 ) . Here we tested for variation in rDNA-gene contact density between LCL and K562 . For the 45S array , we observed 731 genes with fold change in interaction density >2 for the LCL vs K562 comparison ( experiments with the same enzyme and protocol ) ; 97 genes ( FDR < 0 . 05 ) displayed significantly different DI after multiple corrections ( S10 Fig ) . For the analyses of 45S rDNA contacts variation among five ESC related cell types , we observed 435 genes with significantly differential density of rDNA contacts ( FDR < 0 . 05 ) . For the 5S array , we observed 954 genes with DI fold change >2 in the LCL vs . K562 comparison . However , none of these genes reached statistical significance , possibly due to the higher variance emerging from the low coverage and thus limited number of 5S contacts in each gene . There was not enough data for statistical analyses of variation in 5S rDNA contact among the five ESC related cell types . Here we estimated contact densities per base pair in three ways . First , the average contact per base pair across the whole genome was calculated by dividing the total number of mapped rDNA-genome reads by the genome length ( 3 billion base pairs ) . The average contact rate is estimated as 4 . 8 x 10−5 and 3 . 7 x 10−3 contacts per base pair for the 5S and 45S rDNA , respectively ( S5 Table ) . Hence , for the 45S rDNA each base pair in the genome is expected to have 0 . 37 mapped reads . Second , the average contact per base pair was estimated after filtering out bins with pseudogenes . Here we divided the total number of rDNA-genome reads within 1MB segments without a pseudogene by the total sequence length in those segments . This yielded an estimated average contact rate of 2 . 0 x 10−5 and 1 . 7 x 10−3 contacts per base pair for the 5S and 45S rDNA , respectively . These numbers are comparable with those estimates using all rDNA reads and the whole genome . Third , we estimated the average contact rate per base pair in protein-coding genes by dividing the total number of rDNA-gene reads by the total length of nucleotides within genes , after excluding genes with evidence of containing rDNA pseudogenes . This yielded an average contact rate for genic segments of 2 . 2 x 10−4 and 0 . 016 contacts per base pair for the 5S and 45S rDNA , respectively ( S5 Table ) . Collectively , these estimates of contact rate are useful in evaluating regions with putative enrichment or deficit in rDNA contacts . We examined the relationship between various genomic attributes and the density of rDNA contacts . First , the data showed a significant association between the number of 45S rDNA contacts and the A/B compartments . Specifically , the B compartment of closed chromatin displays an enrichment in rDNA contacts , whereas the A compartment of open chromatin displays a deficit of rDNA contacts ( P < 0 . 01 , Chi-square test; Fig 4 ) . In addition , we examined 15 functional annotations; significant enrichments were observed in segments of repressive chromatin , as well as in segments annotated as repetitive or containing insulator regions ( P < 0 . 01 , Chi-square test; Fig 4; S11 Fig ) . Finally , we examined segments of CTCF binding; CTCF is a conserved 11-zinc finger DNA binding protein that regulates chromosome architecture [20] . Using the CTCF database we estimated that CTCF binding segments constitute <7 . 5% of the human genome . On the other hand , we observed that 37% and 29% of all 45S rDNA-genome reads overlapped a CTCF binding segment in LCL and K562 , respectively . These figures are in good agreement with the 35% of all rDNA-genome reads that overlapped a CTCF binding segment in the ESC cell set . These represent a >4-fold enrichment that indicate a significantly higher percentage of 45S rDNA contacts with CTCF binding sites ( P < 0 . 05 , one proportion test ) . We selected a small set of genes to be examined in greater detail . Specifically , we examined genes that are ( i ) known to regulate rDNA function or structure and/or ( ii ) whose expression are associated with rDNA CN variation [21–23] . For instance , the CTCF gene is located on Chr16 and displayed a meager 118 contacts with the 45S rDNA in LCLs , which is significantly lower ( P-value < 0 . 001 , one proportion test ) than the expected 1198 contacts calculated based on the genome wide average contacts per base pair ( 1 . 56% ) and the length of the CTCF gene . Thus , the CTCF gene appears to be in repulsion to the rDNA arrays . Similarly , CBX1 ( Hp1beta ) , Ubf1 , and KDM4B had fewer hits than expected ( P < 0 . 0001 for all of them , one proportion test ) . Thus , we examined the top 400 genes that are positively and negatively associated with rDNA CN variation in LCL [21] . Collectively , however , these genes were neither enriched nor depleted in rDNA contacts , with a distribution of contacts that is undistinguishable from all other genes in the genome ( Fig 5 ) . Nevertheless , nucleolar , mitochondrial , and ribosomal genes were also associated with variation in rDNA array CN [21] , and could reveal a distinct pattern . Accordingly , genes that localize to the nucleolus as well as ribosomal genes showed a distribution of contacts that was significantly shifted towards a greater than average number of contacts with the rDNA array in both LCLs and K562 ( Fig 5 and Fig 6 ) . Next , we addressed if the higher density of rDNA contacts in nucleolar , ribosomal , and mitochondrial genes would emerge as significant gene ontology enrichments when genes with a high CPGN are selected . To address the issue , we examined the genes in the top 5% higher number of 5S and 45S contacts after correction for gene length ( i . e . , CPGN ) . For 5S rDNA-gene contacts in LCL the cell component category of mitochondrion ( GO:0005739 ) emerged on the top of the list , with 56 candidates ( out of 494 genes ) localized to the mitochondrion . The association is functionally intriguing and also emerged in the K562 dataset ( S6 Table ) . The same class emerged among the top 5% in the 45S rDNA in LCL , with 63 candidates in the mitochondrion ( GO:0005739; adjusted P < 0 . 05 , after correction for multiple testing ) . The class includes interesting candidates such as seryl-tRNA synthetase 2 ( mitochondrial SARS2; ENSG00000104835 ) , tRNA 5-methylaminomethyl-2-thiouridylate methyltransferase ( TRMU; ENSG00000100416 ) and tRNA methyltransferase 1 ( TRMT1; ENSG00000104907 ) , Era like 12S mitochondrial rRNA chaperone 1 ( ERAL1; ENSG00000132591 ) . In addition , 10 other functionally coherent cell components emerged for 45S rDNA-gene contacts in LCL ( S7 Table; adjusted P < 0 . 05 , for all classes in LCL; see S8 and S9 Tables for data on K562 and the ESC set ) . Four of those categories are highly significant GO terms containing the protein-components of the ribosome ( GO:0005840~ribosome , GO:0022625~cytosolic large ribosomal subunit , GO:0015935~small ribosomal subunit , and GO:0022627~cytosolic small ribosomal subunit ) . Collectively , the data suggest that highly transcribed genes encoding protein constituents of the ribosome are co-localized in proximity to the rDNA arrays ( Table 5 ) . In addition , one GO term related to nucleolar function ( GO:0005730~nucleolus ) also emerged as significantly enriched with 39 genes in the top 5% of genes with higher numbers of 45S rDNA-gene contacts in LCL . Genes in this set include intriguing candidates such as NOP2 nucleolar protein ( NOP2; ENSG00000111641 ) , FSHD region gene 1 ( FRG1; ENSG00000109536 ) , Sirtuin 6 ( SIRT6; ENSG00000077463 ) , and MDM2 ( ENSG00000135679 ) . Among genes in the bottom 5% CPGN in 45S , we observed seven HOX genes dispersed across several chromosomes ( HOXA1 , HOXA6 , HOXA7 , and HOXA11 on Chr 7 , HOXB5 on Chr 17 , HOXC11 on Chr 12 , and HOXD13 on Chr 2 ) , three of which showed zero 45S rDNA contacts [HOXA7 ( Chr 7 ) , HOXC11 ( Chr 12 ) , and HOXD13 ( Chr 2 ) ] even in the dense LCL map . This suggests that developmentally regulated Hox genes are rarely localized in proximity to the rDNA arrays . Furthermore , we also found several other developmental genes in the set of 67 genes with zero contacts with 45S rDNA genes , further indicating that developmental genes show “repulsion” from the rDNA genes . Interesting candidates include NK2 homeobox 3 ( NKX2-3 ) on Chr 10 , BMP3 on Chr 4 , BMP5 and BMP6 on Chr 6 , as well as NOTCH1 on Chr 9 . Interestingly , the histone cluster 1 H1 family member d ( HIST1H1D ) on Chr 6 also emerged without a single 45S rDNA contact in the dense 45S map of LCLs . Finally , we confirmed the lack of Hi-C contacts between the 5S and 45S arrays [24] . The segments proximal to the 5S array also displayed depletion in 45S rDNA contacts . The gene RHOU , for instance , is located adjacent to the 5S array and emerged in the bottom 3% of the distribution of 45S rDNA contact density .
Multicopy ribosomal DNA arrays are essential components of the genome . Yet ribosomal DNA arrays are also among the most variable segments of the genome . The arrays have lagged behind with limited assemblies and little understanding of their nuclear localization . Here we report a detailed contact map of spatial interactions between the rDNA arrays and the rest of the genome . Although there are huge amounts of HI-C data , analyses of rDNA contact density for specific regions/genes remained a challenge because rDNA reads constitute a fraction of the Hi-C reads . Thus , we combined multiple Hi-C datasets to identify the subset of reads containing information on rDNA contacts . The effort was computational intensive because the fraction of rDNA reads in shotgun Hi-C is very small . This is particularly evident in the case of the 5S rDNA array: the contact data remained sparse even for LCL , the cell line that has by far the largest amounts of data collected from multiple Hi-C experiments . Nevertheless , we identified consistency of rDNA-gene contacts across different cells ( LCL and K562; especially for 45S ) , which point to replicable spatial interactions . Heatmaps enabled visualization of rDNA contacts along the human genome with statistical analyses pinpointing significant differences in the density of contacts . While the approach can be applied to other multicopy genes as well as single copy genes or regions , we caution that the typical resolution of shotgun Hi-C is not sufficiently high . Indeed , limited resolution was apparent for both the 5S rDNA and 45S rDNA arrays , which required combining multiple datasets to ascertain contacts with genic and non-genic segments of the genome . In summary , the LCL map achieved good resolution for 5S and 45S contacts but the K562 set is quite sparse for 5S contacts , and both 5S and 45S maps are very sparse in the case of the ESC cell and ESC-derived cell types . Variation in rDNA contact density across genes reflects variation in proximity to the rDNA arrays . The data displayed over 100-fold variation in contact density across genes and revealed several intriguing patterns . First , the compilation enabled us to conduct statistical tests of the differential density of rDNA interactions between LCL and K562 . These 45S maps are sufficiently dense , with differences in contact density likely reflecting differences in nuclear organization between these cells . These differences are not surprising since the LCLs are immortalized cells derived from lymphocytes whereas K562 is a myelogenous leukemia . K562 has , moreover , undergone genomic rearrangements [25] . While the data also suggested variation across ESC and ESC-derived cell types , greater coverage for these cells is necessary to draw sufficiently dense contact maps for a more fine-grained and meaningful biological contrast . An intriguing suggestion is that the rDNA/nucleolus represents a keystone in nuclear structure around which the rest of the genome is functionally organized [26] [24] . In this case , rDNA-contact differences between cells are bound to emerge and reflect functional variation . Second , as a class , the rDNA proximity with genes previously identified as associated with rDNA CN variation across genotypes in human populations is undistinguishable from the background of genes . This indicates that genes impacted by rDNA CN are not spatially close to the rDNA arrays and are not enriched in direct rDNA contacts . This is not an unexpected observation , because the association of gene expression variation with rDNA CN includes hundreds of genes , with only a fraction of them likely to be directly regulated by the rDNA array ( i . e . genes associated with rDNA CN are presumably modulated by both direct and indirect effects emerging from the rDNA ) . While we suggest that changes in nuclear architecture could be one way to for the rDNA to exert regulatory effects , the mechanisms through which rDNA CN directly modulates gene expression are likely varied and the ratio of direct to indirect effects is unknown . Third , we observed that genes encoding proteins that localize to the mitochondria display a disproportionally large number of contacts with the 45S rDNA . Concordantly , genes localized to the mitochondria also emerged as enriched in 5S rDNA contacts . The data suggests that genes localized to the mitochondria might be collectively regulated through aspects of nuclear architecture that are influenced by the rDNA . Noteworthy , connections between the rDNA array and mitochondrial gene expression and function have been uncovered before . In Drosophila , Paredes et al ( 2011 ) observed that engineered deletions in the rDNA array preferentially impacted the expression of genes whose protein products localized to the mitochondrion [27] . In humans , Gibbons et al ( 2014 ) observed that rDNA CN variation is associated with the expression of genes whose protein product localize to the mitochondrion as well as genes encoding protein components of the mitochondrial ribosome and mitochondrial DNA copy number [21] . Interestingly , in addition to its well-documented role as a structural component of the cytosolic ribosome , the 5S rRNA is also specifically imported into the mitochondria [28 , 29] . Fourth , we observed that genes localized to the nucleolus and encoding protein components of the ribosome were significantly enriched for 45S rDNA contacts . The finding points to the specificity of rDNA-genome interactions and suggests that ribosomal gene regulation might be directly influenced by the rDNA array . This pattern of rDNA-gene contacts might partially explain the observation that genes whose expression was correlated with rDNA CN included several candidates encoding the protein components of the ribosome . Indeed , sequence specific inter-chromosomal interactions between the yeast rDNA array and an intergenic segment adjacent to the largest RNA pol I subunit has recently been demonstrated [30] . All in all , our study identified functionally coherent genes and GO categories that are depleted and enriched in direct rDNA contacts . Ribosomal DNA contacted regions for all chromosomes along the human genome suggest a structural component underlying the global regulatory consequence of rDNA CN variation [21] . Finally , we note that as much as 29% of the 45S rDNA reads have both ends mapped in the 45S rDNA . These partially reflect linear proximity along the 45S rDNA unit but could also emerge from looping substructures with contacts between distant units; looping and contact among non-adjacent units has been suggested to facilitate ultra-structural organization of the array and coordinate transcription among rDNA repeat units [6 , 7 , 24 , 31–35] . Concerted copy number variation ( cCNV ) refers to the correlation in copy number of 5S and 45S rDNA [36] . This co-variation in copy number across genotypes with variable rDNA array size is observed in human lymphoblastoid cells ( LCLs ) and occurs despite 5S and 45S rDNA residence on different chromosomes and lack of sequence homology between 5S and 45S rDNA subunits . Therefore , physical linkage between loci cannot explain the co-variation . On the other hand , spatial co-localization of the arrays as well as cellular processes of recombination such as those of micro-homology mediated end joining could conceivably contribute to the emergence of cCNV . Our results , however , confirmed a lack of direct 5S-45S contacts in Hi-C , an observation that is in agreement with a previous study [24] . This included a lack of 45S rDNA contacts with genes that are adjacent to the 5S rDNA array . The gene RHOU , for instance , is located next to the 5S and emerged in the bottom 3% of the distribution of 45S rDNA contact density . This indicates that the 5S and 45S rDNA are not in close enough proximity or that large protein complexes prevent the formation of 5S-45S Hi-C reads . The findings support the hypothesis that physical interactions occurring between 5S and 45S rDNA arrays are more restricted than previously anticipated . On the other hand , the denser maps presented here indicate that the 5S and 45S arrays share overlapping contact maps and many regions of the genome display a high density of contacts with both rDNA arrays . For instance , the density of 5S and 45S contacts is strongly correlated across genes and 1MB segments in both LCL and K562 cells . Whether or not this overlapping contact map is relevant for cCNV remains to be determined , but the evidence suggests that the two arrays are not completely independent . Coordination between them is likely to be relevant , with costs and benefits to 5S array proximity with the 45S arrays [24] . All in all , the association between contact density for the 5S and 45S arrays suggest that cCNV might be facilitated by structural proximity . Similarly , rDNA mediated structural changes in the nucleus might partially explain the regulatory consequences of naturally occurring variation in rDNA copy number [21] . From an evolutionary perspective , the co-existence of two clusters of rDNA loci ( 5S and 45S ) might incur costs and benefits compared to rDNA residency on a single location . In some plants and yeasts , the 5S and 45S/35S rDNA subunits are spatially adjacent in the genome [7 , 37–41] , whereas in Drosophila and mammals , the 5S and 45S arrays reside on different chromosomes . However , the correlated contact maps for the 45S and 5S rDNA arrays suggest that they preferentially anchor at overlapping domains . This might narrow their spatial distances , and could explain why the 5S and 45S arrays can display apparent proximity to one another in a fraction of the cells as observed in cytological preparations . However , the lack of direct Hi-C 5S-45S contacts might suggest a model of competitive exclusion for similar anchoring sites , and predicts that a segment is in close proximity to either the 5S or 45S rDNA at each time . In cases of cytological proximity between the 5S and 45S arrays , large protein complexes might be present and prevent the emergence of direct 5S-45S inter-chromosomal contacts in the scale captured by Hi-C technology . Furthermore , the enrichment of rDNA contacts with ribosomal protein coding genes is surprising and might help explain the association between rDNA CN and the expression of these genes [21] . It suggests a structural component to the regulatory role of the rDNA and raises the possibility that the arrays might exert direct modulation of some genes via changes in nuclear organization . The data suggest that models that exclusively consider proximity to the rDNA arrays/nucleolus as a repressive modifier of gene expression might be overly simplistic . Rather , the distal and proximal association of genes with the rDNA arrays appears functionally motivated , as in the case of developmental genes or ribosomal genes . For instance , ribosomal gene proximity to the rDNA arrays could help facilitate coordinated Pol I , Pol II and Pol III responses . Collectively , these structural rDNA-mediated associations might have partially evolved to mitigate the fitness costs of dosage imbalances among highly expressed RNA and protein components of the translational machinery .
The human 5S rDNA along with flanking regions ( chr1: 228 , 765 , 135–228 , 767 , 255 ) and the human 45S rDNA ( GenBank reference number U13369 . 1 , with modifications ) were obtained as recently described [21 , 36 , 42 , 43] . The 45S reference comprises the 18S , 5 . 8S and 28S rRNA encoding segments , external transcribed sequences ( ETS ) and internal transcribed segments 1 and 2 ( ITS1 and ITS2 ) , as well as a ~32 Kb non-coding intergenic spacer ( IGS ) . Both 5S and 45S segments contain repetitive elements , such as Alu and Line1; all analysis carried out in this study used 5S and 45S sequences masked for these repeats . Raw Hi-C reads for LCLs and erythroleukemia K562 ( K562 ) cells were downloaded from the Gene Expression Omnibus ( GEO ) repository with accession number GSE63525 [44] . Biological replicates with more than 1 technical replicate were included for a total of 6 , 017 , 877 , 658 reads in LCL and 1 , 366 , 228 , 845 reads in K562 . In addition , raw Hi-C reads for five cell types were obtained from GEO data with SRA Study number SRP033089 [45] . The five cell types comprised the H1 embryonic stem ( ES ) cells and four differentiated cell-types derived from H1 [Mesendoderm ( ME ) cells , Mesenchymal stem ( MS ) cells , Neuronal Progenitor ( NP ) cells , trophoblast-like ( TB ) cells] [45 , 46] . The number of reads studied and recovery rates for 5S and 45S informative reads was summarized in Table 1 . All data were downloaded in SRA format and converted into FASTQ files by the NCBI SRA Toolkit’s command ( fastq-dump ) . FASTQ files were quality and adapter trimmed with Trim Galore . The trimming criteria required minimal quality score ( > 20 ) and length ( >50 bp ) . Next , we identified Hi-C reads that mapped to the 5S rDNA array or the 45S rDNA array . In this step , both forward and reverse reads were mapped independently to the 5S rDNA and 45S rDNA using Bowtie2 [47] . We used unpaired mapping with ‘very-sensitive’ mode ( combinations of parameters: -D 20 -R 3 -N 0 -L 20 -i S , 1 , 0 . 50 ) . The mapping results were sorted and converted into binary format using SAMtools [48] and bed format using BEDTools [49] . We then extracted reads that mapped to the rDNA array and mapped the opposite end to repeat libraries . Reads for which one end mapped to repeats library were excluded . Finally , in order to identify potential confounders due to rDNA pseudogenes , both rDNA references were blasted against the human genome separately . Putative pseudogenes were identified as significant hits ( E-value <1 × 10−4 ) using BLASTN [50 , 51] . A segment of 1 MB was excluded from the analysis if an rDNA blast hit is identified within it . Similarly , a gene was removed from analysis if an rDNA blast hit is identified within its boundaries . To identify spatial variation in genomic contact density along the chromosomes we segmented the human genome GRCh37/hg19 assembly into 3 , 173 bins of 1MB using BEDTools [49] . Bins with rDNA pseudogenes were excluded . Contact densities were summarized for each bin for each of 5 cell types and 2 cell lines . We calculated the number of Contacts Per Million reads ( CPM ) to normalize the data and control for different number of reads in each of the seven conditions . This placed all the data in a comparable scale , to enable visualization of contact density along the human genome using heat maps in the 'gplots' R package [52] . The term of “rDNA-gene contact” refers to reads with one end mapped to rDNA arrays and the other end mapped between the first and the last exon of an annotated gene in the human genome . We extracted coordinates of these reads using BEDTools [49] and the Gene Transfer Format ( GTF ) file: Homo_sapiens . GRCh37 . 75 . gtf from the Ensembl database . GC content and length were also computed for each gene . To normalize contact densities in genes of different length , we computed the number of contacts per gene length in nucleotides ( Contacts reads per gene per nucleotide , CPGN ) . The web based tool DAVID v6 . 8 [53] was used to investigate gene ontology enrichments for the top 5% of genes with greater CPGN for 5S rDNA or 45S rDNA genes . This corresponds to 494 out of 9864 genes for 5S-gene contacts , and 480 out of 9595 genes for 45S-gene contacts . The “one proportion” test [54] was also applied to address whether the number of mapped reads per base pair within a gene is significantly different from the genome wide average . We modeled differential contact density per 1MB and per gene using the edgeR package and statistical approaches adapted from RNA-seq analysis [55 , 56] . Raw counts for physical contacts with rDNA loci within each bin along the human genome are modeled using generalized linear models ( likelihood ratio tests ) implemented in the edgeR package [55 , 56] . These approaches were recently been used to detect differential interaction density ( DIs ) in Hi-C data [19 , 57 , 58] . The models identified statistically significant differences among cell lines/types in rDNA contacts density per MB and within genes . The Benjamini-Hochberg method was used for multiple testing correction [59] , and statistical significance was denoted by FDR < 0 . 05 . We applied the method to ascertain significant differences between LCL and K562 data from a single publication . For statistical comparison , we focused specifically on 11 biological replicates for LCL ( collected with the Mbol enzyme ) contrasted with two biological replicates for K562 ( collected with the Mbol enzyme ) and two biological replicates for LCL ( collected with the DpnII enzyme ) . Each biological replicate consists of multiple technical replicates . We also evaluated variation among the five ES derived cell types , each with two biological replicates . We cross-referenced the rDNA contact map with several sources of functional annotation . First , Hi-C studies proposed the partition of the genome into A and B compartments that are widely interpreted as open and closed chromatin , respectively [60] . A/B coordinates were downloaded for LCL cells and 12 cancer types [60] . Second , coordinates of 15 functional regions identified in hESC using ChromHMM [61] were downloaded . Third , information on replication timing along the genome was downloaded from the Replication Domain Database ( www . replicationdomain . org ) . Finally , CTCF binding coordinates were obtained from the CTCFBSDB database [62] . We extracted the coordinates for all the segments in each annotation and addressed its density of rDNA contacts . BEDTools was used to assess the number of mapped reads that overlapped with each annotated segment for each dataset . The percentage of mapped reads was calculated by dividing the number of reads mapped to the segment by the total number of mapped reads . The genome wide average read per base pair was used to compute the expected number of reads in the functional segment . Statistical significance was assessed with Chi-square tests . In addition , we applied the “one proportion” statistical test [54] to address whether the numbers of mapped reads per base pair within a functional segment ( e . g . , CTCF binding ) is significantly different from the genome-wide average per nucleotide contact rate .
|
The repeated ribosomal DNA ( rDNA ) array gives rise to the nucleolus , an organelle that is involved in key cellular processes such as stress response , cell cycle regulation , RNA modification , and production of more than 70% of all cellular RNAs ( the ribosomal RNAs ) . This critical genome element has remained unassembled and has been excluded from all Hi-C analyses to date . Here we built the first map of 5S and 45S rDNA contacts with the rest of the genome . The map yielded a number of novel results and challenge the expectation that 5S and 45S arrays are close together in the nucleus . The rDNA arrays share common sites of contact across the genome , are biased towards segments of closed , repressed , and late replicating chromatin , and display greater proximity or avoidance to functionally coherent gene sets . The results further our understanding of the rDNA arrays and their localization in the nuclear environment .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"pseudogenes",
"human",
"genomics",
"genome",
"analysis",
"gene",
"types",
"energy-producing",
"organelles",
"mitochondria",
"bioenergetics",
"cell",
"nucleus",
"cellular",
"structures",
"and",
"organelles",
"nucleolus",
"genome",
"complexity",
"proteins",
"gene",
"expression",
"gene",
"ontologies",
"ribosomes",
"homeobox",
"biochemistry",
"rna",
"ribosomal",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"protein",
"domains",
"biology",
"and",
"life",
"sciences",
"genomics",
"non-coding",
"rna",
"computational",
"biology"
] |
2018
|
The long-range interaction map of ribosomal DNA arrays
|
Endocytosis is an essential process by which cells internalize a piece of plasma membrane and material from the outside . In cells with turgor , pressure opposes membrane deformations , and increases the amount of force that has to be generated by the endocytic machinery . To determine this force , and calculate the shape of the membrane , we used physical theory to model an elastic surface under pressure . Accurate fits of experimental profiles are obtained assuming that the coated membrane is highly rigid and preferentially curved at the endocytic site . The forces required from the actin machinery peaks at the onset of deformation , indicating that once invagination has been initiated , endocytosis is unlikely to stall before completion . Coat proteins do not lower the initiation force but may affect the process by the curvature they induce . In the presence of isotropic curvature inducers , pulling the tip of the invagination can trigger the formation of a neck at the base of the invagination . Hence direct neck constriction by actin may not be required , while its pulling role is essential . Finally , the theory shows that anisotropic curvature effectors stabilize membrane invaginations , and the loss of crescent-shaped BAR domain proteins such as Rvs167 could therefore trigger membrane scission .
Endocytosis enables cells to internalize extracellular material and to recycle membrane components [1] . During this process , the plasma membrane is deformed into an invagination progressing inwards , which is severed and eventually released in the cytoplasm as a vesicle . Key endocytic components have been identified in several systems . This process usually involves membrane coating proteins ( such as clathrin ) , their adaptors ( such as epsins ) and actin microfilaments together with associated factors [2] . We focused on the yeast model system , in which endocytosis is well characterized experimentally . The abundance and localization of the principal proteinaceous components have been measured as a function of time both in S . pombe [3 , 4] and S . cerevisiae [5 , 6] . To understand how these components work together to deform the membrane , it is necessary to consider the physical constraints under which the task is performed in vivo . While in animal cells , invaginations are opposed mostly by membrane tension and elasticity [7] , turgor pressure strongly opposes invaginations in plants and fungi . In those cells , the difference of osmolarity with the outside causes a large pressure pushing membrane against the cell wall . The magnitude of the pressure has been measured in different walled cells using various methods . For plant cells , studies converge to the range 0 . 2–1 MPa ( see [9] for a careful review of the subject ) . In the yeast S . pombe , an effective pressure of 0 . 85 ± 0 . 15 MPa was derived from studying the buckling of the rod-like cells in micro fabricated chambers [10] . Based on the variation of volumes upon changes in osmolarity , the pressure in S . cerevisiae was recently estimated to be 0 . 6 ± 0 . 2 MPa [11] , while an older study concluded 0 . 2 MPa for stationary phase cells [12] . It has been suggested that the pressure could be decreased locally by releasing osmolytes at the endocytic patch [13] . This would however practically not induce any local variation of pressure , because hydrostatic pressure gradients equilibrate at the speed of sound ( possibly 1500 m/s ) . Given the size of yeast , this is considerably faster than an endocytic event ( ∼ 5s ) . The pressure , thus of the order of 1 MPa ∼ 1 pN/nm2 , pushes the plasma membrane outward uniformly . This effect is balanced by an equal force from the cell wall , wherever the membrane is in contact with the cell wall ( Fig 1 , left ) . The pressure strongly opposes endocytic membrane invagination , as membrane and cell wall must come apart . The large required force is produced in yeast by an actin machinery , which forms a crosslinked network of rigid filaments around the invagination ( see [8] and Fig 1 , left ) . Actin polymerizes close to the basal plasma membrane [6] . The newly inserted F-actin at the bottom of the network is though to lift the entire network away from the cell wall ( upward on Fig 1 , left ) [13] . For this force to be productive , the actin network should be attached to the tip of the invagination , and this is the role of the protein sla2 , which is required for endocytosis—while several other proteins , including clathrin , are dispensable [5 , 14] . This essential role for actin is further supported by the fact that actin assembly precedes or coincides with membrane deformation [6 , 8] . The interplay between actin and pressure was nicely demonstrated by showing that impairing the arp2/3 complex ( an actin nucleator ) delayed the invagination , while decreasing the hydrostatic pressure ( by adding sorbitol to the media ) had the opposite effect [15] . Coat and associated proteins play a role in endocytosis , notably by recruiting the endocytic machinery , and keeping the actin network physically connected to the membrane . Proteins that bind to the lipid bilayer can also directly induce the membrane to curve [16 , 17] . Two familiar examples in yeast endocytosis are clathrin [1] and Rvs167 [18] . The induced curvature is expected to be qualitatively different for these two proteins , because clathrin proteins form regular triskelion [19] while Rvs167 is shaped as a crescent [20] . Moreover , clathrin adaptors can bind to membrane , cargo , actin and/or clathrin and have essential functions in endocytosis [21] . Additionally , some adaptor proteins can induce membrane curvature [22] . The shape of the membrane can be predicted by minimizing an effective deformation energy [23] . This approach has been used successfully in different systems [24–26] , but the case of yeast endocytosis where the membrane detaches from the cell wall despite a large hydrostatic pressure has not been analyzed theoretically to our knowledge . Most previous work focused on the endocytosis of viruses in animal cells , which occurs by wrapping of the membrane around viral particles in the absence of hydrostatic pressure [7 , 27 , 28]; the constriction of vesicle necks was also studied recently [29] , but without comparison to physiological results . Yeast endocytosis was modelled before , mostly without turgor [30 , 31] . One study considered a difference of pressure across the plasma membrane [32] , but adopted a value of the pressure estimated in yeast spheroplasts [33] . The pressure in these cells , which lack a cell wall , is at least two orders of magnitude lower than under more physiological conditions [10] , and the forces in this study are consequently severely underestimated . Here we integrate the contributions of pressure , coat proteins and membrane properties during the endocytic invagination , providing an estimate of the force that actin must exert to induce the invagination . We study the effect of coat proteins such as clathrin that induce isotropic curvature , and contrast it with crescent-shape proteins such as BAR-domain that induce curvature only in one direction . Finally , we discuss whether the combination of actin-mediated pulling together with the removal of Rvs167 is sufficient to lead to vesicle internalization .
We predict membrane shapes by minimizing a deformation energy , using a Helfrich-type Hamiltonian [24 , 34] , in which membrane deformations are penalized by a bending rigidity κ and tension σ . We write Π the difference of hydrostatic pressure between the inside and the outside of the cell . In addition , we assume a point force fa pulling the apex of the invagination , to represent the driving force generated by the actin cytoskeleton [6 , 13] . We thus implicitly assume that the forces produced by actin polymerization at the base of the invagination , and possibly by myosin motors or other processes , are transmitted to the tip of the invagination over the actin network ( see Fig 1 ) . We note κ the rigidity of the membrane together with its coat of proteins [16 , 17] . Moreover , we consider that the coat proteins curve the membrane either by scaffolding or inserting themselves in the membrane . We first describe proteins such as clathrin , that induce an isotropic curvature C0 , with the same radius of curvature in both directions . To write the membrane deformation energy , we introduce C = ( 1 R 1 + 1 R 2 ) the local curvature , V the volume inside the invagination , S the surface area of the membrane that is not in contact with the wall , and L the height of the invagination . The total energy ℱ of an invagination reads [26]: F = ∫ ∫ S [ κ 2 ( C - C 0 ) 2 + σ ] d S + Π V - f a L . ( 1 ) The rigidity term will tend to make invaginations as large as possible to minimize their curvature , while both pressure and tension will tend to make invaginations smaller , to minimize their volume and surface , respectively . Therefore an invagination dominated by pressure and rigidity will have a typical width R Π = κ / 2 Π 3 while an invagination dominated by tension and rigidity will have a typical width λ ∼ κ / 2 σ . From measured values of Π , σ ( Table 1 ) , and using the rigidity of a naked membrane ( κ ∼ 40 kb T ) , we find RΠ ∼ 3 nm and λ > 9 nm . Since RΠ < λ in this case , we expect that invaginations will have a typical radius RΠ and tension will not significantly affect the shape of the invagination . In this estimate , we have used a lower bound for κ given by the rigidity of a pure lipid bilayer , but the actual value of κ will be higher because it should also include the stiffness provided by coat proteins . However , the statement RΠ < λ will be all the more true for higher values of κ because RΠ will increase slower than λ as κ becomes larger . We have used an upper bound for the membrane tension σ , but in reality it could be significantly smaller , since yeast cell have membrane furrows [35] , which act as membrane reservoirs and limit the surface tension . Moreover , we will see later how fitting the experimental membrane shapes confirms that the contribution of tension is negligible . Since the competition of pressure with membrane rigidity should control the shape of the invagination , the forces due to pressure and rigidity will have the same order of magnitude and will dominate those due to membrane tension . The total force exerted by pressure over the invagination should be ΠS0 , where S0 is the surface area of the wall that is not in contact with the membrane . This area is difficult to delineate in the electron micrographs [8] , but using the dimension at the tip of the invagination Rt ( Table 1 ) to obtain an ersatz π R t 2 , we estimate the magnitude of the force to ∼ 300 pN . Given this force , and a membrane viscosity ηm ∼ 10−8 kg . s−1 [39] , we can calculate the time needed to pull a membrane tube over a distance L ∼ 100 nm [8] . Because the resulting time scale Lηm/f ∼ 10−5 s is considerably shorter than the duration of an endocytic event ( a few seconds [8] ) , the membrane has plenty of time to reach a static equilibrium , at any stage of its evolution . In this quasi-static regime , we can derive the membrane shape equations from minimizing the energy in Eq ( 1 ) , and following earlier work , we also assume that they are axisymmetric for simplicity ( see supplementary information S . I . 1 . 1 in S1 Text ) . To compare our predictions to experimental membrane profiles , the measured shapes ( Fig 1 , right ) were projected to an axisymmetric profile ( Fig 2 ) . For a given set of C0 , σ , RΠ , only one value of fa allows the theoretical profile to match the invagination shape , and therefore three parameters could be varied to fit experimental membrane profiles ( see supplementary information S . I . 1 . 5 in S1 Text for fitting procedure ) . Generally , we found that measured profiles could be fitted precisely ( Fig 2 ) except for long invaginations ( L > 80nm , see comment in the discussion ) . A given experimental profile can usually be fitted by a range of parameters rather than a unique set , but the values of the fitting parameters are remarkably consistent . The tension is negligible in all cases as expected , and the values of RΠ and κ can be interpreted and used to calculate the magnitude of the apical driving force fa . We will usually express lengths as multiples of RΠ and forces in terms of f Π = 4 π Π R Π 2 , since this makes our results independent of the values Π and κ , which are known only within an interval , without any loss of generality . At this stage , we emphasize that the values of κ and C0 cannot be known a priori because these parameters represent properties of the coated membrane which is a complex assembly of interacting lipids and proteins , and not a simple lipid bilayer . In the following , we determine the rigidity κ and curvature C0 from the fit , assuming them to be constant over the membrane . In reality the coated membrane is expected to be inhomogeneous , but allowing these parameters to vary spatially would drastically increase the parameter space , such that fitting would be both underdetermined and computationally expensive . Moreover , the overall quality of the fits obtained by assuming κ and C0 to be homogeneous does not warrant an increase in the complexity of the theory . We discuss the implications of heterogeneity in the membrane later .
The fit always provided a well determined value for RΠ in the range 15–25 nm , and since R Π = κ / 2 Π 3 , this allowed us to estimate the ratio between the bending rigidity and the pressure . Assuming Π ∼ 1 MPa one gets κ ∼ 2000 kB T , while a conservative value of the pressure ( 0 . 2 MPa ) would yield κ ∼ 400 kB T . The coated membrane is therefore much more rigid than a pure lipid bilayer ( κ ∼ 40 kB T ) , leading to wider invaginations . The lower estimate corresponds to the expected rigidity of clathrin coated membranes in vitro ( 300 kB T , [40] ) , while the higher value is also realistic , because other proteins besides clathrin are concentrated at the endocytic site [14] . Adaptor proteins could drastically stiffen the coat by interacting with both clathrin and the membrane [22] . By increasing the effective rigidity , coat proteins enlarge the invaginations , but as a consequence , the overall resisting force is also increased . Importantly , our estimate of the minimal force needed to sustain the invagination , fa ∼ 3000 pN is not based on the parameters that are poorly determined by the fit ( such as σ ) . We were also able to determine the value of the spontaneous curvature C0 ∼ 0 . 4/RΠ . While this parameter is less well determined , it corresponds to a radius of curvature 2/C0 ∼ 100 nm , which is consistent with the known characteristics of the coat proteins . In particular , purified clathrin form in vitro spherical cages with an average radius of 35 nm [41] . The conditions in vivo are likely to be different however , since more proteins coat the membrane , and the membrane itself can also contribute to C0 . Note that we have assumed that curvature applied everywhere to the membrane , rather than in a restricted region as can be expected in reality , but this point will be discussed later . By solving the membrane shape equations for a range of invagination heights L , we could compute the required apical force during different stages of the invagination . While the magnitude of the force is set by f Π = 4 π Π R Π 2 , its variations as a function of L are very instructive . The force has a non-zero value for L = 0 , for the determined values of the spontaneous curvature C0 ∼ 0 . 4/RΠ ( Fig 3 , blue curve ) . fa increases and reaches a maximum for L ∼ RΠ , and then decreases to a minimum value for L ∼ 3RΠ . For longer invaginations a plateau is reached . This is very different from a membrane without spontaneous curvature ( Fig 3 , black line ) or for a tension-dominated membrane [26] , for which the force is minimum for L = 0 . In the presence of spontaneous curvature , the initiation force at L = 0 is high , the force peaks for smaller invaginations , and the pulling force is generally lowered for longer tubes ( compare black and blue lines in Fig 3 ) . The fact that an important initial force f0 = fa ( L = 0 ) is required to start the invagination is biologically meaningful . We estimated f0 ∼ 0 . 8fΠ , corresponding to 60% of the maximal force . Because the membrane is nearly flat if L → 0 , we could obtain analytically f 0 = 2 f Π R Π 2 C 0 2 + 2 σ R Π / κ ( see supplementary information S . I . 2 . 2 in S1 Text ) . With σ negligible , this expression reduces to f0 = 4πκC0 , which summarizes how coat proteins can affect the initial force , either by changing the spontaneous curvature , or the effective rigidity of the coated membrane . By itself , spontaneous curvature is not able to lift the membrane off the wall , but rather increases the force needed to start of the invagination . Because the force rises sharply with the height of the invagination , actin should be needed to lift the membrane already at the earliest steps of endocytosis , as recently observed [6 , 8] . This idea is consistent with the observation that the delay before significant membrane deformation is observed depends on the competition between actin and pressure [15] . It has been suggested that when the membrane is still flat , actin could pull at one site of the membrane while simultaneously pushing on a ring-shaped zone surrounding this site ( Fig 1 , left ) . This scenario was supported by the recent observation that sla1 , an actin organizer , forms rings at endocytic sites on flat membranes , possibly indicating where pushing forces are applied [6] . Because the membrane is still juxtaposed to the cell wall at this stage , the pulling and pushing forces generated by the actin network have to be balanced . Finally , a key feature of the force profile is that the most demanding part of the invagination occurs early during endocytosis ( L < RΠ ) . In mechanical terms , this can lead to a snap-through transition [32] , in which once a critical threshold is reached , subsequent stages spontaneously follow because they do not require additional efforts . This may explain that while the duration of the endocytic early phase is highly sensitive to the competition between actin nucleation and hydrostatic pressure , later stages are largely insensitive to pressure [15] . This snap-through force profile also elegantly explains why so few retraction events were observed experimentally ( less than 1% of endocytic events fail to complete [5] ) , despite the inherent stochasticity of a biological machine such as endocytosis , that only contains a few hundred actin filaments [3] . Above a certain height , invagination shapes predicted by theory exhibited a neck , even though there is no constriction force in the model ( Fig 2 ) . The only external force is a vertical lifting force applied on the apex of the invagination , and the neck appears as a consequence of the spontaneous curvature ( Fig 2 , last profile ) . Higher spontaneous curvatures will produce smaller necks , up to a point where the theory predicts a neck of zero radius , corresponding to a shape instability . The apparition of a localized neck as a result of global spontaneous curvature has been described theoretically before [29] . We have more precisely characterized the nature of the shape instability ( see supplementary information S . I . 3 in S1 Text ) . We can understand its biological consequences in two limit cases: with a given value of C0 with increasing length L , or with a given length L and increasing C0 . Firstly , for small values of the spontaneous curvature , the membrane shape evolves smoothly upon increasing the length L of the invagination , and fa ( L ) is continuous . Above a threshold C 0 * however , we can distinguish two branches: one corresponding to short , quasi-tubular invaginations , and another one corresponding to larger spheroidal invaginations ( Fig 4a–4c , C0 = 0 . 45/RΠ ) . The two branches overlap , i . e . for a range of lengths , there are two possible membrane shapes for a given L , as illustrated on the example profiles ( Fig 4a ) . Above a second threshold C 0 * * , the two branches cease to overlap , such that there is a range of L where there is no equilibrium membrane shape ( Fig 4d–4f , C0 = 1/RΠ ) . The shapes corresponding to the edges of these regions are illustrated in ( Fig 4d ) . The transition from the short invagination to the tall spheroidal shape corresponds to a decrease in total membrane deformation energy ( Fig 4f ) . We could compute the values of C 0 * and C 0 * * as a function of tension ( Supplementary Fig . 4 ) . The same phenomena can be observed for a given length L by increasing C0 . There is a threshold C 0 + ( L ) at which the predicted radius is zero , see Fig 5 . For larger spontaneous curvatures C 0 > C 0 + ( L ) , no stable membrane shape exist . This is in agreement with the existence of a zero-radius neck at a critical value of C0 [29] . This shape instability is not a pearling instability ( a shape instability in membrane tubes with tension [42] ) , because it is inhibited by membrane tension ( see supplementary information S . I . 3 in S1 Text ) . Rather , it stems from the energetic cost of shape defects . Indeed , invaginations cannot be perfectly tubular , and the tip and base of the invagination can be viewed as defects that increase the membrane conformation energy . When the invagination takes an almost spherical shape , the tip defect is eliminated , and the cost of the base defect is minimized by making the neck infinitesimal . Indeed , we can see that the transition from tubular to spheroidal shapes reduces the total energy as the reduction in bending energy exceeds the increase of pressure-associated energy ( Fig 4f ) . This membrane shape instability can facilitate membrane scission , a crucial step of endocytosis . Membrane coating is not homogeneous: clathrin , sla2 locate at the tip of invagination , while Rvs167 is usually located at the neck [6] . Many other proteins can associate to membranes , including the F-Bar protein Bzz1 [43] , epsins such as Ent1 [44] , and others [5 , 45] . While we do not know the exact localization , the mechanical properties and the interactions of all these proteins , it is usually believed that the clathrin-coated tip is more rigid than the base of the invagination [32] . This could also cause a higher spontaneous curvature at the tip , and additional spontaneous curvature could also stem from lipid asymmetry in the membrane leaflets . We modeled this possible heterogeneity in rigidity as a region with higher rigidity κ ( the tip ) , a region with a lower rigidity κmin ( the base ) , and a small transition zone ( see supplementary information S . I . 1 . 4 in S1 Text ) . Even in the absence of spontaneous curvature , there is an instability if κmin is too small compared to κ , above a threshold that depends on the surface area having higher rigidity . This instability resembles the curvature instability described above , and also stem from the destabilization of the base of the invagination ( see supplementary information S . I . 1 . 4 in S1 Text ) . Overall , a membrane that is more flexible at the base is more prone to shape instability . The results that were derived for a homogeneous membrane should thus remain qualitatively valid for a heterogeneous membrane as well . BAR-domain proteins are elongated crescent-like objects [46] . It was shown recently that the number of Rvs167 molecules , a BAR-domain coat protein , increases rapidly in the late stages of the endocytic process [6] . They are thought to induce anisotropic curvature in the membrane . Following previous work , we assumed that they favor curvature only orthogonally to the symmetry axis ( Z ) , i . e . they give a favorite radius of curvature R0 [47] , rather than a favorite total curvature . Noting Γ the rigidity of a membrane coated with BAR-domain proteins , the contribution to the energy Eq ( 1 ) reads: F BAR = ∫ ∫ S Γ 2 ( 1 R - 1 R 0 ) 2 d S ( 2 ) This implicitly assumes that the coverage is uniform on the membrane , whereas in reality Rvs167 proteins are localized at the neck of the invagination [6] . This simplification was necessary however to derive the membrane shape equations ( see supplementary information S . I . 1 . 2 in S1 Text ) and sufficient to understand in essence how anisotropic curvature can affect the invagination . It allowed us to predict the membrane shape in the presence of both isotropic curvature effectors ( with parameters C0 , κ ) and anisotropic curvature effectors ( with parameters R0 , Γ ) . We found that anisotropic curvature inhibits the membrane shape instability . The membrane is therefore stabilized and longer invaginations can grow continuously even for high values of C0 ( Fig 6 , bottom , where we used R0 = RΠ for simplicity ) . This is in agreement with in vivo observations that invaginations without Rvs167 cannot grow longer than about 60 nm . As a corollary , removing BAR-domain proteins is a possible mean of triggering membrane scission . In our theory , this corresponds to Γ → 0 , which indeed destabilizes the membrane ( Fig 6 , top ) . This possibility is supported by the recent observation that membrane scission is synchronous with the disappearance of Rvs167 [6] .
Using a general model for membrane mechanics , we could accurately fit experimental profiles shorter than 80nm , even though we assumed the membrane to be homogeneous , i . e . with constant rigidity and spontaneous curvature over the surface of the deformed membrane . In combination with membrane rigidity , we can expect pressure to be the dominant factor opposing membrane invagination during yeast endocytosis , while membrane tension should be negligible . This statement is derived from the dimension of the invagination , and from the scale of pressure determined experimentally , and is thus independent of the details of the model . We estimated the force required to pull the invagination based on the value of turgor pressure , and on the value of the rigidity that was determined by fitting the experimental curves . While the exact value of the force also depends on other parameters and in particular on the height of the invagination ( see Fig 3 ) , its scale is primarily determined by pressure and the width of the invagination . For the measured range of pressure Π ∼ 0 . 2—1 MPa , the force scale is fΠ ∼ 1000—5000 pN . This is significantly larger than previously estimated ( 1—1000 pN [13 , 48 , 49] ) . The corresponding range of value for the rigidity is κ ∼ 400—2000 kB T . This is much stiffer than a pure lipid bilayer , because the membrane is heavily coated at the endocytic site . It was suggested that phase boundaries could play a role by generating a line tension , favorable to membrane budding . However the typical line tension ( of the order of ∼ 0 . 4 pN [50] ) is much smaller than fΠ , and such phenomenon can therefore be discarded . The fact that the scale of the invagination is determined by the ratio of rigidity and pressure ( R Π = κ / 2 Π 3 ) suggests a possible tradeoff . In the absence of any reinforcement , the naked membrane could only make small invaginations with a width of 3 nm . The coat enlarges the invagination by a factor ∼ 10 by increasing the local rigidity , but at the same time also increases the required pulling force . The mechanical properties of the coat thus likely represent an optimal tradeoff between increasing the radius of the invagination ( making individual endocytic events more productive ) and limiting the driving force required from the actin machinery . The ratio κ/Π is likely to have been adjusted in the course of evolution , and is not expected to be conserved across species . In general , coat proteins can induce a negative tension that can favor tubulation [51] , because of their adhesion energy with the membrane . This adhesion energy has been estimated to ω ∼ 10−4 N/m in the case of clathrin [52] , while other coat proteins may have adhesion energies of ω ∼ 10−3 N/m [7] . Using the largest of the two values , we find a pulling force πωR ∼ 30pN , which is insufficient to drive the invagination against the turgor pressure . Moreover , we found that spontaneous curvature actually increases the initial pulling force in these conditions , and consequently coat proteins do not help to initially lift the membrane . This counter-intuitive result is in agreement with observations that clathrin is not necessary to initiate curvature during endocytosis in yeast [8] . This result is specific to the situation where a membrane subjected to a large pressure has to be pulled away from its supporting wall . In this configuration , the invagination must have regions with positive ( at the tip ) and negative ( at the rim ) curvature , and the energetic cost of the latter are significant in the presence of positive spontaneous curvature . The good quality of the fits ( Fig 2 ) indicates that the experimental invagination profiles are consistent with a pulling point force , directed away from the cell wall . In the cell , the forces generated in the actin network are probably transferred to the tip of the invagination over a finite area corresponding to the sla2 proteins . In the model , the point force is effectively propagated over the tip of the invagination , due to the inferred rigidity of the coated membrane . As long as the coat structure remains sufficiently rigid , these two situations should be equivalent . We find that under physiological conditions of pressure , direct constriction by actin in a mechanism similar to cytokinesis is not necessary to explain the shape of the invagination , or specifically the formation of a neck . This result is similar to what was reported in the absence of turgor [29] , and stems from having an isotropic spontaneous curvature that is everywhere the same on the membrane . With isotropic spontaneous curvature , the membrane can have a shape instability ( Figs 4 and 5 ) . A more complicated model that included some level of inhomogeneity ( S . I . 1 . 4 in S1 Text ) , showed that this shape instability is facilitated if the coated membrane is less rigid at the base than at the tip of the invagination ( S . I . 3 . 2 in S1 Text ) . We thus expect the shape instability to be promoted by the inhomogeneity of the system , in particular by the fact that the membrane tip is covered with clathrin while the base most likely is not . Qualitatively , pressure is pushing on the neck , thus constricting it , while membrane tension is pulling on it . Thus the shape instability is promoted by pressure and inhibited by membrane tension ( S . I . 3 . 1 in S1 Text ) . Crescent-shaped curvature effectors such as Rvs167 were modelled as inducing anisotropic curvature . The analysis showed that their presence stabilizes the invagination , suggesting that Rvs167 removal can cause membrane destabilization and scission . It is known that membrane fission is promoted by amphipathic helix insertion and inhibited by crescent BAR domains [53] , and a theoretical argument was made comparing idealized tubes to spherical vesicles . Here we show that crescent-shaped proteins also stabilize intermediate shapes of endocytosis , in what appears to be a very generic phenomenon . It was reported that Rvs167-deprived cells have shorter invaginations [8] . A possible explanation is that BAR domain proteins stabilize invaginations , as in our theory . An alternative possibility is that Rvs167 would be needed to lower the pulling force required at later stages of endocytosis . However , our calculations indicate that the highest forces are required during the initial stages of endocytosis ( Fig 3 ) . Indeed experimentally , the growth speed of the later stages seems independent from the pressure [15] , which is the main force opposing endocytosis . This indicates that stabilization of the shape by BAR domains , rather than reduction of force , is the most likely explanation of the observed shortening of invaginations in Rvs167 mutants . The homogeneous theory could not fit large membrane deformations ( > 80nm ) . Indeed , Rvs167 appears at the neck for large invaginations [6] , and probably induce inhomogeneous properties near the base . This effect was not included in the inhomogeneous model described in ( S . I . 1 . 4 in S1 Text ) , because introducing several different regions makes it technically intractable . Endocytosis is ubiquitous in nature and in other model systems such as animal cells , it appears to involve a similar set of molecules . The physical conditions can however vary greatly in different cells , and this should be reflected in some of the requirements that have constrained the evolution of the endocytic machinery . In the absence of significant hydrostatic pressure , membrane coating is sufficient to generate invaginations [7] , while dissipative processes involving actin and/or dynamin machinery appear necessary to membrane severing [54 , 55] . For the yeast system that we examined , pulling the membrane away from the cell wall seems sufficient to induce a complete budding event . The next crucial step is to decipher the mechanism by which the actin cytoskeleton is able to exert the required force . Precise quantitative information is available to do so [4 , 6] and theoretical work is under way [13] . Ultimately , this will offer a unified model incorporating both actin cytoskeletal dynamics and membrane mechanics , based on the experimental observations .
|
Cells use endocytosis to intake molecules and to recycle components of their membrane . Even in its simplest form , endocytosis involves a large number of proteins with often redundant functions that are organized into a microscopic force-producing “machine” . Knowing how much force is needed to induce a membrane invagination is essential to understand how this endocytic machine may operate . We show that experimental membrane shapes are well described theoretically by a thin sheet elastic model including a difference of pressure across the membrane due to turgor . This allows us to integrate the different contributions that shape the membrane , and to compute the forces opposing membrane deformation . This calculation provides an estimate of the pulling force that must be generated by the actin machinery in yeast . We also identify a membrane instability that could lead to vesicle budding .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Membrane Mechanics of Endocytosis in Cells with Turgor
|
Lassa virus ( LASV ) is a causative agent of hemorrhagic fever in West Africa . In recent years , it has been imported several times to Europe and North America . The method of choice for early detection of LASV in blood is RT-PCR . Therefore , the European Network for Diagnostics of ‘Imported’ Viral Diseases ( ENIVD ) performed an external quality assessment ( EQA ) study for molecular detection of LASV . A proficiency panel of 13 samples containing various concentrations of inactivated LASV strains Josiah , Lib-1580/121 , CSF , or AV was prepared . Samples containing the LASV-related lymphocytic choriomeningitis virus ( LCMV ) and negative sera were included as specificity controls . Twenty-four laboratories from 17 countries ( 13 European , one African , one Asian , two American countries ) participated in the study . Thirteen laboratories ( 54% ) reported correct results , 4 ( 17% ) laboratories reported 1 to 2 false-negative results , and 7 ( 29% ) laboratories reported 3 to 5 false-negative results . This EQA study indicates that most participating laboratories have a good or acceptable performance in molecular detection of LASV . However , several laboratories need to review and improve their diagnostic procedures .
Lassa fever was first described in 1969 as the cause of a nosocomial outbreak of hemorrhagic fever in Nigeria [1] . Lassa fever is an acute viral infection associated with a wide spectrum of disease manifestations , which range from mild courses to multiorgan failure [1–3] . The etiologic agent of Lassa fever is Lassa virus ( LASV , family Arenaviridae , genus Arenavirus ) [4] . The natural host of LASV is the small rodent Mastomys natalensis , which lives close to human settlements [5] . The rodents can become chronically infected at birth and excrete infectious virus in urine and other body fluids , with subsequent transmission to humans [6] . There is evidence of human-to-human transmission in both hospital and community settings [7] . The fact that LASV may be transmitted from human to human gives rise to nosocomial or community-based outbreaks . LASV is endemic in the countries of Nigeria , Liberia , Sierra Leone , and Guinea [8 , 9] and was detected in Mali [10 , 11] . Seroepidemiological studies and imported cases of Lassa fever indicate that arenaviruses circulate somewhere in the region comprising Côte d’Ivoire and Burkina Faso [12] . The annual incidence is estimated at 300 , 000 cases , with 5 , 000 fatalities per year [13 , 14] . Additionally , LASV has been introduced several times into Europe , Japan , and North America . Among the hemorrhagic fever viruses of risk group 4 ( such as Crimean-Congo hemorrhagic fever , Ebola , and Marburg virus ) , LASV has been most frequently imported [15] . The virus usually is imported by returning travelers [16 , 17] . Within Europe , LASV infections have been imported to Germany [18 , 19] , The Netherlands [20] and the United Kingdom [21] . Laboratory testing is required to establish a diagnosis , as Lassa fever can hardly be distinguished from other febrile diseases based on clinical symptoms [14 , 22] . A suspected case must be rapidly ruled out or verified to facilitate appropriate case management , including treatment , the implementation of isolation measures , or the tracking of contact persons [18] . The method of choice for early detection of LASV in blood is reverse transcription ( RT ) -PCR [23–29] . However , the high degree of genetic variability of the virus poses a problem with the design of RT-PCR assays for the reliable detection of all virus strains [30] . The performance of the different techniques applied for molecular diagnosis of LASV may vary between laboratories . External quality assessment ( EQA ) studies for LASV molecular diagnostics have not been performed since 2004 [31] . An EQA study allows the participating laboratories to monitor the quality of their diagnostics and to identify problems with particular diagnostic assays . For these reasons , an EQA study for the molecular diagnosis of LASV was conducted by the European Network for Diagnostics of ‘Imported’ Viral Diseases ( ENIVD ) ( http://www . enivd . org ) in 2013 . ENIVD is concerned with the development of laboratory diagnostic capacities for imported virus infections , quality control , standardization of laboratory procedures , and training of laboratory staff [32] . Based on the results of this study , the quality of LASV diagnostics may be improved .
Twenty-eight laboratories involved in diagnostics of viral hemorrhagic fevers were invited to participate in this study . Invitees were selected from the register of ENIVD network members and from the list of national and regional reference laboratories for rare , emerging , and dangerous viruses . The participation in the study was free of charge . Participants permitted publication of the results in a comparative and anonymous manner . This EQA was coordinated by ENIVD according to the established procedures of the network [33–35] . The proficiency test panel included 13 LASV preparations derived from culture supernatants of Vero E6 cells ( ATCC—American Type Culture Collection ) infected with 4 different LASV strains . Virus in cell culture supernatant was inactivated by heat ( 1 h at 60°C ) followed by gamma irradiation ( 25 kilo gray ) . The test panel consisted of six samples of LASV strain Josiah from Sierra Leone , obtained by serial 10-fold dilution of cell culture supernatant ( 1:10 to 1:106 ) , three samples of LASV strain Lib-1580/121 from Liberia ( dilutions 1:103 to 1:105 ) , LASV strain CSF from Nigeria ( dilution 1:103 ) , and LASV strain AV from Cote d’Ivoire or Burkina Faso ( dilution 1:103 ) . The samples were freeze-dried in 3% mannitol based formulation using an EPSILON 2-6D Pilot Freeze Dryer ( Martin Christ GmbH , Osterode am Harz , Germany ) . In addition , we included one sample containing LASV strain Josiah at a dilution of 1:104 ( sample #14 ) that was prepared with a new dry stabilizer method ( Biomatrica , Inc . , San Diego , CA , USA ) , and one sample containing LASV strain Lib-1580/121 at dilution of 1:104 ( sample #6 ) that was prepared using a liquid stabilizer ( Biomatrica ) . A sample containing lymphocytic choriomeningitis virus ( LCMV ) , the prototype member of the family Arenaviridae , as well as two negative control sera were included in the test panel as specificity controls . After lyophilized sample preparation , the samples were tested and quantified by an in-house real-time PCR assay for quality control purpose . The assay was performed by employing 12 . 5 pmol of forward primer LaV F2 ( CCACCATYTTRTgCATRTgCCA ) , 13 pmol of reverse primer LaV R ( gCACATgTNTCHTAYAgYATggAYCA ) and 5 pmol of probe LaV TM ( FAM-AARTggggYCCDATgATgTgYCCWTT-BBQ ) . The real-time PCR assay was carried out in one-step format on ABI 7500 real-time PCR system using the AgPath-ID One-Step RT-PCR Kit according to manufacturer´s instruction . Plasmid standards were used for the quantification of the genome copies of Lassa virus RNA . The EQA was performed according to National Ethical regulations . Before dispatching the panel , samples were sent to the reference laboratory for testing the quality and obtaining the reference results . Reference laboratory used RT-PCR protocol described by Ölschläger et al . , 2010 [27] . Samples were resuspended in 100 μl of water and the RNA was extracted using the QIAamp viral RNA kit ( Qiagen , Hilden , Germany ) . The presence of LASV or LCMV RNA in the samples was ascertained by RT-PCR and sequencing . The number of LASV genome copies present in these samples was determined by qRT-PCR . Samples were shipped by regular mail at ambient temperature . Participating laboratories were instructed to resuspend the samples in 100 μl of water and to analyze the material like serum samples potentially containing LASV using their routine nucleic acid detection assays . The EQA panel was accompanied by documentation including instructions and an evaluation form for results . Participants were asked to report the assay protocol , the result for each sample , the LASV strain identified , the number of genome copies as well as any problem encountered . To guarantee anonymous data evaluation and reporting , each participating laboratory was coded with an identifier . The results were scored according to detection rate and specificity as in previous EQA studies of ENIVD [33–35] . We assigned one point for correct results; false-negative , false-positive , and indeterminate results did not count . Results were classified as “good”—when all results were correct; “acceptable”—when 1 to 2 results were incorrect; and “need for improvement”—when more than 2 results were incorrect . Results for the sample containing LCMV ( sample #3 ) were not included in the score , as verification of the sequence was optional . In addition , we excluded from scoring the sample containing LASV strain Josiah at a dilution of 1:106 ( sample #15 ) as this concentration is likely to be below the 95%-detection limit of most assays . Thus , obtaining a positive or negative result becomes a matter of chance . Each laboratory received the complete summary of the results in an anonymous way , by which only the own laboratory was recognizable .
This EQA study indicates that most participating laboratories , located in various countries around the world , have a good or acceptable performance in molecular detection of LASV . However , several laboratories need to improve their performance , in particular with respect to detection of the Liberian strain . The data allow the participating laboratories to identify the weakness in their diagnostic procedures and to review and improve their protocols . One published protocol has achieved 100% detection rate reported by single participant . However , the reference laboratory recommends Ölschläger et al . , 2010 published protocol for LASV detection as most commonly used with good detection rate and ability to detect all described Lassa virus strains . The main aim of this EQA study was not to compare published protocols , rather to give chance to participating laboratories to evaluate their testing performance and provide practical exercise for molecular detection of LASV . There should be a follow-up EQA for molecular detection of LASV to evaluate a possible improvement .
|
A proficiency test panel for molecular diagnostic of Lassa virus provides objective evidence of testing quality of International diagnostic laboratories . Since there are no commercial assays available , it is very important to assess the quality of diagnostic test used as well as evaluate detection sensitivity and specificity performance . Participating laboratories have received samples containing different inactivated Lassa virus strains as well as two negative controls . Participants were asked to provide information on diagnostic test procedure and protocols used for analysis of samples of Lassa virus External Quality Assessment ( EQA ) . Based on received information we were able to compare and evaluate the quality of diagnostic profile and facilitate further improvement . Participating laboratories may use results of Lassa virus EQA to become accredited for Lassa virus molecular diagnostic . Since different Lassa virus strains are not available for most of the laboratories , participants achieved very advanced training for diagnostic of rare and imported viruses .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results"
] |
[] |
2015
|
International External Quality Assessment Study for Molecular Detection of Lassa Virus
|
Trinucleotide repeat expansions are the genetic cause of numerous human diseases , including fragile X mental retardation , Huntington disease , and myotonic dystrophy type 1 . Disease severity and age of onset are critically linked to expansion size . Previous mouse models of repeat instability have not recreated large intergenerational expansions ( “big jumps” ) , observed when the repeat is transmitted from one generation to the next , and have never attained the very large tract lengths possible in humans . Here , we describe dramatic intergenerational CTG•CAG repeat expansions of several hundred repeats in a transgenic mouse model of myotonic dystrophy type 1 , resulting in increasingly severe phenotypic and molecular abnormalities . Homozygous mice carrying over 700 trinucleotide repeats on both alleles display severely reduced body size and splicing abnormalities , notably in the central nervous system . Our findings demonstrate that large intergenerational trinucleotide repeat expansions can be recreated in mice , and endorse the use of transgenic mouse models to refine our understanding of triplet repeat expansion and the resulting pathogenesis .
Trinucleotide DNA repeat expansion is the mutational cause of at least 14 neurological , neurodegenerative , and neuromuscular diseases in humans , including Huntington disease , many spinocerebellar ataxias , fragile X syndrome , and myotonic dystrophy type 1 ( DM1 ) . While expansion of other repeats can also be pathological , the majority of the triplet diseases are linked to instability of CTG•CAG repeat sequences [1 , 2] . The size of the inherited expansion is critically linked to the severity of the disease and the age of onset . The most surprising feature of this peculiar type of mutation is the marked tendency of the expanded repeat to further increase in size from one generation to the next , resulting in increased severity of the symptoms and earlier age of onset ( anticipation ) [3 , 4] . Anticipation is particularly evident in DM1 and is explained by the dramatic intergenerational instability of the CTG trinucleotide repeat: very large intergenerational expansions ( “big jumps” ) of several hundred repeats are frequently observed in DM1 families [5] . The repeat , normally ranging between 5–37 CTG , is expanded in DM1 patients , reaching up to 4 , 000 CTG in the most severe congenital form of the disease [6] . In addition , the disease-associated repeat expansion is also unstable in somatic tissues throughout the lifetime of the individual [7 , 8] . Modeling trinucleotide repeat instability in transgenic mice has been a challenge for more than a decade . Transgenic mice have been generated to provide in vivo models to study repeat instability in the germline and during somatic development , as well as to provide models of disease pathogenesis . Initial attempts using cDNA transgenes containing expanded CAG•CTG repeats failed to recreate repeat dynamics . A second generation of mice , carrying longer CAG•CTG tracts , or moderately sized expansions within their native genomic DNA context , reproduced intergenerational and somatic instability [9] . However , large intergenerational length increments of several hundred repeats have never been observed in these animals ( Figure 1A ) . The smaller magnitude of intergenerational expansions in transgenic models raised the question of the adequacy of mice to fully model trinucleotide repeat instability . We report , to our knowledge for the first time , the occurrence in a transgenic mouse model of DM1 of CTG trinucleotide repeat big jumps in association with a strong phenotype and molecular abnormalities . This work demonstrates that large intergenerational mutations can accumulate in a mouse model , and provides a unique tool to further improve our understanding of the metabolism of unstable trinucleotide repeats associated with human disease .
We have previously generated the DM300–328 transgenic mouse line , which carries over 300 CTG repeats embedded in 45 kb of human genomic sequence from the DM1 locus . This line displays the highest frequency of intergenerational instability in mice reported until now ( ∼90% expansions in the offspring ) . Nevertheless , repeat changes never reached more than +60 CTG with a mean of +9 and +20 CTG for maternal and paternal transmissions , respectively [10] . Recently , two different mice derived from the DM300–328 line , carrying 430 and 460 CTG ( male and female , respectively ) , transmitted big jumps of +250 CTG and +480 CTG in a single generation , producing offspring with 680 CTG and 940 CTG ( Figure 1A and 1B ) . The male , carrying a remarkably long 680-CTG tract , transmitted larger expansions to 82% of its offspring ( 11% of transmissions resulted in repeat contraction ) . Among expansions , 11% consisted of further big jumps with a mean repeat gain of +290 CTG . We derived mouse sub-lines called “XL” for mice carrying ∼600–700 CTG and “XXL” for mice carrying ∼900-1000 CTG . Further large size gains were observed in these lines , and a male carrying 1 , 230 CTG was obtained , resulting from an additional +270-CTG expansion . Very large increments of CTG repeats seem to arise more frequently through male than female transmissions , with an overall frequency of 5% ( Figure 1C ) . We have previously shown that mice carrying 350–500 CTG display a mild phenotype , consistent with a trans-dominant effect of toxic RNA transcripts carrying CUG expansions [11–13] . A phenotype was only detected in homozygous mice for the low expressing transgene , suggesting a dose effect of the mutant RNA . Hemizygous mice for the very large expansions of 900 to 1 , 230 CTG did not show an obvious phenotype , indicating that the dramatic increase in CTG repeat size is not sufficient to overcome the low expression of the mutant RNA . However , homozygous mice carrying 700 CTG on one allele and 900 or 1 , 230 CTG on the other displayed a severe phenotype with very high mortality; 60% animals died before seven months of age . These animals exhibited severe growth retardation from birth ( Figure 2 ) and marked splicing abnormalities in the central nervous system and muscle , through expression of toxic CUG-containing expanded myotonic dystrophy protein kinase ( DMPK ) transcripts ( Figure 3 ) . Notably , abnormal splicing of glutamate receptor , ionotropic , N-methyl D-aspartate 1 ( Grin1/Nmdar1 ) and microtubule-associated protein tau ( Mapt/Tau ) transcripts was detected in the brain of homozygous transgenic mice carrying >700 CTG trinucleotide repeats within their human genomic context . In addition , a change in the distribution of muscleblind-like ( Mbnl1 and Mbnl2 ) isoforms either carrying or not carrying exon 7 was observed in the brain of these mice ( Figure 3 ) . To our knowledge , this is the first report of RNA splicing abnormalities in the central nervous system of a mouse model of DM1 . The insulin receptor ( Insr ) and chloride channel 1 ( Clcn1 ) transcripts also showed abnormal splicing patterns in skeletal muscle of the same animals ( Figure 3 ) . In summary , we report remarkable , very large intergenerational changes in CTG repeat length in a mouse model of DM1 . Striking expansions of several hundred repeats in a single generation resulted in the XL and XXL mouse lines , which carry the largest expansion ever observed in mice and display severe phenotypic and molecular abnormalities when homozygous for the very large repeat tract . The first two very large expansions arose in the same line but in mice raised in different animal facilities . The occurrence of these events in the line that we have the most extensively studied , is probably associated with a higher probability of getting very large expansions when the size of the repeat in the transmitting parent is larger , as Figure 1A appears to suggest . However , further mouse breeding strategies , especially with DNA repair–deficient lines , are needed to decipher the mechanisms involved in the formation of these very large expansions . The observation that hemizygous mice carrying very large repeats ( over 1 , 000 CTG ) do not display an obvious phenotype suggests that the expanded RNA dose is not sufficient to induce significant abnormalities . It is possible that the expression of the DMPK transgene might be lowered by the expanded CTG repeat , although this has not been described in DM1 patients . Furthermore , we cannot exclude a heterochromatinization spreading effect of the large repeat on the transgene and/or on the adjacent mouse genomic sequences . If only revealed by homozygous animals , this effect could contribute , at least partially , to the phenotype of mice carrying two very large CTG repeats . Future experiments will address these possibilities . Nevertheless , our results directly challenge the assumption that dramatic intergenerational triplet repeat expansions cannot be recreated in mouse models of unstable DNA . This important proof of principle provides an experimental foundation to further refine our understanding of the metabolism of disease-associated trinucleotide repeats . The lines generated in our study afford a unique tool to explore the complex dynamics of simple trinucleotide repeats , the increasing phenotypic severity through generations , as well as the molecular bases of RNA toxicity in disease pathogenesis .
The transgenic mice used in this study carried 45 kb of human genomic DNA cloned from a DM1 patient as described by Seznec et al . [10] . Animals were bred onto a mixed background ( C57BL6/129/OLA/FVB ) . Transgenic status was assayed by PCR using DMHR4 and DMHR5 oligonucleotide primers [10] . Tail DNA was amplified by PCR with oligonucleotide primers 101 and 102 , and CTG repeat size was determined after electrophoresis of amplification products through a 4% polyacrylamide denaturing gel [10] . For the very large expansions ( ≥700 CTG ) , repeat size was measured after resolving the amplification products on a large 0 . 7% ( w/v ) agarose gel ( 20 × 35 cm ) . Animal care was performed according to institutional guidelines and approved by the Police Prefecture of Paris . Total RNA was extracted from microdissected brain regions using the RNeasy Mini Kit ( Qiagen , http://www . qiagen . com ) . cDNA synthesis was primed with random hexamers . cDNA samples were treated with RNase A at 37 °C for 20 min . Most PCR amplifications were carried out for 21–26 cycles , within the linear range of amplification for each gene . The PCR amplifi-cation for Mapt exon 10 alternative splicing analysis was performed for 28 cycles , given the low abundance of the isoform lacking exon 10 . The following oligonucleotide primers were used: Grin1 , 5′-ATGCCCCTGCCACCCTCACTTTTG-3′ and 5′-GCAGCTGGCCCTCCTCCCTCTCA-3′; Mapt exons 2 and 3 , 5′-ACTCTGCTCCAAGACCAAG-3′ and 5′-TGTCTCCGATGCCTGCTTC-3′; Mapt exon 10 , 5′-CACCAAAATCCGGAGAACGA-3′ and 5′-CTATTTGCACCTTGCCACCT-3′; Mbnl1 exon 7 , 5′-GCTGCCCAATACCAGGTCAAC-3′ and 5′-TGGTGGGAGAAATGCTGTATGC-3′; Mbnl2 exon 7 , 5′-ACCGTAACCGTTTGTATGGATTAC-3′ and 5′-CTTTGGTAAGGGATGAAGAGCAC-3′; Insr exon 11 , 5′-GAGGATTACCTGCACAACG-3′ and 5′-CACAATGGTAGAGGAGACG-3′; Clcn1 exon 7a , 5′-CTTTGTAGCCAAGGTG-3′ and 5′-ACGGAACACAAAGGCACTGA-3′; and TATA box binding protein ( Tbp ) , 5′-GGTGTGCACAGGAGCCAAGAGTG-3′ and 5′-AGCTACTGAACTGCTGGTGGGTC-3′ . PCR products were resolved through 2 . 5% ( w/v ) agarose gels and stained with ethidium bromide .
The National Center for Biotechnology Information GenBank ( http://www . ncbi . nlm . nih . gov/Genbank ) ID numbers for the genes discussed in this article are mouse Clcn1 ( 118425 ) , human DMPK ( 1760 ) , mouse Grin1 ( 14810 ) , mouse Insr ( 16337 ) , mouse Mapt ( 17762 ) , mouse Mbnl1 ( 56758 ) , mouse Mbnl2 ( 105559 ) , and mouse Tbp ( 21379 ) .
|
Many neurological and/or neuromuscular diseases , such as myotonic dystrophy , Huntington disease , and fragile X mental retardation are caused by an increase in the size of a repeated DNA sequence within a specific gene . These repetitive DNA sequences are prone to expansion , increasing in size when transmitted from one generation to the next , which results in more severe symptoms and earlier age of onset . In myotonic dystrophy , the DNA repeat can undergo very large increments of several hundred units ( frequently called “big jumps” ) , usually associated with the most severe clinical picture . Until now , big jumps have not been observed in mice carrying the disease mutation , leading to questions about the adequacy of mice to fully model DNA repeat instability . We now report that these large increments in the size of DNA repeats can occur in transgenic mice , resulting in animals that carry extremely large repeated sequences . These mice are remarkably small and display abnormalities in the metabolism of multiple messenger RNAs , notably in brain and muscle . Our findings strongly support the use of transgenic mice to resolve the complex dynamics of simple repetitive DNA sequences associated with human inherited diseases , and to investigate the molecular events that underlie the development of disease symptoms .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"mus",
"(mouse)",
"molecular",
"biology",
"genetics",
"and",
"genomics",
"neurological",
"disorders"
] |
2007
|
CTG Trinucleotide Repeat “Big Jumps”: Large Expansions, Small Mice
|
Memory storage in the brain relies on mechanisms acting on time scales from minutes , for long-term synaptic potentiation , to days , for memory consolidation . During such processes , neural circuits distinguish synapses relevant for forming a long-term storage , which are consolidated , from synapses of short-term storage , which fade . How time scale integration and synaptic differentiation is simultaneously achieved remains unclear . Here we show that synaptic scaling – a slow process usually associated with the maintenance of activity homeostasis – combined with synaptic plasticity may simultaneously achieve both , thereby providing a natural separation of short- from long-term storage . The interaction between plasticity and scaling provides also an explanation for an established paradox where memory consolidation critically depends on the exact order of learning and recall . These results indicate that scaling may be fundamental for stabilizing memories , providing a dynamic link between early and late memory formation processes .
Memory function consists of different , temporally overlapping stages , roughly divided into working memory , short-term and long-term memory , which are distinguishable by their increasing capacity and storage duration [1] , [2] . Especially long-term memory requires lasting changes which involve synaptic plasticity and , subsequently , other complex and slow physiological and anatomical network processes . Furthermore , the formation of long-term memories relies on memory consolidation ( [3] , for a review see [4] ) . Consolidation , in turn , seems to rely on the intrinsic activation of the network that happens during sleep [5]–[7] . Commonly one distinguishes between two types of consolidation [4] , [8]–[10]: ( i ) systems consolidation which transfers memories from one brain area to another ( e . g . , from hippocampus to neocortex ) and ( ii ) synaptic consolidation which stabilizes memories within a brain area . However , even after consolidation , memories are not ‘frozen’ , thus , new memories learnt can disrupt memories previously learnt and , furthermore , the recall of a memory can destabilize this memory [4] , [11]–[13] . Memories have to be ( re ) consolidated several times to achieve permanence [4] . It is an intriguing problem how the nervous system is capable of distinguishing between memories of different storage duration within the same brain area . Given that memories are represented by synapses [14] , [15] , somehow candidate synapses for long storage duration ( named in the following long-term storage LTS to not confuse this with long-term memory ) must respond differently to those that are involved in short-term storage ( STS ) only . In particular , one would expect that LTS-candidate synapses should be susceptible to synaptic consolidation , while STS-candidates should not . All this happens mainly in the cross-section of the hippocampal and cortical networks , a highly dynamic system continuously driven by inputs as well as by intrinsic activity patterns . In spite of this dynamic volatility , the network is capable of maintaining the synaptic integrity of LTS-candidates for a long enough time such that systems consolidation and other processes can set in . Many computational and psychological memory models describe the dynamics of systems consolidation between hippocampus and neocortex by introducing different time scales for plasticity [16]–[20] . By contrast , experimental evidence [21] indicates that the time scales are about the same . For synaptic consolidation the underlying central difficulty , which makes it hard to design more realistic memory models , is that synaptic plasticity operates at time-scales of seconds to minutes while consolidation takes days . The first steps after memory formation are the processes of protein synthesis [4] and tagging [22]–[24] distinguishing short- from long-term plasticity . They occur on a time scale of minutes to hours after learning . However , synaptic consolidation consists of several steps [4] , [10] and experimental evidences point out that NMDA- and AMPA-receptor reactivations [25]–[27] and sleep [6] , [28] are needed even days later to ( synaptically ) consolidate a new learnt memory . Thus , there is a time-gap between neuronal physiology ( synaptic plasticity; minutes ) and consolidation ( days ) . A physiologically plausible , fully dynamic memory model that bridges such time-spans ( from learning to consolidation ) such that LTS-candidate synapses properly respond to synaptic consolidation , while STS-candidates do not , is still missing . Here we work towards bridging this gap by considering one additional , well-established physiological component which naturally operates at a longer time scale: synaptic scaling [29] . Synaptic scaling has primarily been associated with the homeostatic regulation of activity in a network [30] . Overly active networks will – on a time scale of hours up to days – down-scale their activity and vice versa , which is a result of synaptic scaling , where synaptic weights are regulated by the deviation from a homeostatic level of activity . In the following , we show that neural circuits , which combine synaptic scaling with conventional plasticity [31] , [32] such as long-term potentiation ( LTP; [33] ) , long-term depression ( LTD; [34] ) , or spike-timing-dependent plasticity ( STDP; [35] ) , naturally exhibit a transition from short- to long-term storage , where LTS-candidate synapses are consolidated and maintain their integrity through unspecific , “sleep-like” activation , while STS-candidates fade . This bi-modal characteristic is due to an intrinsically arising nonlinearity that induces – without any addition assumption – a natural bifurcation in the dynamics of the system . Intriguingly , this bifurcation can also explain experimental results [36] on the apparently paradoxical effect of memory destabilization during reconsolidation protocols [11] , [37] , where the recall of a previously learnt aspect actually disrupts its memory . Our model does not attempt to implement any of the complex and still little understood mechanisms for systems consolidation or other long-term processes , which would lead to true long-term memories . Instead , the goal of this study is to present a generic mechanism for dynamically maintaining synaptic integrity of LTS-candidates in the network by synaptic ( re ) consolidation . Thus , this study suggests a solution to the long standing problem of synaptic stability in a fully dynamic network by proposing a bifurcation scenario resulting from combined plasticity and scaling .
Consider , for instance , a model neural circuit ( see Materials and Methods ) of locally connected rate-coded units . Each unit is described by a leaky membrane potential and a firing rate or activity which depends in a non-linear way ( here sigmoidal ) on the unit's actual membrane potential ( ) . This formulation allows for a general interpretation of each unit as either a rate-coded neuron [40] , [41] or a population of neurons [42]–[44] . Thus , the here presented results are independent of the spatial scale of the neural circuit . In the following , we will use the terms ‘unit’ and ‘neuron’ synonymously . In the basic state every neuron receives a small noisy background input of about . For a certain period of time ( here about two hours ) , only a local patch of neurons receives an external input of stronger intensity ( see green striped area in Figure 1 A and green pulse ‘L = Learning’ in Figure 1 B , C; all inputs are noisy ) while others do not and serve as control . This input mimics localized rate-coded signals from the environment or other brain areas delivered to the circuit . In the circuit plastic excitatory synapses to the nearest neighbor neurons exist ( see , for instance , in Figure 1 A the purple area regarding the blue unit ) , as well as , short- and long-range lateral inhibition with unchanging synaptic strengths ( purple and bluish gray area ) . For simplicity , we assume that each unit provides excitatory and inhibitory synapses . The dynamics of the excitatory synapses between neuron and is governed by the combination of synaptic plasticity and scaling defined as [31]: ( 1 ) where is the activity , a time constant of synaptic changes , the ratio of plasticity and scaling time constants , and the desired homeostatic level of activity . As shown in previous work [31] , [32] , the quadratic weight-dependency of the synaptic scaling term guarantees convergent synaptic weights without the need of additional constraints [45]–[49] . The synaptic plasticity part consists only of a correlation-based LTP-term . Analytical and numerical results demonstrate ( see below and Text S1 ) that a synaptic plasticity rule consisting of a combination of LTP and LTD does not alter the general dynamics , we will discuss in the following . Depending on the intensity of the external input , differently strong synaptic weights between the stimulated units are induced by the combined rule of plasticity and scaling ( bottom panels in Figure 1 B , C ) . Thus , the units of the stimulated patch form a local cell assembly similar to those found in recent experiments [50]–[52] and represent a memorized version of the local external input . Small differences in input intensity ( vs . ) induce large differences in weights ( bottom panels , red curves ) . The gray curves represent the controls from neurons that do not receive the strong external input . As we show below , these strong weights differences ( red curves ) arise from a generic nonlinear property of the network , where weight-formation follows a saddle-node bifurcation . This nonlinearity exhibits an intriguing phenomenon: When all units in the circuit ( within and outside the cell assembly ) receive a strong ( ) but brief input ( here about 15 minutes; yellow needles , ‘C1 , C2 = consolidation’ , in panels B , C ) only the strong synapses will recover ( panels C ) , while the weak ones continue to decay ( panels B ) . Here this brief and global input takes the role of the coherent , but unspecific neural activation during slow-wave-sleep , which is commonly considered as a potential basis of synaptic consolidation [5] , [7] . This observation is the first indication that the combination of plasticity and scaling in a simple dynamic model allows differentiating between synapses for short-term storage , which decay , from those for long-term storage , which can be recovered ( or rather consolidated ) . Furthermore , we note that the network has only increased activity during external stimulation . Such a stimulation yields an imbalance in neuronal circuit activity depending on the recurrent synaptic weights . Thus , the learnt cell assemblies are stronger activated than controls and the memory contents stored in the network are read-out ( see below ) . As soon as the external input is not present any more and only background input remains , all activities relax back to background firing rate ( ) although recurrent weights are still high ( Figure 1 B , C ) . This is an important difference to attractor memory models [53]–[55] , which will continue to be active after stimulus withdrawal for ( theoretically ) infinitely long time . This persistent activity is important for explaining the dynamics of working memory ( seconds ) but contradicts the idea of long-term memories which are not permanently active . Here , the memory content is transferred from the input to the synaptic weights [14] . The activities can relax back to background state . We remark that the emergence of the here shown phenomena does not rely on saturation effects and fine tuned topology ( see Text S1 ) . A detailed quantification is provided below . First , we show the impact of a memory recall on the spatial structure of the LTS-synapses . During recall the spatial distribution of weights and activities reveals an interesting competitive effect ( Figure 2 ) , that is important for the formation of different memory cell assemblies and also leads to the paradox of memory loss during recall ( [36] , see below ) . Initially , during learning only the a local patch of units is stimulated and the synapses of their target neurons all grow ( purple square in Figure 2 A; L-phase in Figure 1 C ) , where we have used a strong and local stimulus to drive all synapses into the LTS-regime . Consolidation stimulates the complete network and all synapses within the assembly recover or exceed their initial strengths ( Figure 2 B; C1 , C2-phase in Figure 1 C ) . The process of remembering ( recalling ) a memory is often understood as partial stimulation of an assembly and potentially of some other neurons [56]–[58] . By ways of its learnt connections the assembly produces a filling-in and generates a spatially quite complete excitation pattern including most of its members ( so-called pattern completion ) . According to the literature [1] , [14] , [56]–[60] this represent the behaviorally relevant recall activity . Therefore , only a randomly selected subset of assembly-neurons receives a stimulation ( we used here about with some outliers ) . The resulting network activity clearly shows a filled-in spatial assembly structure ( Figure 2 C; Please note that due to the partial stimulus all units of the assembly are stronger active than the control ones . Thus we can assume that the assembly is completed . ) , where , however , sometimes strongly active neurons are neighbors of weakly active ones . For such constellations the different activities induce a dissimilar weight dynamic . Consider a pair of mutually connected neurons ( see hatching in panel C ) . The weakly active neuron ( but still more active than controls ) induces a small synaptic plasticity term and synaptic scaling is weak , too . By contrast , the synaptic scaling term for the strongly active neuron is large and , thus , dominates the dynamics . As a consequence , the corresponding weight shrinks substantially ( Figure 2 C , inset , yellow bars; see also Text S1 for equations ) . We remark that such network structures with generic lateral inhibition admit separation of different assemblies from each other if learning stimuli do not overlap too much . On the other hand - as soon as overlap exists - activation imbalances , as described above , may lead to interference and competition between different memories . The consequence of this will be discussed in conjunction with the paradox of memory loss during recall [36] at the end of this study . The difference between STS- and LTS-synapses in Figure 1 is a non-linear phenomenon , which is due to a saddle-node bifurcation and as such robust against changes in the stimulation patterns , representing different learning protocols . We tested a range of different input strengths and pulse protocols ( Figure 1 D ) . Generally , for small external inputs the resulting synaptic weights depend roughly linear on the intensity ( Figure 1 E ) with a sudden jump to high values above a certain input intensity . The critical value , where this transition takes place , is insensitive to details in the pulse protocol ( indicated by the strong weight differences shown in Figure 1 F1 , F2 ) . The mechanism inducing this phenomenon is readily understood by investigating the dynamics of this system in more detail . We first analytically calculated the characteristic Weight-Input curve of this system . In the following we will show in an abbreviated form the analytical calculations ( see Text S1 for more details ) . We assume that the long-range inhibition separates the circuit into two ( or more ) subnetworks: ( i ) the externally stimulated local patch ( es ) and ( ii ) the unaffected control units . This enables us to average Equation 1 over all units within such a subnetwork . To calculate the fixed point of the resulting mean field differential equation we set it equal to zero and solve it . As result we receive the weight-nullcline of the system ( The weight-nullcline is a set of states where weights do not change under the given dynamics . ) : ( 2 ) with as averaged value of variable . Equation 2 describes the resulting strength of the synaptic weights within a subnetwork given the dynamic of plasticity and scaling and a mean neuronal activation . As the maximal activation of each unit can not exceed ( given by the input-output function ) , the maximal possible synaptic weight is given by . The resulting weight-activity function in the phase space is shown in Figure 3 B , C ( blue line ) for the parameters used in Figure 1 . Of course , the course of the function depends on the used synaptic plasticity rule ( the numerator in Eq . 2 ) , but it also shows that the LTP-term ( ) dominates and that additional plasticity mechanisms ( e . g . , LTD [34] or short-term plasticity [61] ) do not alter the basic dynamic ( see Figure S1 in Text S1 ) . The average activity within a subnetwork induces certain synaptic strengths ( Equation 2 ) . In turn , the mean external input ( multiplied by the input weight ) and the average recurrent synaptic weights themselves adapt the average activity . The resulting fixed point of this dynamic is calculated by the mean field differential equation of the membrane potential ( Eq . 4 ) . This yields the activity-nullcline ( In analogy to the weight-nullcline , the activity-nullcline is a set of states where activities do not change . ) : ( 3 ) with membrane resistance and average excitatory ( ) and inhibitory ( ) number of connections per unit within the subnetwork . As and are the only topology-related parameters in this equation ( and Eq . 2 ) , the described dynamics are independent of the detailed topology ( see Figure S2 in Text S1 ) . The activity-nullcline follows roughly the sigmoidal shape of the activation function ( Eq . 5 ) . Furthermore , it shows that external inputs of different intensity delivered to the circuit change the neuronal activation ( see green line in Figure 3 B for compared to the red line in panel C for ) and , therefore , ( via Eq . 2 ) the synaptic weights . The direct influence of the external input on the synaptic weights within a subnetwork can be assessed by calculating the intersections between both nullclines . These intersections are the fixed points of the whole subnetwork ( activity as well as weights ) . The resulting fixed point equation has no closed-form solution and , therefore , has to be solved numerically . Direct simulations of the whole circuit ( Euler-method ) match our theoretical predictions ( Figure 3 A ) . Specifically , we find a saddle node bifurcation where different fixed points are reached for low as compared to high input intensities . For the particular setting displayed in Figure 3 , a continuous regime of fixed points for the weights exists for firing rates below approximately ( Short-Term Storage , STS; green , Figure 3 A ) , while above this frequency , the system jumps to a fixed point regime with substantially larger weights ( Long-Term Storage , LTS; red , Figure 3 A ) . The gray area below STS represents the range of weights found for the randomly stimulated control neurons ( targets of the yellow neurons in Figure 1 A ) . Note , to obtain this curve we assumed that the circuit consists of several roughly independent subnetworks . This means that in one circuit different fixed points are reached at different spatial locations . For instance , in Figure 1 C after local stimulation the ( local ) patch is in the LTS-regime ( about in Figure 3 A ) while the control units are weakly stimulated and , therefore , they are in the gray control regime ( about ) with small synaptic weights . The bifurcation is essential for the dynamics discussed here . Using different parameter values for the system does not change the fixed point curve significantly ( see , e . g . , Figure 4 B and Figure S3 in Text S1 compared to the used setting shown in Figure 4 A and Figure 1 B , C ) . However , if one parameter is changed dramatically an adequate adaption of the other parameters can still guarantee the desired circuit dynamics ( see Figure 4 C , D ) . Thereby , the range of parameters remains in a physiological regime . The emergence of the ( desired ) form of the vs . function can be explained by the changing locations of the nullclines in the phase space ( Figure 3 B , C ) . For small input frequencies , the nullclines intersect at three different points and , therefore , two stable and one unstable fixed points exist ( green and red markers in Figure 3 B ) . As weights gradually start to grow from low values , the system gets trapped in the lower stable fixed point in the STS-domain . For high input frequencies only one stable fixed point exists which is in the LTS-domain ( Figure 3 C ) . As soon as the strong external input ends , only the lower fixed point exists and the weights start to decay and , without further inputs , reach control values after maximally ten days ( Figure 3 D and Text S1 ) . However , brief consolidation inputs prevent this as discussed next . Bifurcation analysis also helps to understand why synapses with values in the upper fixed point regime ( LTS-synapses ) respond to global and unspecific consolidation inputs while others do not . Weight changes strongly differ for differently strong initial weights when presenting a single consolidation stimulus ( Figure 5 A and Figure S6 in Text S1 ) . Weights above the bifurcation threshold ( dashed line ) are increased substantially , while those under the threshold are almost unaffected ( close beneath threshold they rather decrease due to the lateral inhibition , see Figure 1 A , top ) . This phenomenon is robust against the duration of the consolidation stimulus ( Figure S4 in Text S1 ) . As a consequence , while all weights decay after learning , consolidation will recover those above bifurcation threshold . Hence , consolidation must not come too late , or also those weights might have dropped beneath threshold from which they cannot be recovered ( Figure 5 B ) . Note , this phenomenon is not “history dependent” , which means it does not matter whether learning or consolidation had driven the weights into the LTS-regime before decay has set in ( Figure S5 in Text S1 ) . We remark that our model solely captures dynamic network effects and that we do not attempt to model systems consolidation , which relies on complex and little-understood physiological processes . It appears , however , important that the here observed dynamic properties of such a network allow synapses to maintain ( and regain ) their stability such that systems consolidation or other processes may find a stable substrate to operate on . The wide parameter range within which this happens ( Figure 5 C ) supports this argument , because recovery is robust and stable . Only if the consolidation input is too short or too late , forgetting sets in . As consolidation is a sleep-induced effect [5] , [6] , [10] , little is known about the actual activity characteristics of the consolidation process . Input intensities required for consolidation are similar to those for initiation ( similar to Figure 1 E ) , but emphasis lies on the fact that for consolidation the whole network is stimulated in an unspecific way and that the consolidation stimuli can be shorter ( in Figure 1 B , C about 15 minutes of total duration ) . Additionally , similar to during sleep induced activations ( e . g . , spindles or ripples [6] ) , the memory-related cell assembly is reconstructed ( “replayed”; see , e . g . , [62] for review ) during the consolidation input ( Figure 2 B ) . The recall of a previously well-learnt memory item may lead to the paradoxical phenomenon that this memory will be less well remembered than a newly learnt one . In the literature , this phenomenon is widely interpreted as memory destabilization or rather disruption [4] , [11]–[13] , [37] , [63] and has been found in some studies [11] , [64] , but not in others [65] , [66] . Thus , the question arises what the dynamical processes are that underlie it and especially also why memory destabilization/disruption depends on details of experimental protocols . In one specific experimental paradigm [36] destabilizing happens due to the interference of a new memory item with the previously learnt first memory , but only if the first memory was recalled before the second was learnt . In this protocol the first memory is impaired , while the new one is now susceptible to consolidation . In the following , we show that combined plasticity and scaling also naturally accounts for this paradox . We compare the experimental paradigm with the collective dynamics of our model system and highlight reasons for the ambivalence about the emergence of this phenomenon [11] , [64]–[66] . In a series of elegant experiments , Walker et al . [36] have shown that destabilization of memory happens during a motor learning task . In a control experiment ( Figure 6 , Protocol 1 ) human subjects were first trained only on one motor sequence ( learning , L1 , blue , day one ) and then tested once on day two ( recall , R2 ) and day three ( recall , R3 ) . Significant improvement in accuracy was observed at day two , but not at day three ( Figure 6 A ) . In the second control experiment ( Figure 6 , Protocol 2 ) subjects had been trained on the first sequence on day one ( L1 , blue ) and on a different , second sequence on day two ( L2 , red ) , hence 24 h later . Testing was done on day three ( R3 , blue and red ) and performance had improved for both sequences equally ( Figure 6 B , blue and red bars ) . Both observations ( panels A and B ) were explained [36] by the overnight consolidation ( C1 , C2 ) of the memory . In the third experiment ( Figure 6 , Protocol 3 ) subjects learnt the first sequence on day one and were – as above – tested on day two ( R2 , blue ) showing the same clear improvement ( Figure 6 C , left blue bar ) . Immediately after testing they had to learn sequence two ( L2 , red ) . When re-tested on the third day ( R3 , blue and red ) performance had significantly improved for sequence two but dramatically dropped for sequence one ( Figure 6 C , right blue and red bars ) . This indicates that the second memory interferes with the first but only when the first is activated before the second was learnt . In our model setting , we performed an identical set of experiments , i . e . , with the same learning and testing sequences as used for the human subjects . The model was set up with two cell assemblies , partially overlapping at a corner . Assembly one ( blue ) was trained on one input sequence and assembly two ( red ) on another sequence . For recall – as explained above ( Figure 2 ) – we stimulate only a randomly selected subset of 30% of the original neurons . Connectivity and all other parameters were the same as before ( Figure 1 ) . Training of either sequence leads to increased synaptic weights which are in the LTS-domain , hence , large enough to allow for consolidation . Consolidation stimuli , C1 and C2 , were applied “at night” , where we briefly ( three times 15 min ) stimulated the whole network ( similar to the procedures in Figure 1 ) , as indicated by the dashed arrows in panels G–I . In these panels one can also see the development of the synaptic weights for the first ( blue ) and the second ( red ) cell assembly for all three experiments . Performance indices of the model ( Figure 6 D–F ) are similar to those for the human experiments and we find that data points for the two control experiments match ( Figure 6 A , D and B , E ) . Moreover , also the non-trivial effect on memory disruption is robustly reproduced by the model ( Figure 6 C , F ) . The weight growth normally happening at consolidation C2 is only visible in the control protocols ( Figure 6 G , H ) . By contrast , the readout that happens for protocol 3 at R2 effectively prevents the first memory from consolidation ( Figure 6 I ) . This phenomenon based on the intrinsic competitive effect arising from activation imbalances already discussed for Figure 2 ( see inset in panel C ) above . This can be seen in panel G here ( see box with magnification ) , as the recalls R2 and R3 yield a reduction of the average weight curve , without inducing transitions from the LTS- to STS-regime . Learning the second memory acts for the first assembly “like a recall” , due to the partial overlap between assemblies . This is visible in panel H ( box ) . Thus , learning a second memory can reduce the average weights of the first one . In panel H all weights are far above threshold and both assemblies can be consolidated . This is different for the last experiment ( panel I ) . Recall R2 together with learning the other sequence L2 pushes the blue curve down more strongly ( see box ) than in panels G and H such that it has dropped under the bifurcation threshold when consolidation C2 happens . Close beneath threshold we remember that consolidation acts disruptive ( see negative parts of the curve in Figure 5 A ) , which leads to a further weight decrease at time point C2 . Panels J–L show the time courses of the fraction of synapses of each cell assembly that are in the LTS-domain , which corresponds to the above discussed effects . We remark that we have set all parameters in this simulation purposefully so that we can in panel I exactly depict the critical bifurcation point , where at C2 the red weights are just above threshold while the blue ones are just below and the first memory is disrupted . This is meant to emphasize that the transition from the LTS- to the STS-regime , which is a qualitative change , is sensitive to the experimental parameters . This might underly the fact that destabilization , which leads to an actual memory disruption , is not always found in real experiments [65] , [66] . While recall and learning of other memories can robustly destabilize a memory , it is the relation of the weight-values relative to the bifurcation threshold , which can give rise to memory disruption ( or not ) . A detailed parameter analysis of the destabilization phenomenon , confirming its robustness , is provided in the supplemental information ( Figure S7 in Text S1 ) . This analysis shows that only , if weights are too big or stimulation for recall is too broad and not competitive enough , transitions from the LTS- to the STS-domain do not happen as the system will not travel through the bifurcation . We remark that several recalls briefly after each other affect the same subset of synapses and , therefore , a destabilized memory can not be destabilized further by applying more recalls . More specifically , we observe that the overlap between the cell assemblies , related to the fraction of reactivated neurons during recall , is the most critical factor which determines whether one assembly can be destabilized ( Figure 7 ) . Zero overlap - trivially so - leads to no disturbance ( not shown ) , small overlap represents the situation which is most strongly susceptible to the disruption of a long term memory ( Figure 7 , left rows ) , where more synapses move from the LTS- to the STS-domain than vice versa . By contrast , for a large overlap both assemblies drive each other up into the LTS-domain ( Figure 7 , right rows ) . Intuitively this makes sense . Large overlap means that both memories are very similar , hence they might as well begin to couple themselves in an associative ( hebbian ) way . For small overlap the ( dis- ) similarity of the memories might rather be “confusing” and an agent ( animal/human ) might benefit from forgetting one of them not being able to decide whether they are the same or different . It would be interesting to investigate this from a psychophysical point of view . We expect that memory similarity is the crucial factor which determines the capabilities of the system for memory maintenance versus destabilization .
Previous theoretical studies have shown that synaptic scaling could play a key role in neural network dynamics . For instance , synaptic scaling assures competition [67] between synapses at the same dendrite and , therefore , can help to distinguish different inputs [68] , [69] . Furthermore , scaling can outbalance neuronal heterogeneities in a way that the performance at working memory tasks is improved [70] . In this study we have shown that synaptic scaling appears a viable candidate mechanism to bridge the large temporal gap between synaptic plasticity ( minutes ) and synaptic consolidation ( days ) , where we have investigated simulated 24 h sleep-waking cycles . Scaling operates on time scales of hours to days [29] and synaptic plasticity on seconds to minutes [33] . Processes on other time scales , for example short-term plasticity [61] , long-term depression ( LTD , [34] ) , or synaptic tagging [22] , [23] , can influence synapses without great impact on the dynamics of our model , because these mechanisms are “temporally close” to the synaptic plasticity part of the learning rule used here ( see Eq . 2 ) . Our analytical and numerical results indicate ( Text S1 and [31] , [32] ) that a different formulation of the synaptic plasticity part will not interfere with the final dynamics as long as the weight-nullcline obeys with which holds for many generic plasticity rules [31] . This constraint also holds for the more complex dynamics of spike-timing-dependent plasticity ( STDP; [35] , [71] ) as strong neuronal activations lead to long-term potentiation ( LTP ) independent of the exact timing of spiking [72]–[74] . In an intermediate activity regime we would expect that STDP together with scaling could yield the emergence of even more complex cell assembly structures which could store spatial-temporal patterns [75]–[79] . Over longer time scales ( on average ) the dynamic of STDP can be simplified by the BCM-rule [47] , [80] , [81] . This rule consists of an LTP- and an LTD-term and , therefore , the phenomena revealed in this study are maintained ( compare also Figure S1 in Text S1 ) . As an important consequence , the bifurcation is preserved under these conditions . Thus , our model with such additional faster synaptic modification mechanisms would exhibit only changed time-courses of the transient synaptic dynamics , for example the learning- or decay times , or more complex structures of cell assemblies , but this would not modify the bifurcation scenario qualitatively and , therefore , the consolidation paradigm presented here . However , not only different plasticity mechanisms can be used , but also the homeostatic term ( here , synaptic scaling ) could be another ( slow ) mechanism adapting synaptic weighs . Note , not every homeostatic term ( e . g . , [46]–[48] ) fulfils the above stated weight-constraint . We considered a class of models of general form ( see Materials and Methods ) . Together with the analytical results this indicates that the phenomenon of synaptic consolidation and differentiation between two storage durations within one network is nearly independent of the underlying network topology ( see Figure S2 in Text S1 ) , plasticity rule considered ( see above ) , details of neuronal and network properties , and type of stimuli . The main requirements , which have to be fulfilled , are: ( i ) a learning rule which guarantees stable synaptic weights depending on the neuronal activity ( ) as assured by the combination of LTP and scaling , ( ii ) leaky , non-linear units ( single neurons or ensembles of neurons ) , ( iii ) an excitatory recurrent network with , on average , long-range inhibition , and ( iv ) ‘local’ external stimuli with increased firing rate . Therefore , the bifurcation and consolidation mechanisms described here are not restricted to a certain brain area . Instead , they can occur in every brain area fulfilling the above requirements . Commonly on assumes for memory the neocortex and hippocampus [4] , [42] , [52] , [82] . Furthermore , the area has to have global activations during sleep [6] which could then serve as the consolidation stimulus . Furthermore , the learning stimulus in this model depends on the input frequency . This means that the cell assembly or memory in this model can correspond to a wide variety of long-term memories represented by Hebbian cell assemblies in the brain [14] , [15] . This includes declarative as well as non-declarative memory types . Often ( computational ) memory models are currently based on attractor neural networks [53] , [54] , [57] , [83]–[85] . In these networks , after the withdrawal of the external input , the activity of a reactivated memory persists for a longer duration [55] , [86] . This feature allows for the use of attractor models to reproduce the ( relatively ) short neuronal dynamics during working memory tasks ( up to ten seconds ) . However , without additional external stimuli these networks are even longer persistently active than the working memory time scale . This means that a reactivated memory in an attractor network will stay active for several minutes or days . Therefore , other mechanisms , as , for instance , inhibitory plasticity [58] , are considered to deactivate the recalled memory . All this seems physiologically problematic . By contrast , in our model activity drops back to the background state after a short period ( Figure 1 ) as the memory is not an attractor of the activity dynamics . This is another important property of our system , which combines dynamic behavior with the possibility for synaptic recovery by consolidation . To enable working memory dynamics within this circuit , our model could be extended by the mechanisms of short-term plasticity [61] , [87] , [88] . However , the drop in activity results in a decay of weights which , due to further mechanisms , could be probabilistic as already proposed by Fusi et al . [89] . The decay of synaptic weights can be avoided by repeatedly delivering brief and global consolidation signals to the network . Here , we assume that such signals can arise during sleep , especially by spindles and ripples [6] . Experimental findings show that , for instance , the disruption of ripples impairs memory consolidation [90] and , furthermore , that synaptic weights are , as in the model , increased after slow-wave sleep or rather spindles [7] . Although we did not include the rich dynamics induced by sleep , our model suggests a potential basis for synaptic consolidation happening during sleep . Furthermore , other experimental studies [25] , [26] show that , even six months after learning , memory needs repetitive inductions of plasticity ( reconsolidation ) . The biological mechanisms of this phenomenon are slightly different to initial synaptic consolidation [91] . However , as in this model , the functional properties of these two events are assumed to be similar [13] , [63] . The dynamics presented here also yield the fact that the model – similar to the real system – remains susceptible to perturbations and we explicitly reproduced the elusive effect of memory disruption by recall [36] . Similar , drug-induced effects had also been reported in a few studies [11] , [37] but others failed to obtain it [65] , [66] . Furthermore , learning something new shortly before or after recall seems to increase the chance of perturbing the old memory [12] , [13] . This ambivalence is hard to account for with other existing memory models but finds a possible explanation in the bifurcation scenario found here . The bifurcation scenario also predicts that relearning of the disturbed memory should be much faster than before as weights are still larger than without learning . Furthermore , memory similarity ( here “assembly overlap” ) has a non-trivial effect on consolidation versus destabilization ( Figure 7 ) . This is a novel and intriguing prediction which may well be tested in psychophysical experiments . In general , it seems that memory has to be repeatedly consolidated [4] , [25] , which could happen during sleep [5] , until it is increasingly stabilized . To achieve the latter , systems consolidation , which also begins during sleep [6] , performs a transition from a dynamic to a more static memory representation . By this , the stored information is transferred to the neocortex [4] . The process suggested here is capable of repeatedly recovering LTS-candidate synapses , while STS-candidates fade . This may , thus , essentially contribute to providing a stable substrate for systems consolidation and other processes .
The network consists of a circuit ( Figure 1 A ) with units . Each unit receives an external input with fixed weight . Furthermore , each unit has plastic excitatory connections to its nearest-neighbors ( purple area in Figure 1 A regarding blue unit ) and constant inhibitory connections to its nearest and next-nearest neighbors ( bluish gray and purple area in Figure 1 A ) . We remark that the specific layout of this topography is not relevant for the results obtained here ( see Figure S2 in Text S1 ) , as long as there is a competition between local excitation and longer-ranging inhibition . Each neuron in the circuit is defined by its leaky membrane potential which changes according to ( 4 ) with membrane time constant , resistance , and external input given by with unchanging input weights . The input is modulated by a noise term drawn each time step from a normal distribution with mean zero and standard deviation . In all simulations the ( abstract ) membrane potential ranges from values about to . The membrane potential is non-linearly transformed to a firing rate by a sigmoidal-function: ( 5 ) where is the maximum firing rate , the steepness of the sigmoidal function , and its inflexion point . All parameters combined specify the input-output behavior of the unit . Only the excitatory synapses in the second layer are modified using the “Synaptic Plasticity and Synaptic Scaling” ( SPaSS ) -rule [31]: ( 6 ) where defines the plasticity rate and the ratio between plasticity- and scaling rate . The desired ‘target’ firing rate of synaptic scaling is given by . A detailed analysis of the properties of this rule is provided elsewhere [31] , [32] . All equations are solved analytically in a mean field approach ( see Results section and Text S1 ) and numerically with the Euler method ( ) . In the following , we provide the parameters used ( if not stated otherwise ) . For numerical simulations , we set , , , thus , the circuit is a 2-d grid . The inhibitory and projection weights are proportional to the maximal possible weight: and with . The neuronal parameters are , , , , and . The here shown results are independent of . Although , a smaller value would be biological more reasonable , we took as this avoids numeric instabilities ( ) . The plasticity parameters are , , and . To avoid boundary effects , we used periodic boundary conditions resulting in a toroidal network topology . In Walker et al . [36] training and recall of memory items differ in the number of blocks each consisting of 30 seconds task followed by 30 seconds rest . Here we use 36 blocks for a training session and 10 blocks for recall . Throughout the task a stimulus of intensity is given to the memory-related neurons ( ) . Consolidation signals consist of three blocks with 15 minutes whole network stimulation ( ) followed by 15 minutes pause . Every time step gaussian noise is added to the external stimuli as mentioned above but with a standard deviation of . For Walker et al . [36] as well as for model results all values in the insets of Figure 6 are average values over 10 trials . Data points ( black dots ) in the main panels have been calculated from the bar plots by us; connecting lines are for graphical reasons only . Performance indices of the model are calculated as time- and space-averages of the synaptic weights across all neurons of the respective sub-populations . The time averages have been obtained over five blocks . These are the five last task blocks used for recalls or learning ( the learning pulses define the value ) .
|
The ability to form memories of the past is a main feature of the brain . Memories are formed by learning . However , the biological mechanisms for learning , which change the synaptic weights by synaptic plasticity , act on a different time scale ( minutes ) than those that lead to memory consolidation ( days ) . Experimental results of the last 15 years show that there exists another mechanism , named synaptic scaling , which also influences synaptic weights but on an intermediate time scale ( hours ) . In this study , we analyse whether this process could bridge the time gap and to what degree it can be used to link the processes of synaptic changes with the slow processes of memory formation ( and forgetting ) . Furthermore , the combination of synaptic plasticity and scaling provides a possible explanation for the effect that memory recall can destabilize existing memories . Thus , our results suggest that synaptic scaling is a fundamental mechanism for the dynamic processes of memory .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Synaptic Scaling Enables Dynamically Distinct Short- and Long-Term Memory Formation
|
The mechanosensitive channel of large conductance ( MscL ) is capable of transducing mechanical stimuli such as membrane tension into an electrochemical response . MscL provides a widely-studied model system for mechanotransduction and , more generally , for how bilayer mechanical properties regulate protein conformational changes . Much effort has been expended on the detailed experimental characterization of the molecular structure and biological function of MscL . However , despite its central significance , even basic issues such as the physiologically relevant oligomeric states and molecular structures of MscL remain a matter of debate . In particular , tetrameric , pentameric , and hexameric oligomeric states of MscL have been proposed , together with a range of detailed molecular structures of MscL in the closed and open channel states . Previous theoretical work has shown that the basic phenomenology of MscL gating can be understood using an elastic model describing the energetic cost of the thickness deformations induced by MscL in the surrounding lipid bilayer . Here , we generalize this elastic model to account for the proposed oligomeric states and hydrophobic shapes of MscL . We find that the oligomeric state and hydrophobic shape of MscL are reflected in the energetic cost of lipid bilayer deformations . We make quantitative predictions pertaining to the gating characteristics associated with various structural models of MscL and , in particular , show that different oligomeric states and hydrophobic shapes of MscL yield distinct membrane contributions to the gating energy and gating tension . Thus , the functional properties of MscL provide a signature of the oligomeric state and hydrophobic shape of MscL . Our results suggest that , in addition to the hydrophobic mismatch between membrane proteins and the surrounding lipid bilayer , the symmetry and shape of the hydrophobic surfaces of membrane proteins play an important role in the regulation of protein function by bilayer membranes .
The biological function of membrane proteins is determined by a complex interplay between protein structure and the properties of the surrounding lipid bilayer [1]–[6] . In particular , the bilayer hydrophobic core couples to the hydrophobic regions of membrane proteins [7]–[10] . The resulting deformations in the lipid bilayer membrane from its unperturbed state can be described quantitatively [11]–[17] using the continuum elasticity theory of membranes [18]–[20] . The energetic cost of protein-induced membrane deformations depends on the protein conformational state as well as on the bilayer material properties , which allows [11]–[17] the lipid bilayer to act as a regulator of protein function . A widely-studied model system for the coupling between membrane protein function and the elastic deformation of lipid bilayers is provided by mechanosensitive ion channels . Mechanosensitive channels are capable of transducing membrane tension into an electrochemical response [21]–[23] by switching from a closed to an open conformational state with increasing membrane tension , allowing cells to sense touch , sound , and pressure . A paradigm of mechanosensation is the prokaryotic mechanosensitive channel of large conductance ( MscL ) [24]–[26] . In particular , biophysical approaches such as patch-clamp experiments and reconstitution of MscL in artificial lipid bilayer vesicles have allowed [24]–[34] a systematic analysis of the relation between lipid material properties and the gating probability of MscL with increasing membrane tension . However , despite its central significance , even basic issues such as the physiologically relevant oligomeric states and molecular structures of MscL remain a matter of debate [26] , [35]–[37] . So far , the oligomeric state and molecular structure of MscL have mainly been studied [24]–[27] , [33] , [35]–[47] using crystallographic , biochemical , and computational approaches . This has led to the identification of a number of possible oligomeric states and molecular structures of MscL . In particular , early low-resolution electron microscopy studies suggested that MscL is a hexamer [39] , while more recent high-resolution x-ray crystallographic studies demonstrated pentameric [40] and tetrameric [46] MscL structures . Do the various reported stoichiometries of MscL induce distinct membrane deformations , yielding distinct functional responses to membrane tension ? More generally , theoretical studies of the energetic cost of protein-induced membrane deformations [11]–[15] have mostly focused on membrane inclusions with a cylindrical or conical hydrophobic shape . But experimental surveys of the protein content in the membranes of , for instance , synaptic vesicles [48] and Acinetobacter baumannii [49] suggest [50] that membrane proteins exhibit great diversity in their oligomeric state and transmembrane shape . What is the relationship between the oligomeric state and hydrophobic shape of a membrane protein and the elastic energy required to accommodate the membrane protein within the lipid bilayer ? In this article we address the above questions on the basis of the continuum elasticity theory of lipid bilayer membranes [18]–[20] . In particular , we generalize the standard framework for calculating the energetic cost of protein-induced membrane deformations [11]–[15] , which was employed previously to understand the basic phenomenology of MscL gating [51]–[54] , to account for non-circular cross sections of membrane proteins . Our methodology establishes a quantitative relationship between the oligomeric state and hydrophobic shape of a membrane protein and the elastic energy required to accommodate the membrane protein within the lipid bilayer membrane . We make quantitative predictions pertaining to the gating characteristics associated with various structural models of MscL and , in particular , show that different oligomeric states and hydrophobic shapes of MscL yield distinct membrane contributions to the gating energy and gating tension . Generally we find that the oligomeric state and hydrophobic shape of a membrane protein are reflected in the energetic cost of the lipid bilayer deformations necessary to accommodate the protein within the membrane . Our results suggest that , in addition to the hydrophobic mismatch between membrane proteins and the surrounding lipid bilayer [11]–[15] , the symmetry and shape of the hydrophobic surfaces of membrane proteins play an important role in the regulation of protein function by bilayer membranes . The results and predictions of our model calculations are described in the Results and Discussion sections . The Models and Methods section provides a detailed mathematical formulation of our analytic methodology linking the hydrophobic shape of membrane proteins to the elastic deformations in the surrounding lipid bilayer membrane .
The basic experimental phenomenology of mechanosensitive gating is captured by a two-state Boltzmann model [27]–[33] describing the competition between the closed and open states of MscL . The central quantity in this model is the channel opening probability ( 1 ) where , in which is Boltzmann's constant and is the temperature , is the total free energy difference between the open and closed states of MscL , is the membrane tension , and is the area difference between the open and closed channel states . Equation ( 1 ) implies that , for a fixed , a given channel is more likely to be in the open state for larger values of the membrane tension and provides a simple description of experimental data on MscL gating [14] , [25] , [27]–[33] , although a more detailed description of MscL gating would need to take into account the existence of multiple conductance states [31] , [33] , [43] , [44] . A deeper understanding of Eq . ( 1 ) in terms of the physical mechanisms underlying MscL gating hinges on a quantitative description of the various contributions to . To this end it is useful [51]–[54] to write as the sum of protein and lipid bilayer contributions , ( 2 ) where denotes the difference in internal protein free energy between the open and closed channel states , and denotes the difference in membrane deformation energy between the open and closed states . In general , depends on the oligomeric state and hydrophobic shape of MscL in the closed and open channel states , as well as on bilayer material properties such as the bilayer hydrophobic thickness and bending rigidity . In the remainder of this article we focus on the membrane deformations induced by MscL . To simplify our notation we therefore drop the subscript in and denote by the membrane deformation energy associated with MscL . The continuum elasticity theory of membranes [18]–[20] provides a general framework for evaluating bilayer-protein interactions [11]–[15] , [55]–[63] and , hence , the membrane contribution in Eq . ( 2 ) . On this basis , the elastic membrane deformations required to accommodate MscL within the bilayer membrane were estimated previously [51]–[54] under the assumption that the transmembrane region of MscL is cylindrical in the closed and open channel states . In particular , it was found that thickness deformations , where and are spatial coordinates along the bilayer membrane , are the dominant elastic membrane deformations induced by MscL . The quantitative details of this previous model of MscL gating , which forms the foundation for the work presented here , are summarized in the Models and Methods section . The overall conclusion of the cylinder model of MscL [51]–[54] is that can be of the same order of magnitude as the measured values of [27] , [29] , [30] , [33] in Eq . ( 1 ) , with both and being ( much ) larger than the thermal energy . This suggests that membrane mechanics plays a central role in mechanotransduction and the biological function of MscL . This conclusion is also consistent with experiments measuring the dependence of MscL gating on membrane composition [27] , [34] . We emphasize , however , that in general the protein contribution to the free energy difference in Eq . ( 2 ) must be considered , and may very well dominate over the membrane contribution . The calculation of the membrane contribution to the gating energy merely represents one step in drawing up a general energy budget of gating . As mentioned above , the determination of the oligomeric state and , more generally , molecular structure of MscL in different conformational states is a problem of intense experimental interest [24]–[27] , [33] , [35]–[47] . How do the observed discrepancies in the oligomeric state and molecular structure of MscL relate to the mechanosensitive gating characteristics relevant for the biological function of MscL ? In order to address this question from the perspective of membrane mechanics we formally divide into two contributions , ( 3 ) where corresponds to the membrane deformation energy associated with the idealized cylinder model of MscL [14] , [51]–[54] , which we employ as our point of reference when estimating the membrane deformations induced by different oligomeric states of MscL , and corresponds to the modification of due to deviations of the hydrophobic cross section of MscL from the circle . In particular , depends on the oligomeric state ( symmetry ) of MscL . We have obtained the analytic solution of the general elastic equations describing bilayer deformations induced by MscL in the limit of weak perturbations about the cylindrical reference shape , thus providing a general framework for estimating for arbitrary oligomeric states . The mathematical details of these calculations are described in the Models and Methods section . As discussed below , we find that the oligomeric state and hydrophobic shape of MscL can have a considerable effect on the membrane deformation energy . Thus , based on the membrane deformation energy , distinct boundary shapes and , in particular , distinct oligomeric states of MscL are predicted to yield distinct mechanosensitive gating curves . A variety of different approaches have been employed [24]–[27] , [33] , [35]–[47] to study the molecular structure of MscL in different conformational states . Figure 1 shows examples of the molecular structures of MscL obtained for Staphylococcus aureus ( SaMscL ) and Myobacterium tubercolosis ( MtMscL ) . In particular , Fig . 1 ( A ) displays the tetrameric structure of SaMscL solved most recently [46] using x-ray crystallography . This structure may correspond to an expanded state which is intermediate between the closed and open states of MscL . Figure 1 ( B ) shows pentameric structures of the closed and open states of MscL proposed for MtMscL using crystallographic , biochemical , and computational approaches . The closed state of MscL displayed in the left-hand panel of Fig . 1 ( B ) was obtained on the basis of x-ray crystallography [40] , while the right-hand panel displays a molecular model suggested for the open state of MscL [43]–[45] . For MscL in Escherichia coli ( EcoMscL ) , hexameric [38] , [39] as well as pentameric [43]–[45] molecular models have been proposed . The contour lines approximating the cross sections of the transmembrane domains in Fig . 1 represent the bilayer-MscL boundary curves used in our membrane-mechanical model of MscL gating . Similar fits are obtained for the hexameric [38] , [39] and pentameric [43]–[45] models proposed for EcoMscL ( in particular , see Fig . 3 in Ref . [39] and Fig . 5 in Ref . [44] ) . The subscript in denotes the oligomeric state ( symmetry ) of MscL with tetrameric , pentameric , and hexameric structures of MscL corresponding to , , and , respectively . As discussed further in the Models and Methods section , we express the bilayer-MscL boundary curves in terms of the variables and , which are the radial coordinate and the polar angle associated with a polar coordinate system having the MscL protein at its center . The cylinder model of MscL [51]–[54] corresponds to choosing and in the closed and open states of MscL , where and are the cylinder radii in the closed and open channel states . However , as apparent from Fig . 1 , the proposed hydrophobic cross sections of MscL [24]–[27] , [33] , [35]–[47] often deviate from a circle . Indeed , inspired by the structural models of MscL in Fig . 1 and Refs . [38] , [39] , [43]–[45] , we distinguish between two basic shapes of boundary curves . The “polygonal boundary curves” correspond to the tetragonal boundary curve shown in Fig . 1 ( A ) ( see Fig . 5 in Ref . [44] for examples of pentagonal boundary curves ) , while the “clover-leaf boundary curves” correspond to the pentameric propeller shapes in Fig . 1 ( B ) ( see Fig . 3 in Ref . [39] for examples of hexameric clover-leaf shapes ) . Following the approach summarized in Eq . ( 3 ) , we employ the cylinder model of MscL [51]–[54] as a means to isolate the role played by the oligomeric state and hydrophobic shape of MscL [24]–[27] , [33] , [35]–[47] in the regulation of MscL by the surrounding lipid bilayer . In particular , in the simplest model of MscL the hydrophobic thickness of MscL is assumed to be constant when transitioning between closed and open channel states [51] , [52] , while a more general model [53] , [54] allows for changes in the hydrophobic thickness of MscL [43] , [44] , [47] . We consider here both models of the hydrophobic thickness of MscL but , to systematically study the role played by MscL shape in MscL gating , focus on the case of a constant hydrophobic thickness ( see the Models and Methods section for details ) . In either case we always use the same hydrophobic thickness when making comparisons between different shapes of MscL so as to isolate the role played by MscL shape . Moreover , in order to compare membrane inclusions of equal size , and in light of the central role played by the protein area in Eq . ( 1 ) , we generally contrast different oligomeric states and hydrophobic shapes of MscL for a fixed area of the hydrophobic cross section . This assumption allows us to make direct comparisons with previous work on bilayer-MscL interactions [51]–[54] , and eliminates any spurious effects resulting from MscL occupying different membrane areas in different oligomeric states , but would need to be relaxed for a more detailed description of the membrane deformations induced by MscL . In particular , we use for the closed and open states of MscL the cross-sectional areas and with nm and nm , which were estimated previously [51] , [52] , [54] for the cylinder model of MscL on the basis of the available structural models of MscL [27] , [33] , [40]–[45] , [47] . Setting the cross-sectional area equal to or fixes the size of the polygonal and clover-leaf shapes , with all other parameters in determined by the respective symmetries and morphologies of the MscL boundary curves . For comparison , we also consider polygonal shapes having the same circumference , rather than the same area , as the cylindrical reference shape in the closed and open channel states . Figure 2 shows the difference in the membrane deformation fields induced by some of the structural models of MscL in Fig . 1 and Refs . [39] , [44] and the cylinder model of MscL [51]–[54] . As described in greater detail in the Models and Methods section , we estimated the membrane deformation field due to a given oligomeric state and molecular structure of MscL by minimizing the elastic membrane energy with respect to the thickness deformation field in the limit of weak deviations from the cylindrical reference shape . In particular , Figs . 2 ( A ) , 2 ( B ) , and 2 ( C ) show the difference in the thickness deformation fields induced by the tetragonal , pentagonal , and pentameric clover-leaf models of MscL in Fig . 1 and Ref . [44] and the cylinder model of MscL . The cross sections of all membrane inclusions in Fig . 2 are of the area corresponding to the closed state of the cylinder model of MscL . The deformation profiles in Fig . 2 demonstrate that the symmetry and shape of the hydrophobic surface of a membrane protein are reflected in the structure of the membrane deformations required to accommodate the protein within the lipid bilayer . Figure 2 allows us to gain some intuition regarding the membrane deformations associated with different oligomeric states and hydrophobic shapes of MscL . First consider the deformation fields in Figs . 2 ( A ) and 2 ( B ) due to polygonal boundary curves . Tetragonal and pentagonal boundary curves yield membrane deformations exhibiting four- and five-fold symmetry , respectively . However , while polygonal boundary curves of four-fold and lower-order symmetry produce considerable deviations from the deformation field of the cylindrical reference shape , the shallow angles of pentagonal boundary curves only produce relatively small deviations . Indeed , for hexagonal and higher-order symmetries the deviations from the cylindrical deformation field are even smaller than those shown in Fig . 2 ( B ) . For clover-leaf shapes , however , the overall deviation from the deformation field induced by the cylinder model of MscL increases with increasing symmetry of the oligomeric state . As illustrated in Fig . 2 ( C ) , clover-leaf shapes of pentameric and higher-order symmetry can , in addition to clover-leaf shapes of lower-order symmetry , yield substantial modifications of the deformation field associated with cylindrical membrane inclusions . Thus , for the polygonal structures of MscL in Fig . 1 and Refs . [39] , [44] the overall deviation from the elastic deformation footprint of the cylinder model of MscL decreases with increasing symmetry , but for clover-leaf shapes the overall deviation becomes more pronounced with increasing symmetry . Figure 3 ( A ) shows the difference in membrane deformation energy between some of the structural models of MscL in Fig . 1 and Refs . [39] , [44] and the cylinder model of MscL [51]–[54] as a function of lipid tail length ( bilayer hydrophobic thickness ) . Irrespective of the oligomeric state or hydrophobic shape of MscL , deviations of the cross section of MscL from the circle , and the corresponding non-trivial structure of the membrane deformation field , are seen to increase the elastic energy required to embed MscL within the bilayer membrane . Consistent with the deformation profiles in Fig . 2 , the elastic energy difference between polygonal shapes of MscL and the cylinder model of MscL is largest for the tetragonal structure in Fig . 1 ( A ) and decreases with increasing symmetry of the oligomeric state , with hexagonal and higher-order boundary curves inducing elastic membrane deformations of essentially the same energetic cost as the cylinder model of MscL . These conclusions do not change if we consider polygonal models of MscL which have the same circumference , rather than the same cross-sectional area , as the cylindrical reference shape . The pentameric clover-leaf shape of MscL in the closed state [see Fig . 1 ( B ) ] induces membrane deformations which carry a greater energetic cost than any of the polygonal shapes considered in Fig . 3 ( A ) . In contrast , due to its decreased deviation from the cylindrical reference shape , the hexameric clover-leaf shape of MscL in Ref . [39] carries a relatively small cost in membrane deformation energy . Overall , Fig . 3 ( A ) shows that the various structural models of MscL proposed in previous studies [24]–[27] , [33] , [35]–[47] , and the polygonal or clover-leaf boundary shapes associated with these structural models , yield considerable differences in the membrane deformation energy required to embed MscL within a lipid bilayer membrane . In Fig . 3 ( B ) we compare the elastic energy difference between the open and closed states of MscL for the structural models of MscL gating in Fig . 1 and Refs . [39] , [44] ( tetragonal shapes in light blue , pentagonal shapes in orange , pentameric clover-leaf shapes in purple , and hexameric clover-leaf shapes in red ) and the cylinder model of MscL [51]–[54] ( black ) . For completeness , we also consider in Fig . 3 ( B ) transitions between a closed pentagonal shape and an open pentameric clover-leaf shape of MscL ( dark blue ) , as well as the reverse case of transitions between a closed pentameric clover-leaf shape and an open pentagonal shape of MscL ( green ) . For all of these plots we used the parameter values characterizing bilayer-MscL interactions estimated in Refs . [51] , [52] with zero membrane tension . As discussed in greater detail in the Models and Methods section , this parameterization of bilayer-MscL interactions allows the systematic study of the effect of the structure of membrane deformations on the gating characteristics of MscL , without the further complications introduced by MscL having different hydrophobic thicknesses in the closed and open channel states . We also include in this plot the total free energy differences between the open and closed states of EcoMscL estimated by Perozo et al . [27] for PC16 , PC18 , and PC20 bilayers at zero membrane tension . In the case of transitions between the polygonal structures in Fig . 1 and Ref . [44] , we again find that the deviation from the cylindrical reference shape is more pronounced for tetragonal shapes than for pentagonal shapes , and that in either case the free energy of gating is increased relative to cylindrical membrane inclusions . In addition , Fig . 3 ( B ) shows that , for transitions between the pentameric clover-leaf shapes in Fig . 1 , the difference in membrane deformation energy between the open and closed states of MscL is strongly decreased relative to cylindrical inclusions . We attribute this to the larger deformation of the circular boundary curve for the closed pentameric clover-leaf shape in Fig . 1 ( B ) [see also Fig . 3 ( A ) ] as compared to the corresponding open pentameric clover-leaf shape . Allowing for ( hypothetical ) transitions between different families of boundary curves , the situation becomes more complex . Transitions from a closed pentagonal to an open pentameric clover-leaf shape show a strongly increased gating energy , whereas transitions from a closed pentameric clover-leaf shape to an open pentagonal shape carry a small penalty as far as the elastic membrane deformation energy is concerned . This trend is amplified if pentagonal shapes of the same circumference , rather than of the same cross-sectional area , as the cylindrical reference shape are considered . In summary , Fig . 3 ( B ) indicates that , for the proposed structural models of MscL gating [24]–[27] , [33] , [35]–[47] , the term in Eq . ( 3 ) is generally of the same order of magnitude as , with different structural models of MscL displaying a characteristic dependence of the sign and numerical value of on the bilayer hydrophobic thickness . Figure 4 provides a systematic comparison of the membrane deformation energy associated with different oligomeric states of MscL for the polygonal and clover-leaf boundary shapes inspired by the molecular models in Fig . 1 and Refs . [39] , [44] ( see Fig . S1 ) . As in Fig . 3B , we used for Fig . 4 the same hydrophobic mismatch for closed and open states of MscL [51] , [52] . For the clover-leaf shapes in Fig . 4 we considered shapes which were perturbed by the same amplitude about the cylindrical reference shape in open and closed states . The left-hand panel of Fig . 4 ( A ) shows a clear progression in membrane deformation energy as a function of the oligomeric protein state , with lower-order clover-leaf shapes being energetically favorable compared to higher-order clover-leaf shapes . All clover-leaf shapes induce a membrane deformation energy which is greater than the deformation energy associated with the cylinder model of MscL [see Fig . S2 ( A ) for more comprehensive results] . The elastic energy differences between the open and closed states of clover-leaf shapes are displayed in the right-hand panel of Fig . 4 ( A ) . We find that the gating energy of clover-leaf shapes decreases with increasing channel symmetry . Intriguingly , oligomeric states of high enough symmetry yield a gating energy which is reduced relative to cylindrical inclusions of the same cross-sectional area ( see Fig . S3 for more comprehensive results ) . The left-hand panel of Fig . 4 ( B ) illustrates the membrane deformation energy of the closed state of MscL for trigonal , tetragonal , pentagonal , and hexagonal boundary curves . In contrast to clover-leaf shapes , the membrane deformation energy corresponding to polygonal inclusion shapes decreases with increasing symmetry , and eventually approaches the deformation energy associated with cylindrical inclusions . For membrane inclusions of equal circumference the convergence of the membrane deformation energies induced by polygonal and cylindrical inclusions is rendered more rapid as compared to membrane inclusions of the same cross-sectional area [see Fig . S2 ( B ) ] . The elastic energy differences between the open and closed states of polygonal boundary curves are illustrated in the right-hand panel of Fig . 4 ( B ) , and exhibit characteristics which are qualitatively different from the corresponding results for clover-leaf shapes in the right-hand panel of Fig . 4 ( A ) . For polygonal shapes the energy difference between the open and closed channel states decreases with increasing symmetry of the membrane inclusion , and is always greater than the elastic gating energy associated with the cylindrical reference shape . These conclusions hold for membrane inclusions of equal circumference as well as inclusions of the same cross-sectional area ( see Fig . S3 ) . Polygonal boundary curves with six-fold or higher-order symmetry yield , for the parameter values appropriate for MscL [51] , [52] , a gating energy which closely approaches the corresponding gating energy associated with the cylinder model of MscL ( see Fig . S3 for more comprehensive results ) . Thus , Fig . 4 predicts systematic trends in the total membrane deformation energy required to accommodate MscL ( or other membrane proteins with comparable hydrophobic surfaces ) within the bilayer membrane , and in the elastic gating energy , as the oligomeric state and protein shape are being varied . We now turn to the dependence of the channel opening probability in Eq . ( 1 ) on the oligomeric state and hydrophobic shape of MscL . It should be emphasized that we thereby focus solely [11]–[15] , [51]–[54] on the lipid bilayer contribution to the total free energy difference between the open and closed channel states , and neglect any contributions to the gating energy due to changes in the internal protein conformation . While it was argued previously [51]–[54] that , in certain situations , the total free energy difference between the open and closed states of MscL can be of the same order of magnitude as the difference in membrane deformation energy between the open and closed states of MscL , other contributions to the free energy difference must generally be considered . Note , however , that our results in Fig . 3 ( B ) indicate that the term in Eq . ( 3 ) capturing contributions to the membrane deformation energy due to deviations of the hydrophobic cross section of MscL from the circle is generally of the same order of magnitude as the elastic energy difference calculated previously using the cylinder model of MscL [51]–[54] . Thus , the structure of lipid bilayer deformations associated with different oligomeric states and shapes of MscL is expected to affect the gating characteristics of MscL . In order to facilitate the systematic investigation of the connection between the oligomeric state and the gating energy of MscL in Fig . 3 ( B ) we employed the parameterization of bilayer-MscL interactions in Refs . [51] , [52] and used the same hydrophobic mismatch for closed and open states of MscL . Applying these parameter values to the fits to the structural models in Fig . 1 and Refs . [39] , [44] we found the gating curves shown in Fig . 5 ( A ) . The tetragonal model of MscL in Fig . 1 ( A ) is seen to gate at a larger tension than the pentagonal model of MscL in Ref . [44] , with both models yielding a larger gating tension than the cylindrical reference shape . In contrast , the pentameric clover-leaf model of MscL in Fig . 1 ( B ) produces a smaller gating tension than the hexameric clover-leaf model of MscL , the cylinder model of MscL , as well as the tetragonal and pentagonal models of MscL . Moreover , for a pentagonal shape of MscL in the closed state and a pentameric clover-leaf shape in the open state , Fig . 5 ( A ) predicts a relatively large gating tension . In contrast , the reverse case of a pentameric clover-leaf shape in the closed state and a pentagonal open state yields a markedly smaller gating tension than any of our other models of MscL gating motivated by Fig . 1 and Refs . [39] , [44] . Figure 5 ( B ) displays the same gating curves as Fig . 5 ( A ) , but using the distinct values of the hydrophobic thickness of the closed and open states of MscL suggested by structural studies of MscL [40] , [41] , [47] . In this parameterization of bilayer-MscL interactions [53] , [54] , closed and open states of MscL are distinguished not only by their hydrophobic cross section but also by their hydrophobic thickness . As a result , gating is driven by a more complex interplay between the energetics of thickness deformations and the structure of membrane deformations induced by a non-circular cross section of MscL . In comparison to Fig . 5 ( A ) , the gating curves in Fig . 5 ( B ) are shifted to a larger tension into the regime of the measured gating tension [30] , [33] for which in Eq . ( 1 ) . Moreover , for the parameter values used in Fig . 5 ( B ) , the gating tension associated with the structural models in Fig . 1 and Refs . [39] , [44] is generally larger than the gating tension of the cylinder model of MscL . Contrary to Fig . 5 ( A ) , Fig . 5 ( B ) implies that the hexameric clover-leaf model of MscL gates at a smaller tension than the pentameric clover-leaf model of MscL . Similarly as Fig . 5 ( A ) , however , Fig . 5 ( B ) predicts that the tetragonal model of MscL gates at a larger membrane tension than the corresponding pentagonal model . Moreover , Figs . 5 ( A ) and 5 ( B ) both imply that for a pentagonal shape of MscL in the closed state , and a pentameric clover-leaf shape in the open state , the gating tension is increased relative to most other scenarios suggested by Fig . 1 and Refs . [39] , [44] , with the reverse result for the case of a closed pentameric clover-leaf shape and an open pentagonal shape of MscL . Collectively , Fig . 5 shows that , even if only membrane contributions to the gating energy are considered , different oligomeric states and hydrophobic shapes of MscL yield considerable and distinctive modifications of the gating characteristics of MscL . In analogy to Fig . 4 , we have also carried out a systematic comparison between the gating characteristics associated with different oligomeric states of MscL for the polygonal and clover-leaf boundary curves inspired by Fig . 1 and Refs . [39] , [44] ( see Fig . S4 ) . For this comparison we used , as in Fig . 5A , the same hydrophobic mismatch for closed and open states of MscL [51] , [52] . As already suggested by the results in Fig . 4 we found that , for clover-leaf shapes , higher-order oligomeric states gate at a smaller membrane tension . Moreover , depending on the oligomeric state considered , clover-leaf membrane inclusions can gate at a smaller or at a larger tension than the cylinder model of MscL . For polygonal shapes , higher-order oligomeric states are also found to gate at a smaller membrane tension than lower-order oligomeric states but , in contrast to clover-leaf shapes , polygonal channels always gate at a larger tension than the cylindrical reference shape . These features of the gating characteristics of polygonal membrane inclusions do not change if inclusions of equal circumference , rather than equal cross-sectional area , are compared , although the differences in the gating tensions associated with the various oligomeric states of polygonal inclusions become less pronounced .
Inspired by structural studies of MscL [24]–[27] , [33] , [35]–[47] we have determined the membrane deformation energy associated with a variety of oligomeric states and hydrophobic shapes of MscL . Our analysis focused on the limit of weak perturbations about the cylinder model of membrane proteins , which was employed previously to study bilayer-protein interactions for MscL [51]–[54] as well as for a number of other membrane proteins [11]–[15] . It would desirable to complement the analytic approach developed here with numerical schemes allowing the accurate solution of the elastic membrane equations for complicated protein shapes . Such numerical schemes will be crucial for connecting membrane-mechanical models of bilayer-protein interactions more closely to the shapes of real membrane proteins . Moreover , in our analysis we have focused solely [11]–[15] , [51]–[54] on contributions to the total gating energy due to thickness deformations of the bilayer membrane . In particular , we did not consider contributions to the free energy difference between the open and closed states of MscL due to changes in the internal protein free energy . While it has been argued [51]–[54] that , at least for some strains of MscL [27] , the thickness deformation energy may play a dominant role in MscL gating , other contributions to the free energy budget must generally be considered . Our mathematical approach for determining the energetic cost of membrane deformations associated with different oligomeric states and hydrophobic shapes of MscL is general and directly applicable to other membrane proteins . Thus , the methodology developed here establishes a quantitative relationship between the oligomeric state and hydrophobic shape of a membrane protein and the elastic energy required to accommodate the membrane protein within the lipid bilayer membrane . However , the quantitative details of our predictions depend on the parameter values characterizing the hydrophobic shape of the membrane protein under consideration . In particular , crucial inputs for our model are the hydrophobic thickness and cross section of membrane proteins . Recent experimental results [3]–[6] , [9] , [10] on bilayer-protein interactions suggest that it may be feasible to substantially refine these model inputs to arrive at a more realistic description of protein-induced membrane deformations . For instance , we assumed here that the hydrophobic surface of MscL is perpendicular to the bilayer membrane and of a constant thickness , while a more realistic description of bilayer-MscL interactions would allow [64] for variations in the hydrophobic thickness of MscL along the bilayer-MscL interface . The physiologically relevant oligomeric states and molecular structures of MscL remain a matter of debate [26] , [35]–[37] , with tetrameric [46] , pentameric [40] , and hexameric [39] states of MscL having been reported . The oligomeric state and molecular structure of MscL have so far mainly been studied [24]–[27] , [33] , [35]–[47] using crystallographic , biochemical , and computational approaches . Our results suggest that , for cases in which there is a significant membrane contribution to the gating energy , functional properties of MscL , such as the predicted discrepancies in the gating energy and gating tension between different oligomeric states and structural models of MscL [24]–[27] , [33] , [35]–[47] , may also be used to shed light on the physiologically relevant oligomeric states and molecular structures of MscL . While we have illustrated our approach for MscL , the methods developed here are general and applicable to other membrane proteins . We predict that the oligomeric state and hydrophobic shape of a membrane protein are reflected in the energetic cost of the lipid bilayer deformations necessary to accommodate the protein within the membrane . Thus , our results suggest that , in addition to the hydrophobic mismatch between membrane proteins and the surrounding lipid bilayer [11]–[17] , the symmetry and shape of the hydrophobic cross section of membrane proteins , and resulting structure of elastic membrane deformations , play an important role in the regulation of protein function by bilayer membranes .
In accordance with the standard framework for describing elastic bilayer-protein interactions [11]–[15] , [51]–[63] , we model MscL as a rigid membrane inclusion inducing bilayer deformations as a result of a hydrophobic mismatch between lipid bilayer and membrane protein . In mathematical terms , the lipid bilayer is represented within the Monge representation of curved surfaces using the functions and , which define the positions of the hydrophilic-hydrophobic interface at the Cartesian coordinates in the top and bottom ( outer and inner ) membrane leaflets . Focusing on thickness deformations induced by MscL [14] , [51]–[54] , we consider the elastic energy [18] , [20] , [56] ( 4 ) where the thickness deformation field is defined by ( 5 ) in which is the equilibrium thickness of the unperturbed bilayer , is the bending rigidity , is the stiffness associated with thickness deformations , and is the membrane tension . Energy functionals of the form in Eq . ( 4 ) have been employed in a range of studies [11]–[15] , [51]–[63] of membrane deformations induced by MscL as well as other membrane proteins . The terms and in Eq . ( 4 ) provide lowest-order descriptions of the energetic cost of membrane bending and compression or expansion of the lipid bilayer , respectively . For generality we allow for the two tension terms and in Eq . ( 4 ) , which were employed previously to describe the effects of membrane tension on lipid surface area [18] , [53] , [54] and on membrane undulations [18]–[20] , [51] , [52] , [56] . While Eq . ( 4 ) provides a simple description of protein-induced membrane deformations , more sophisticated models of membrane deformations can be developed [20] , [52] , [55]–[59] in order to account for detailed properties of lipid bilayers such as lipid structure and spontaneous curvature . Finally , the elastic model of bilayer membranes in Eq . ( 4 ) is completed by accounting for the midplane deformations ( 6 ) To leading order , midplane deformations decouple from thickness deformations in the total membrane elastic energy [56] . It was found previously [14] , [51]–[54] that energetic contributions to MscL gating due to midplane deformations can generally be neglected relative to energetic contributions due to thickness deformations , and we therefore focus here on Eq . ( 4 ) . The specific properties of MscL enter Eq . ( 4 ) through the boundary conditions at the bilayer-MscL interface [12]–[15] , [51]–[54] . For convenience , we specify these boundary conditions along some boundary curve using polar coordinates: ( 7 ) ( 8 ) where is the unit normal vector along the bilayer-inclusion interface . If MscL is described as a cylindrical membrane inclusion [51]–[54] , is a constant and . The quantity corresponds to one-half the hydrophobic mismatch between MscL and the surrounding lipid bilayer , and corresponds to the gradient of the thickness deformation field at the bilayer-inclusion interface . We denote the values of and associated with the closed and open channel states by and , and by and , respectively . The crystallographic structure of the closed state of MscL suggests [40] , [41] , [54] nm , while it has been proposed [41] , [47] , [54] that nm for the open state of MscL . To our knowledge , no experimental estimates of the values of and are available for MscL but , within the membrane-mechanical model of MscL gating , these parameters were found previously [51]–[54] to play a minor role compared to and , and are commonly set to zero . We set in all calculations presented here . An approach alternative to that in Eq . ( 8 ) would allow [55] , [57]–[63] for a free contact slope along the bilayer-inclusion interface . The membrane-mechanical model of bilayer-MscL interactions outlined above yields a qualitative framework for understanding MscL gating , is in broad agreement [14] , [51]–[54] with available experimental data , and provides a machinery for making quantitative predictions . In particular , within the framework of this model , MscL gating is understood on a qualitative level as driven by two competing physical mechanisms . On the one hand , closed channels generally leave a smaller elastic deformation footprint in the membrane , which makes the closed state favorable compared to the open state . On the other hand , in membranes under tension , the increase in membrane area associated with open channels makes this state favorable compared to the closed state . Put differently , MscL gating harnesses the mechanical properties of lipid bilayers for channel function , which penalize the more pronounced membrane deformations which are generally necessary to accommodate larger channels , but favor the relaxation of the tension-inducing loading device [26] , [54] brought about by an increased channel area . This physical picture of mechanosensitive gating [14] , [51]–[54] relies on the implicit assumption that , in the closed state of MscL , and in Eqs . ( 7 ) and ( 8 ) are of a similar or smaller magnitude as in the open state of MscL . While the elastic model in Eq . ( 4 ) provides a general description of membrane shape [12]–[15] , [18]–[20] , quantitative tests of the relevance of this model for mechanosensitive gating rely [14] , [51]–[54] on comparing theoretical estimates of to measured values of . In the absence of reliable measurements of in Eq . ( 2 ) , and presence of large experimental uncertainties , any such comparison can only be of broad character . In the simplest case , the closed and open states of MscL are assumed to take cylindrical shapes with the same hydrophobic thickness , which is then fitted to experimental data . In agreement with the experimental results in Ref . [27] , it is thus found [51] , [52] that varies from to as the lipid tail length is varied from 16 carboxyl groups to 20 carboxyl groups , and that this variation approximately takes the shape of a quadratic function . This result is obtained at zero tension with the fitted hydrophobic mismatch nm , which corresponds to a hydrophobic thickness of MscL matching a PC12 bilayer and lies in between the aforementioned values of and proposed on the basis of the crystallographic structure of the closed state of MscL [40] , [41] and molecular modeling of the open state of MscL embedded in doped bilayers [41] , [47] . For a finite tension , which approximately corresponds to the critical gating tension at which in Eq . ( 1 ) , one finds [54] for the cylinder model of MscL with the values of and proposed on the basis of structural studies of MscL [40] , [41] , [47] that for a model lipid bilayer . This estimate does not involve any free parameters , and agrees quite well with the corresponding experimental estimate in Refs . [30] , [33] . We employ the fitted value nm [51] , [52] in Figs . 2–4 and 5 ( A ) , as well as Figs . S2 , S3 , S4 , for our systematic study of the effect of protein shape on the membrane deformation energy and gating tension . This parameterization of bilayer-MscL interactions allows us to avoid any spurious effects resulting from different hydrophobic mismatches in the closed and open channel states . In Fig . 5 ( B ) we use the estimates nm and nm suggested in Refs . [40] , [41] , [47] , [54] . We follow Refs . [11]–[15] , [51]–[63] and use Eq . ( 4 ) with the boundary conditions in Eqs . ( 7 ) and ( 8 ) as our basic model of the membrane deformations induced by MscL . The Euler-Lagrange equation associated with Eq . ( 4 ) is given by ( 9 ) To proceed , we introduce the function ( 10 ) in terms of which Eq . ( 9 ) reduces to ( 11 ) where ( 12 ) The solution of Eq . ( 11 ) is of the form [11] , [65] ( 13 ) where are solutions of the Helmholtz equations ( 14 ) For the exterior of a circle of radius , the above Helmholtz equations are readily solved by separation of variables [65] , [66] . Thus , for the exterior of a circle , the solution of Eq . ( 11 ) can be written as the Fourier-Bessel series ( 15 ) in which ( 16 ) where and are constants , are modified Bessel functions of the second kind , and we have assumed that membrane deformations decay away from the membrane inclusion [57] . At each order in the Fourier-Bessel series in Eq . ( 15 ) , two boundary conditions at the membrane-inclusion interface are required to fix all constants and . Boundary curves are obtained by fitting the Fourier representation of , ( 17 ) in which we take ( 18 ) and , to the transmembrane cross sections of MscL in Fig . 1 and Refs . [39] , [44] . We focus here on the weak perturbation limit of Eq . ( 17 ) and only consider leading-order terms in . The molecular structures in Fig . 1 and Refs . [39] , [44] suggest two basic families of as models of the hydrophobic cross section of MscL: polygonal boundary shapes and clover-leaf boundary shapes . Polygonal shapes are obtained using the Fourier representation of regular -gons in the complex plane [67] , ( 19 ) in which is the imaginary unit and the tetragonal and pentagonal oligomeric states in Fig . 1 ( A ) and Ref . [44] correspond to and , respectively . Higher orders of in Eq . ( 19 ) yield increasingly sharp polygonal corners . For all polygonal shapes in this manuscript we considered terms up to in Eq . ( 17 ) . As described in the Results section , all parameters in Eq . ( 17 ) are then fixed for polygonal shapes by setting the areas of polygonal shapes equal to the cross-sectional areas of closed and open MscL suggested by structural studies [27] , [33] , [40]–[45] , [47] and used in previous membrane-mechanical models of MscL gating [51] , [52] , [54] . The clover-leaf shapes in Fig . 1 are obtained using boundary curves of the form ( 20 ) where the pentameric and hexameric clover-leaf shapes in Fig . 1 ( B ) and Ref . [39] correspond to and , respectively . As for polygonal shapes , the overall coefficient in Eq . ( 20 ) is determined by fixing the area of clover-leaf shapes in closed and open channel states [27] , [33] , [40]–[45] , [47] , [51] , [52] , [54] . For the clover-leaf shapes considered in Figs . 2–5 , we determined through fits to the models of MscL shape shown in Fig . 1 and Ref . [39] , yielding ( closed pentameric clover-leaf shape ) , ( open pentameric clover-leaf shape ) , and ( closed and open hexameric clover-leaf shapes ) . For the model clover-leaf shapes shown in Figs . S1 , S2 , S3 , S4 we used for closed states and for open states so that the amplitude of perturbations about the cylindrical reference shape , , took the same magnitude in closed and open states . In general , and in the boundary conditions in Eqs . ( 7 ) and ( 8 ) at may both exhibit an angular dependence , and our approach is able to handle such cases . Here we focus on the effect of deviations from the circular shape on the elastic membrane deformations induced by MscL . For simplicity , we therefore take and to be constants . Assuming small deviations from circularity in Eq . ( 17 ) , we use a perturbative approach and expand [68] at the boundary curve around to leading order in , ( 21 ) in which ( 22 ) from the general solution in Eq . ( 15 ) to in , where ( 23 ) for . Note , in particular , that any term in Eq . ( 15 ) involving an angular dependence must at least be of in . Similarly , ( 24 ) to leading order in , in which ( 25 ) from the general solution in Eq . ( 15 ) to in , where ( 26 ) for , and is determined by the terms in Eq . ( 15 ) . Thus , using Eqs . ( 21 ) and ( 24 ) , we can recast the boundary conditions in Eqs . ( 7 ) and ( 8 ) for non-cylindrical inclusions as boundary conditions for cylindrical inclusions of variable hydrophobic thickness , ( 27 ) ( 28 ) to leading order in . Matching Eqs . ( 27 ) and ( 28 ) with Eq . ( 15 ) at each order in the Fourier-Bessel series , we find ( 29 ) ( 30 ) ( 31 ) where , for , and . Equations ( 29 ) – ( 31 ) together with Eq . ( 15 ) constitute , in the limit of weak perturbations about cylindrical inclusion shapes , the general solution of the membrane deformation profile for arbitrary oligomeric states of MscL . The membrane deformation energy associated with the equilibrium deformation profile in Eq . ( 15 ) with Eqs . ( 29 ) – ( 31 ) is obtained by evaluating the surface integral in Eq . ( 4 ) . To this end , we note from Eq . ( 11 ) that ( 32 ) Hence , we can use Gauss's theorem in the plane to transform the surface integral in Eq . ( 4 ) to a line integral: ( 33 ) where is a constant . For simplicity , we choose the zero of the energy such that . To evaluate the integrals in Eq . ( 33 ) it is convenient to note that . Substituting the Fourier-Bessel series in Eq . ( 15 ) into Eq . ( 33 ) then generates integrals of the form ( 34 ) Thus , we find the elastic thickness deformation energy ( 35 ) where ( 36 ) ( 37 ) ( 38 ) ( 39 ) ( 40 ) ( 41 ) for . Equation ( 35 ) with Eqs . ( 36 ) – ( 41 ) and Eqs . ( 29 ) – ( 31 ) provides the general solution of the thickness deformation energy in Eq . ( 4 ) for arbitrary oligomeric states of MscL in the limit of weak perturbations about cylindrical inclusion shapes . The deformation profiles in Fig . 2 were obtained from Eq . ( 15 ) with Eqs . ( 29 ) – ( 31 ) , the energy curves in Figs . 3 , 4 , S2 , and S3 were obtained from Eq . ( 35 ) with Eqs . ( 36 ) – ( 41 ) and Eqs . ( 29 ) – ( 31 ) , and the gating curves in Figs . 5 and S4 were obtained from Eq . ( 1 ) together with Eq . ( 35 ) , Eqs . ( 36 ) – ( 41 ) , and Eqs . ( 29 ) – ( 31 ) . For all plots we used the elastic moduli [54] and , with for Figs . 2–4 , S2 , and S3 . The results in Figs . 2–4 , 5 ( A ) , and S2 , S3 , S4 were obtained with nm [51] , [52] . For Fig . 5 ( B ) we used the estimates nm and nm [40] , [41] , [47] , [54] . We used a bilayer hydrophobic thickness corresponding to PC14 lipids for Fig . 1 , to PC18 lipids for Figs . 4 , 5 ( A ) , and S4 , and to PC14 lipids for Fig . 5 ( B ) . We related membrane hydrophobic thickness to PC lipid tail length using the simple interpolation described in Ref . [51] . The primary accession numbers ( in parentheses ) from the Protein Data Bank are: Pentameric MscL ( 2OAR , formerly 1MSL; Resolution of 3 . 50 Å; Ref . [40] ) and tetrameric MscL ( 3HZQ; Resolution of 3 . 82 Å; Ref . [46] ) .
|
A fundamental property of living cells is their ability to detect mechanical stimuli . Microbes , in particular , often transition between different chemical environments , leading to osmotic shock and concurrent changes in membrane tension . The tension of microbial cell membranes is detected and controlled by membrane molecules such as the widely-studied mechanosensitive channels which , depending on the tension exerted by the surrounding lipid bilayer , switch between closed and open states . Thus , the biological function of mechanosensitive channels relies on an interplay between bilayer mechanical properties and protein structure . Using a physical model of cell membranes it was shown previously that the basic phenomenology of mechanosensitive gating can be understood in terms of the bilayer deformations induced by mechanosensitive channels . We have generalized this physical model to allow for the molecular structures of mechanosensitive channels reported in recent experiments . Our methodology allows the calculation of protein-induced membrane deformations for arbitrary oligomeric states of membrane proteins . We predict that distinct oligomeric states and hydrophobic shapes of mechanosensitive channels lead to distinct functional responses to membrane tension . Our results suggest that the shape of membrane proteins , and resulting structure of membrane deformations , plays a crucial role in the regulation of protein function by bilayer membranes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Models",
"and",
"Methods"
] |
[
"physics",
"interdisciplinary",
"physics",
"biology",
"computational",
"biology",
"molecular",
"cell",
"biology",
"biophysics"
] |
2013
|
Connection between Oligomeric State and Gating Characteristics of Mechanosensitive Ion Channels
|
Recent discoveries of direct acting antivirals against Hepatitis C virus ( HCV ) have raised hopes of effective treatment via combination therapies . Yet rapid evolution and high diversity of HCV populations , combined with the reality of suboptimal treatment adherence , make drug resistance a clinical and public health concern . We develop a general model incorporating viral dynamics and pharmacokinetics/ pharmacodynamics to assess how suboptimal adherence affects resistance development and clinical outcomes . We derive design principles and adaptive treatment strategies , identifying a high-risk period when missing doses is particularly risky for de novo resistance , and quantifying the number of additional doses needed to compensate when doses are missed . Using data from large-scale resistance assays , we demonstrate that the risk of resistance can be reduced substantially by applying these principles to a combination therapy of daclatasvir and asunaprevir . By providing a mechanistic framework to link patient characteristics to the risk of resistance , these findings show the potential of rational treatment design .
Hepatitis C virus ( HCV ) affects approximately 170 million people and chronic infections can lead to cirrhosis and hepatocellular carcinoma [1 , 2] . Recently , development of direct acting antivirals ( DAAs ) against HCV infection has revolutionized the field of HCV treatment , because of their high potency , broad applicability and mild side effects [3 , 4] . Combination therapies of DAAs have achieved remarkably high rates of sustained virological response in clinical trials [5–10] . However , most DAAs have relatively low genetic barriers [11–13] , with the exceptions of a few pan-genotypic , yet high-cost DAAs [6] . Because of the high intrinsic mutation rate of HCV [14 , 15] and the resulting high viral diversity [1 , 16 , 17] , combined with the reality of suboptimal treatment adherence [18 , 19] , viral resistance is still a clinical and public health concern [13 , 20] . This is especially true for high-risk groups such as patients with psychiatric disorders or depression [21] , and in resource-limited settings where patients have limited access to clinical cares and cannot afford the expensive pan-genotypic DAAs with high genetic barriers [22 , 23] . If treatment is not properly managed , resistance could quickly develop to combination therapies and render these new DAAs useless , as observed for other antimicrobial treatments , squandering the potential health gains from these recent breakthroughs [24–26] . Suboptimal patient adherence to dosing regimens is a crucial risk factor for resistance development in both HIV and HCV treatments [18 , 19 , 27 , 28] . Although high rates of sustained virological response have been achieved in clinical trials , adherence levels may vary substantially among the vast population of infected patients , owing to long treatment periods , complicated regimens associated with DAA combination therapies and limited access to health care [18 , 19 , 29–31] . Rational design of combination therapy that achieves viral eradication in patients and maximizes the durability of available DAAs in the presence of suboptimal adherence is a research priority [18 , 32–34] . In addition , theories that guide individualized regimens based on the genetic composition of a patient’s infection and real-time adjustments for missed doses are needed to avoid resistance . Mathematical models are well suited to address this problem . Previous modeling studies for HIV infections have illuminated potential mechanisms underlying treatment failure and explained puzzling clinical observations [35 , 36] . However , HCV is a curable disease and its infection , goal of treatment and mechanism of resistance differ from HIV in many respects [37] , including no known latent reservoir and a finite treatment period to eradicate the virus . Here , by integrating pharmacokinetics/pharmacodynamics ( PK/PD ) and viral dynamics into mathematical models , we develop the first general theory to assess the impacts of suboptimal adherence on the outcome of DAA-based therapies for HCV infection . We derive design principles that can be generalized to therapies involving different classes and different numbers of drugs . Using large-scale data from in vitro resistance assays and human clinical trials , we apply this framework to a combination therapy of daclatasvir and asunaprevir [38] , and derive evidence-based adaptive treatment strategies for treatment protocols over time according to resistance profiles and adherence patterns .
The fitness of a particular strain in a treated patient is determined by the PK/PD of the drug , the level of resistance of the strain , and the availability of target cells , i . e . uninfected hepatocytes for HCV ( Fig 1B ) . We can integrate all these factors ( for any class of DAA therapy ) into a single number , the effective reproductive number under treatment , Reff ( t ) ( Fig 1C ) . Reff ( t ) is the average number of cells infected by viruses produced by a single infected cell . It acts as a measure of viral fitness , and can be calculated as: Reff ( t ) = ( 1−ε ( t ) ) ⋅R0⋅h ( t ) ( 1 ) where t is time since treatment starts , τ is the time since last dose , ε ( τ ) is the efficacy of the drug combination at time τ during the dosing cycle , R0 is the reproductive number of the virus in the absence of treatment , and h ( t ) is the normalized abundance of target cells ( see S1 Text ) . Under effective treatment , the availability of target cells , h ( t ) , increases quickly to reach the infection-free level [40] , and therefore , the overall viral fitness increases over time as h ( t ) increases under effective treatment ( Fig 1B and 1C ) . When adherence is optimal ( i . e . no missing doses ) , the value of Reff for a partially resistant mutant is always less than 1 ( i . e . viral suppression ) ; however , if doses are missed , drug concentration declines exponentially and Reff can become greater than 1 ( i . e . viral growth ) ( Fig 1C ) . Note that , although we consider the fitness of a single ‘partially resistant’ mutant here , the competition between different mutants is described implicitly in Eq 2 by the target cell availability , h ( t ) : if a ‘partially resistant’ mutant rises to a high abundance due to missing doses , then h ( t ) will decrease to a low level again , leading to decreases in fitness for all viral mutants . We now consider how suboptimal adherence impacts the dynamics of partially resistant mutants . As an illustration , we contrast simulations assuming perfect adherence versus suboptimal adherence . Missing doses leads to rapid decreases in drug concentration , and thus increases in Reff of a partially resistant mutant ( Fig 2A–2C ) . This means that extra doses are needed to compensate for the missed doses to suppress the mutant to extinction ( Fig 2D ) , and also that the number of newly infected cells rises substantially , which increases the opportunity for de novo resistance ( Fig 2E ) . We approximate the time-varying values of Reff ( t ) during periods when doses are missed , by calculating the average effective reproductive number , Rave , m , as ( see Materials and Methods ) : Rave , m ( t ) ≈ ( 1−εave , m ) ⋅R0⋅h ( t ) ( 2 ) where t is the time when the patient starts to miss doses , m is the number of consecutive doses missed and εave , m is the average drug inhibition during the period when m consecutive doses are missed . This allows us to generalize our theory to any DAA combinations for which εave , m can be either estimated from pharmacokinetics/pharmacodynamics data or calculated from mutant resistance profiles [36] . We then ask , if m consecutive doses are missed beginning at time t , how many extra doses , Nm , are needed to compensate ? This number , which we denote ‘compensatory doses’ , can be approximated as ( see Materials and Methods ) : Nm ( t ) ≈m⋅Rave , m ( t ) ≈m⋅ ( 1−εave , m ) ⋅R0⋅h ( t ) ( 3 ) This allows us to estimate the total duration of treatment needed to clear infection for a given adherence pattern . Furthermore , since h ( t ) increases over time under effective treatment [40] , Eq 3 shows that a higher number of extra doses are needed to eliminate the infection if doses are missed later in treatment . To assess the risk that a partially resistant lineage will give rise to full resistance , we calculate the expected number of target cells , Φm , that become infected by fully resistant mutant viruses due to de novo mutation during a period when m consecutive doses are missed . This quantity is the product of the cumulative number of cells newly infected by a partially resistant mutant and the effective mutation rate from that mutant to the fully resistant mutant , μeff ( see Materials and Methods ) : Φm ( t ) ≈μeff•I ( t ) •Rave , m ( t ) Rave , m ( t ) −1• ( e ( Rave , m ( t ) −1 ) •δ•m•T−1 ) ⏞Θ ( t ) ⏟cumulative number of cells newly infected by a partially resistant mutant ( 4 ) where I ( t ) is the number of cells infected by the partially resistant mutant at time t when the first dose is missed , and Θ ( t ) represents the potential to generate new infections . δ is the death rate of infected hepatocytes , and T is the scheduled interval between two doses . Φm quantifies the risk that a fully resistant mutant infects target cells , but whether it emerges and becomes established within the host depends on its fitness and the stochastic dynamics of invasion [41–43] . The strong dependence of Φm on μeff predicts that designing combination therapies to increase the genetic barrier to full resistance , e . g . using DAAs with higher genetic barrier or adding an extra drug into the combination , can reduce Φm by orders of magnitude or more , thus it would lead to drastic reductions in the probability of generating full resistance ( compare trajectories a and b in Fig 3A ) . Eq 4 also allows us to assess when during treatment it is most risky to miss doses , which can inform treatment guidelines . Changes in two quantities , I ( t ) and Θ ( t ) , determine changes in Φm over the course of a treatment regimen . For as long as adherence is perfect , I ( t ) decreases exponentially , while Θ ( t ) increases over time since Rave , m ( t ) increases as the abundance of target cells rises over time ( Fig 3B ) . Thus the value of Φm first increases ( due to rapid recovery of target cells ) and then decreases exponentially ( due to decrease of infected cells ) . This leads to a high-risk window period , during which missing doses is especially risky for generating full resistance ( Fig 3A ) . This qualitative finding is robust to changes in model parameters , though quantitative predictions of the risk of full resistance depend on the fitness of the mutant ( R0 ) , the half-life of infected cells ( δ ) , and the rate at which the target hepatocytes become available upon treatment ( S1 Fig ) . These results suggest principles for designing combination therapies and rational optimization of treatment outcomes . First , the genetic barrier to full resistance to a therapy is an important determinant of the risk of resistance . Assessment of treatment readiness has been a low-cost routine clinical practice for HIV treatment [44] . Similar strategies can be implemented for HCV treatment . Based on the assessment , for patients who are predicted to maintain high adherence , combinations of DAA that ensure the fully resistant mutants are not pre-existing would be sufficient . For patients with risk factors for low adherence , therapies should be designed by selecting drug combinations that impose a higher genetic barrier than required to suppress all pre-existing mutants . Second , we have shown there exists a high-risk window period where the risk of de novo resistance is high . Intervention efforts to ensure a high level of adherence during the high-risk window period ( indicated by the value of Φm ) would reduce the risk of resistance and treatment failure . Third , because of the exponential growth of ‘partially resistant mutants’ when doses are missed , missing a number of doses consecutively leads to a much higher risk of de novo resistance than missing the same number of doses separately [36] . Thus , missing a block of doses should be avoided . Adaptive treatment strategies could be developed based on the theoretical findings shown above . If doses are missed during treatment , the patient should be treated with extra doses , computed as the maximum value of the Nm values calculated for all partially resistant mutants . For the lowest risk of de novo resistance , the prescribed number of compensatory doses ( Nm ) should be taken , uninterrupted , immediately after doses are missed . Otherwise the infected cell population may rebound to a high level , which can make further missed doses very risky for resistance . To demonstrate the practical applicability of our theory , we consider a recently developed interferon-free combination therapy based on an NS5A inhibitor , daclatasvir , and an NS3 protease inhibitor , asunaprevir [38] . In clinical trials , a large proportion of patients infected with HCV genotype-1b achieved sustained virological response ( i . e . viral eradication ) when treated with daclatasvir and asunaprevir for 24 weeks , although viral breakthrough and viral relapse occurred in a small fraction of patients [45 , 46] . We first consider patients with the wild-type virus at baseline , i . e . the wild-type virus is the dominant strain before treatment . Using the PK/PD data for each drug [47–49] and the resistance profiles data measured for genotype-1b HCV [50 , 51] , we predicted which mutants are potentially fully-resistant to this combination therapy and calculated the values of Nm and Φm for each of the partially resistant mutants ( Fig 4A and 4B ) ( see Supplementary Materials for more detail ) . Choosing the highest values of Nm and Φm among all the partially resistant mutants allows us to project the overall risk arising from missed doses over the course of treatment , and we found required numbers of compensatory doses were modest and the risk of de novo resistance is low ( S2A Fig ) . To demonstrate that the theoretical approximations represent the full viral dynamics accurately , we simulated a multi-strain viral dynamics model ( see Materials and Methods ) , assuming 1–3 day blocks of consecutive doses are missed randomly within a treatment regimen lasting 24 weeks . The model predicts that relapse of L31M+Y93H or L31W would be observed when overall adherence is less than 90% ( Fig 4C and 4D ) . Indeed , the L31M+Y93H mutant has already been detected in one relapse patient in a clinical trial [46] . There is excellent agreement between simulation results and theoretical predictions ( based on Eqs 3 and 4 ) for the number of cells infected by different mutants after 24 weeks of treatment and the cumulative number of cells infected by partially resistant mutants over the treatment period ( Figs 4D and S3 ) . We then simulated outcomes when the doses are guided by the adaptive treatment strategy ( guided dosing; see Methods for detailed simulation procedure ) . Because the risk of de novo resistance when doses are missed is low , there is no high-risk period for de novo resistance in this case ( Fig 4B ) . If patient dosing is guided , i . e . all the required doses and the extra doses to compensate for the missed doses are taken , the infection can be cleared successfully ( Fig 4E ) . Again , we find excellent agreement between simulation results and theoretical predictions ( Fig 4F ) . Many patients bear the Y93H mutation at baseline and this mutation reduces the genetic barrier to full resistance by one nucleotide[46] . Our theory suggests that reducing the genetic barrier to full resistance will drastically increase the risk of treatment failure . We repeated our analysis for patients with Y93H at baseline , to test how our adaptive treatment strategy works when the risk of resistance is high . As predicted , many more days of treatment are needed to compensate for missed doses , and the risks of generating full resistance de novo are high ( >0 . 01 ) during the first 3 weeks of effective treatment if 2 consecutive doses are missed ( or first 4 weeks if 3 doses are missed; Fig 5A and 5B and S2B Fig ) . De novo full resistance is likely if doses are missed randomly and adherence is less than 90% ( dark red area in Fig 5C ) . The predicted number of infected cells agrees well with simulation , except when adherence is very low such that viral load rebounds back close to the pre-treatment level ( Fig 5D and S4–S6 Figs ) . In stark contrast , when doses are guided , the risk of de novo resistance becomes much lower ( compare Fig 5C with 5E ) . Again , for patients who do not clear infection after 24-week treatment , extended periods of treatment as predicted by our theory ( using Eq 3 ) can clear infection with low risk of resistance . The efficacy of the adaptive treatment strategy is robust across different parameter values ( S7–S12 Figs and S1 Text ) . Therefore , our treatment strategy can improve clinical outcomes substantially by adjusting on-going treatment based on patient adherence patterns .
In this study , we integrate PK/PD parameters and viral dynamics into a unified framework to assess the impacts of suboptimal treatment adherence on the risk of treatment failure . Using simulations incorporating PK/PD and resistance profile data collected previously [48 , 50 , 51] , we showed that treatment outcomes of combinations therapies of daclatasvir and asunaprevir can be improved by this adaptive treatment strategy , especially when the Y93H mutant is the dominant strain before treatment begins . We have identified several factors that influence the risk of de novo resistance to a combination therapy . Among these factors , the genetic barrier to full resistance plays a dominant role . Thus , for patients with risk of low adherence , combinations that impose a higher genetic barrier are recommended . This is especially important in resource-limited settings where patients have limited access to health care and adherence is not closely monitored . The recently developed HCV entry inhibitors [52] , which inhibit host factors that are required for viral entry ( instead of viral factors ) , may offer a promising direction for HCV combination therapy , because of their high genetic barriers to resistance , and their synergistic interactions with other classes of DAAs . For situations where therapies with low genetic barriers to resistance are used , we have identified a high-risk window period during which de novo resistance is likely if doses are missed . Intervention efforts should focus on enhancing patients’ adherence during this period . Additional complementary strategies could further reduce the risk of treatment failure . First , if doses are missed during the high-risk window , the immediate addition of another drug with a different mechanism of action from existing drugs may eliminate any low level of fully resistant mutants that has arisen . Alternatively , a patient could be treated preemptively using additional drugs during the entire high-risk period and switched to fewer drugs afterwards . Another important factor is the number of consecutively missed doses as shown previously [36] . Consecutively missed doses lead to exponential growth of ‘partially resistant’ mutants , and thus substantially increase the risk of de novo resistance . Our theory also predicts the number of compensatory doses ( Nm ) needed to compensate for missed doses , in order to eliminate preexisting mutants . Interestingly , clinical trials have shown that adherence levels tend to decrease over time [19 , 31]; we show that more doses are needed to compensate for missed doses that occur later in treatment because of the rebound of target cells . While many previous studies have focused on average adherence [18 , 19 , 29–31 , 36] , we emphasize that the timing of the missed doses is also a critical determinant of treatment outcome and the risk of resistance . There exist substantial heterogeneities among patients owing to variation in HCV genotypes , patient viral loads , death rates of infected cells [40 , 53] and effectiveness of drug penetration [47] . Our analysis has identified several factors that influence the impact of suboptimal adherence , particularly the rebound rate of target cells under treatment , the half-life of infected cells and the overall viral fitness , R0 . We used the best available estimates of these parameters , but further empirical work is needed . If resistance profiles and viral parameters could be measured directly from a specific patient , then our framework linking these factors could be tailored to give patient-specific guidelines . Certain model assumptions reflect uncertainties in our current knowledge of HCV infection . First , our prediction about time to viral extinction should be treated cautiously . We predict the time of extinction ( as in other models [54–56] ) by assuming that infected cells decline at a rate set by their death rate , and infection is cleared when the number of infected cells is below one . However , factors such as pressures from the immune system and infections in different tissue compartments may influence the extinction threshold . Furthermore , if DAA treatment causes intracellular viral RNA to decay with negligible replication [57] , the decline of infected cells may result from a combination of cell recovery and death of infected cells . Indeed , sustained virological response has been observed in clinical trials of DAA combination therapies with shorter durations of treatment [5 , 6] . Our model can be adjusted easily once the decay dynamics of infected cells are understood better . Second , our model captures the main features of pharmacodynamics and viral dynamics by assuming quasi-equilibrium for viral populations and drug penetration into liver cells . Further work that incorporates detailed intracellular interactions [57] and different body compartments may improve model accuracy , once pertinent parameters are measured . However , a more detailed model may become analytically intractable . This quantitative framework is a step towards developing a tool ( for example , see Ref . [58] ) for clinicians to design combination therapies and adaptively manage treatment regimens to achieve favorable clinical outcomes . It highlights the importance of characterizing resistance profiles of HCV , assessing readiness for treatment , and monitoring adherence patterns during treatment , so that treatment can be designed and adjusted in an evidence-based manner . This framework can be adapted easily to combination therapies based on interferon , entry inhibitors [52] or other DAA candidates , or treatments of other curable diseases without a latent reservoir .
To analyze the dynamics of the virus , we constructed an ordinary differential equation ( ODE ) model to describe the long-term within-host dynamics of a single HCV strain under drug treatment , based on an established model developed by Neumann et al . [53] ( see Supplementary Material ) . In the model , ε represents the proportion by which the therapy reduces viral growth ( ε is in the range of 0 and 1 ) . Then , the fitness of the virus , Reff ( t ) , is the product of the complement of the therapy’s efficacy ( 1- ε ( τ ) ) , the reproductive number of the virus , R0 , and the availability of target cells , h ( t ) ( Eq 1 ) . To approximate the time-varying viral fitness , Reff ( t ) , during the period when m consecutive doses are missed , we assume that the abundance of target cells stays constant . This is a good approximation , because the length of the period when consecutive doses are missed tends to be short compared to the time scale of target cell rebound . Then the only time-varying quantity in Eq 1 is ε ( τ ) . We can calculate the average level of drug inhibition during the period when m doses are missed , εave , m , by incorporating parameters for pharmacokinetics and pharmacodynamics ( for example , see Wahl and Nowak[36] ) . Then the time-average effective reproductive number , Rave , m ( t ) , for a mutant when m consecutive doses are missed starting at time t can be expressed as Eq 2 . In practice , because the precise number of target cells at time t is hard to estimate , we can approximate Rave , m ( t ) by setting h ( t ) = 1 , and then Rave , m ( t ) becomes Rave , m ( t ) ≈ ( 1 – εave , m ) ∙ R0 . Because h ( t ) ≤1 , this always overestimates the viral fitness and thus is a conservative estimate in terms of guiding treatment . Note that the assumption that h ( t ) = 1 is valid only when the viral load at time t is much lower than it was before treatment , which is the case if adherence is not too low . Otherwise , h ( t ) would decrease significantly due to large amount of infection . To calculate Nm for each mutant , we make the simplifying assumption that the dynamics of the viral populations are at quasi-equilibrium , because changes in the viral populations occur much faster than changes in infected hepatocytes . Then , the dynamics of the number of cells infected by mutant viruses , I ( t ) , are described by: dI ( t ) d ( t ) = ( Reff ( t ) −1 ) ⋅δ⋅I ( t ) ( 5 ) where δ is the death rate of infected hepatocytes . If we approximate Reff ( t ) using the constant Rave , m for the period when doses are missed , Eq 5 can be solved analytically . Then , the number of infected cells after missing m consecutive doses starting at time t0 can be expressed as: I ( t0+m+T ) ≈I ( t0 ) ⋅exp ( ( Rave , m ( t0 ) −1 ) ⋅δ⋅m⋅T ) ( 6 ) We now consider the situation when m consecutive doses are missed , and ask how many uninterrupted doses ( compensatory doses ) must be taken so that the number of cells infected by the mutant is suppressed to a same number as if the m doses had not been missed . We first calculate the number of infected cells if the m consecutive doses are taken , i . e . if no doses is missed: Ioptimal ( t0+m⋅T ) ≈I ( t0 ) ⋅exp ( ( Rave , 0 ( t0 ) −1 ) ⋅δ⋅m⋅T ) ( 7 ) where I ( t0 ) is the number of cells infected by the mutant at time t0 , Rave , 0 is the average effective reproductive number of the mutant when all doses are taken , and T is the scheduled interval between doses . We then analyze the situation where a patient skips m consecutive doses , starting at time to , and then takes Nm compensatory doses immediately afterwards . In this case , assuming the number of target cells does not change much during this period , we can approximate the number of cells infected by the mutant at the end of the Nm doses as: Isuboptimal ( t0+m⋅T+Nm⋅T ) ≈I ( t0 ) ⋅exp ( ( Rave , m ( t0 ) −1 ) ⋅δ⋅m⋅T ) ⋅I ( t0 ) ⋅exp ( ( Rave , 0 ( t0 ) −1 ) ⋅δ⋅Nm⋅T ) ( 8 ) By equating the right hand sides of Eqs 7 and 8 and solving the equation , we derive the expression for Nm: Nm ( t0 ) ≈Rave , m ( t0 ) −Rave , 0 ( t0 ) 1−Rave , 0 ( t0 ) ⋅m ( 9 ) For potent therapies , usually Rave , 0 ( t0 ) ≈ 0 . Then we get Eq 3 . In the derivation above , we have assumed that the target cell abundance stays constant during the period under consideration . This would be a good approximation if only a few days of doses are missed or if the target cell has already rebounded to the infection-free level . If the abundance of target cells changes considerably during the period under consideration , an alternative , conservative approach would be to assume h ( t ) = 1 and take Nm , max ( t0 ) ≈ m ∙ ( 1 − εave , m ) ∙ R0 compensatory doses after missing m consecutive doses of treatment . One important application of Nm is to predict the number of remaining doses needed to eradicate a mutant , Nerad , in a patient during treatment . This number can be calculated as follows . If adherence is perfect , the number of infected cells declines exponentially at a rate set approximately by the death rate of infected cells , δ: ( t ) ≈ I0 ∙ exp ( −δ ∙ t ) , where I0 is the number of cells infected by a mutant of interest before treatment . If we assume that a mutant goes extinct if the expected number of infected cells in a patient goes below 1 , the number of doses needed to eradicate a mutant before treatment ( assuming adherence is perfect ) , Nerad , 0 , is calculated as: Nerad , 0≈log ( I0 ) δ∙T . When doses are missed during treatment , it is clear from the calculation of Nm above that Nm–m extra doses of treatment are needed to eradicate the virus . Therefore , if a patient has taken a total of x doses and has had k instances of missing doses before time t , with mi days of doses missed in the ith instance ( i = 1 , 2 , … , k ) , then the number of remaining doses needed to eradicate the mutant is calculated as: Nerad=Nerad , 0−x+∑i=1k ( Nm , i−mi ) ( 10 ) We can use Eq 10 to predict the number of cells infected by a mutant as: I ( t ) ≈ exp ( δ ∙ Nerad ( t ) ∙ T ) . In our model , and a patient is cleared of infection when all mutants are driven to extinction . The accuracy of this approximation is shown in Figs 4D and 4F and 5D and 5F . To calculate the risk of full resistance during the period when m doses are missed , we first calculate the number of cells newly infected by a partially resistant mutant when m doses are missed , Ωm ( t ) . Again , we use Rave , m ( t ) to approximate Reff ( t ) , the total number of cells infected by the mutant virus , starting at time t . Ωm ( t ) can be expressed as an integration of new infections during the period of missing doses ( according to Eq 5 ) : Ωm ( t ) ≈Rave , m ( t ) ⋅δ⋅∫tt+m⋅TI ( x ) dx=I ( t ) ⋅Rave , m ( t ) Rave , m ( t ) −1⋅ ( e ( Rave , m ( t ) −1 ) ⋅δ⋅m⋅T−1 ) ( 11 ) The expected number of target cells that become infected by fully resistant mutant viruses , Φm , is a product of the effective mutation rate from the partially resistant mutant to the fully resistant mutant ( μeff ) and the total number of cells infected by the partially resistant mutant ( Ωm ) : Φm ( t ) = μeff ∙ Ωm ( t ) , as shown in Eq 4 . Note that we track the population of newly infected cells to assess the risk of de novo generation of full resistance . This assumes implicitly that the fully resistant mutant is selected only when it enters a cell . This is a good assumption for DAAs that act on intracellular stages of the viral life-cycle , such as viral genome replication or assembly . However , in situations where the drug blocks viral entry into the cell , the mutant virus may have a selective advantage for entering a cell . Then the viral population should be tracked instead , but the results presented here still can be applied to drugs that block cell entry by multiplying with a simple scaling factor [59] . We constructed a simulation model considering the dynamics of the baseline virus and all the potentially partially resistant mutants ( see Supplementary Material ) . This simulation model follows a hybrid approach used previously to simulate the evolutionary dynamics of HIV [60] . It considers the dynamics of multiple strains of HCV deterministically ( using ODEs ) while treating the extinction and mutation processes as stochastic events ( see Supplementary Material for detail ) . In the simulation , a patient is treated for a total period of 24 weeks . We generate two types of dosing patterns: random dosing and guided dosing . For the random dosing pattern , doses are missed in blocks of 1–3 days at times chosen randomly with equal probability during the treatment period . This probability is set as a constant in each run , but varied across runs such that different overall levels of adherence are generated . In each simulation , we assume that at least one-day treatment is taken immediately after each dose-skipping event ( i . e . 1 , 2 or 3 consecutive missed doses ) , to ensure that two dose-skipping events do not occur consecutively ( otherwise , longer blocks of doses would be missed than intended ) . For guided dosing , we ensure that doses are always taken during the high-risk window period predicted by our theory . After this high-risk window period , we set a constant probability of missing doses in blocks of 1–3 days . Immediately after a block of doses is missed , we ensure a sufficient number of uninterrupted doses ( calculated as Nm ) are always taken . If the virus is not eradicated after the 24-week treatment period , the patient is treated with an uninterrupted number of doses as predicted by our theory . The outcome of the simulation at the end of the procedure is reported .
|
Hepatitis C virus ( HCV ) affects approximately 170 million people world-wide and chronic infections can lead to cirrhosis and liver cancer . New combination therapies of direct acting antivirals have achieved remarkably high cure rates in clinical trials . However , high mutation rates and high diversity of HCV populations , combined with the reality of suboptimal treatment adherence , make drug resistance a clinical and public health concern . By constructing a mechanistic framework to assess the risk of drug resistance , we provide guidelines for rational design and adaptive management of these promising new therapies . In particular , we identify a high-risk period when missing doses is particularly risky , and quantify the number of extra doses needed to compensate when doses are missed . This framework is a step towards developing a tool for clinicians to design combination therapies and adaptively manage treatment regimens to achieve favorable clinical outcomes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Rational Design and Adaptive Management of Combination Therapies for Hepatitis C Virus Infection
|
Eukaryotic cells assemble viscoelastic networks of crosslinked actin filaments to control their shape , mechanical properties , and motility . One important class of actin network is nucleated by the Arp2/3 complex and drives both membrane protrusion at the leading edge of motile cells and intracellular motility of pathogens such as Listeria monocytogenes . These networks can be reconstituted in vitro from purified components to drive the motility of spherical micron-sized beads . An Elastic Gel model has been successful in explaining how these networks break symmetry , but how they produce directed motile force has been less clear . We have combined numerical simulations with in vitro experiments to reconstitute the behavior of these motile actin networks in silico using an Accumulative Particle-Spring ( APS ) model that builds on the Elastic Gel model , and demonstrates simple intuitive mechanisms for both symmetry breaking and sustained motility . The APS model explains observed transitions between smooth and pulsatile motion as well as subtle variations in network architecture caused by differences in geometry and conditions . Our findings also explain sideways symmetry breaking and motility of elongated beads , and show that elastic recoil , though important for symmetry breaking and pulsatile motion , is not necessary for smooth directional motility . The APS model demonstrates how a small number of viscoelastic network parameters and construction rules suffice to recapture the complex behavior of motile actin networks . The fact that the model not only mirrors our in vitro observations , but also makes novel predictions that we confirm by experiment , suggests that the model captures much of the essence of actin-based motility in this system .
The directed assembly of actin networks drives the motility of most eukaryotic cells [1] . Specialized cellular factors assemble actin into different network types , each with a unique architecture and cellular function [2] . One of the most well-studied actin assembly factors is the Arp2/3 complex , a seven-subunit protein complex that nucleates new filaments from the sides of pre-existing filaments to create entangled , dendritic filament arrays [3] , [4] . These arrays behave like viscoelastic gels with an elasticity that depends on the degree of branching , and which break or rip under relatively low stress [5] . In vivo , dendritic networks built by Arp2/3 complex form the lamellipod that drives the movement of eukaryotic cells [3] , [6] as well as the “comet tails” whose assembly drives the intracellular movement of endosomes [7] , [8] and intracellular pathogens [9] such as Vaccinia virus [10] and Listeria [11] . Construction of these motile networks in vivo requires a set of highly conserved accessory proteins , including capping protein , cofilin , and profilin , that function together with the Arp2/3 complex in a simple biochemical cycle converting monomeric actin into crosslinked polymer and back again [6] , [12] . Motile , dendritic actin networks can also be constructed in vitro by recombining purified components of the actin assembly cycle [13]–[16] . These reconstituted actin networks have become a powerful tool for studying how individual protein–protein interactions control the large-scale behaviors of cytoskeletal systems . The simplest way to initiate assembly of such motile , dendritic actin networks in vitro is the “bead motility” system , in which micron-sized beads are uniformly coated with factors that activate the Arp2/3 complex to nucleate actin networks at their surfaces [16] , [17] . These networks form spherically symmetric shells that eventually “break symmetry” and produce stable , asymmetric comet tails that propel the bead along , maintaining direction [14] , [16] , [18] , moving smoothly or pulsing depending on conditions [19] , [20] . In this work , we concentrate on how a geometrically and biochemically symmetric bead can first break symmetry then maintain asymmetry to produce directed smooth or pulsatile motion . Spatially localized nucleation of actin filaments combined with global inhibition of filament elongation by capping protein restricts filament growth to a well-defined zone , e . g . , the Listerium surface [21] , lamellipodial plasma membrane [22] , etc . On the spatial scale of filaments , a Brownian ratchet mechanism has been proposed [23] , [24] to explain how actin polymerization uses the energy of ATP hydrolysis to rectify Brownian fluctuations , exerting force at the surface , as new actin monomers , as new actin monomers add onto existing filaments and extend the network . Although the specific details may vary [25]–[27] , spatially localized network extension fueled by ATP hydrolysis is the basis of all polymerization-driven motility models . Several theoretical frameworks have been proposed to explain actin-based symmetry breaking and bead motility ( reviewed in [28] ) . Some are based on filament-scale descriptions of actin assembly and crosslinking [29] , [30] , while others take a more coarse-grained approach based on the bulk mechanical properties of crosslinked polymer networks [17] , [19] , [20] , [31]–[34] . One such coarse-grained model is the Elastic Gel model [19] , [31] , which provides an intuitive explanation for symmetry breaking . In this model , symmetry breaking occurs when new actin network , continuously deposited at the surface of the bead , displaces older portions of the network radially outward . Expansion of the older network stretches it like the surface of an inflating balloon until , at a critical threshold , circumferential stress causes a rupture in the network ( either by melting [33] or cracking [35] the shell ) and breaks the symmetry of the system . This mechanism fits the experimental observations of symmetry breaking [16] , [19] better than mechanisms inferred from filament-based descriptions of the network [30] . Pulsatile motion has been suggested to result from an unstable balance between the pushing forces and the drag from attached filaments [20] . Explaining the smooth directional motility of symmetrically coated beads has proved more challenging . One attempt , the Soap-Squeezing model [31] , is an extension of the Elastic Gel model that offers an explanation of propulsive force . In this model , surface-associated polymerization stretches older network outwards orthogonal to the direction of motion , storing energy , which it releases by contracting orthogonally , squeezing the bead forward like a hand squeezing a wet bar of soap . However , photobleaching data showing the movement of the network as it leaves the bead demonstrate that orthogonal squeezing does not occur [17] , and whereas treating the network as an incompressible fluid flowing from the bead surface can explain the observed motion [17] , this violates the elastic nature of the gel required to explain the initial symmetry breaking . How , then , does sustained motility occur ? In this paper , we examine the essence of actin-based bead motility by reconstituting it in silico from the network's fundamental viscoelastic properties . Just as reconstituting actin-based motility in vitro from a minimal set of purified protein components demonstrates their necessity and can show how they contribute to the large-scale behavior , reconstituting actin-based motility in silico allows us to demonstrate the necessity and specific contributions of a minimal set of higher-level network properties ( e . g . , elasticity , crosslinking , etc . ) , and demonstrate the mechanisms of motility on a mesoscopic scale . To do this , we use a framework we call the Accumulative Particle-Spring model ( APS model ) in which the viscoelastic actin network is represented simply as a set of particles , subject to viscous drag and coupled by springs that break when strained beyond a certain limit . New Particle-Spring network is created at the bead surface , just as the in vitro actin network polymerizes at the bead surface [16] , and we find that this simple system is sufficient to reproduce a range of the behaviors of actin networks , including symmetry breaking and motility . Our simulations enable us to explore the feasibility of hypothesized mechanisms of force and movement generation , using Ockham's razor to determine the essence of the behavior by exploring the minimal requirements to produce the observed results . We validate the model by checking the results and predictions of the simulations with in vitro experiments in which we reconstitute symmetry breaking and motility from purified proteins . To the extent that the model is valid , we are able to make explanatory claims for the mechanisms involved in symmetry breaking and motility , determining 1 ) the stress and strain distributions in a growing symmetric actin shell and in a comet-like tail , 2 ) where the symmetry break is initiated ( outer or inner surface of the actin shell ) , 3 ) the 3-D structure and dynamics of the break , 4 ) what determines the transition from smooth to pulsatile motility , and 5 ) how symmetry breaking occurs for nonspherical objects .
To perform our in vitro bead motility experiments , we evenly coated 5-µm diameter beads with ActA and added them to motility mix ( see Materials and Methods ) . ActA activates Arp2/3 to nucleate an actin network that grows in a tightly localized zone at the bead surface , breaks symmetry , and propels the bead on an actin comet tail ( Figure 1A–1D and Video S1 ) . To find out how well bead motility can be explained simply by the viscoelastic properties of the network , we created a computational model that simulates the behavior of a generic viscoelastic network deposited stochastically at the surface of a bead . The model starts at t = 0 with no network , then nucleates nodes at a constant rate and with an even distribution across the bead surface , crosslinking new nodes to their neighbors with links that behave as simple Hookean springs that break if extended too far ( Figures S1 and S2 ) . See Materials and Methods and Section S1 of the supporting text ( Protocol S1 ) for full details of the model , and Tables S1 and S2 for the experimental bases for the model assumptions . We tuned the model parameters ( spring constant , crosslinking probability , etc . ) to produce qualitatively similar observations to the in vitro system ( see Model Robustness , Section S3 of the supporting text ( Protocol S1 ) for the effects of varying each parameter . Table S3 lists the corresponding names in the code for simulation parameters mentioned in the main text ) . This simple model exhibits both symmetry breaking and motility behavior that reproduces the sequence of events seen in vitro ( Figure 1E–1H , Video S2 ) . Our experimental observations and our simulations share several features . As the shell grows , it becomes denser near the surface of the bead . When the thickness of the shell reaches approximately the radius of the bead , a clear crack develops , and the bead exits the shell , then the shell opens , crescent-like , and motility proceeds , leaving a low-density and somewhat irregular comet-like tail behind the bead . Figure 1I–1L show the underlying 3-D nature of the simulated network , with the network links colored by tensile stress ( Videos S3 and S4 ) . Although the simulations share many of the features of the experiments , we noticed that the shell shows a close to perfect arc for the experimental conditions in Figure 1 , but the simulations robustly show a more V-like shape with a dent in the center of the inner high-density region of the shell ( compare Figure 1C and 1D with 1G and 1H ) . This implies either a failure of the simulation to capture an essential behavior of the network , or a condition of the in vitro system that we did not include in the simulations . To determine the cause of the dent , we examined the 3-D mechanics of symmetry breaking in our simulations . Figure 2A and 2B show 3-D top and side views of a representative simulated shell after the bead has moved away from the shell , demonstrating that even though the bead is unconstrained in three dimensions , the symmetry break and shell opening occur along only one axis . A rip in the outer shell often accompanies the dent , as seen in Figure 2A ( arrow ) and the corresponding 2-D projection view shown in Figure 2C . To understand why symmetry breaking occurs within one plane , we looked at how the shell cracks . Figure 2-D shows an earlier 3-D view of the same simulation , just as the crack completely fractures the shell; isosurfaces show the densest region of the network in green to highlight the shape of the shell , and the extent of the lower-density actin network ( semitransparent ) . The symmetry-breaking crack is a straight line , as opposed to either lightning-like fracture ( s ) along the weakest regions of the network , or a circular hole opening to allow the bead to escape . The consequence of this straight-line break is that the 3-D stresses in the network are relieved in a 2-D manner—essentially splitting the 3-D spherical shell into two hemispheres that open apart from one another like a clamshell , causing large stresses at the hinge . When this 3-D geometry is viewed from above , the hinge appears as a dent , seen in Figure 2A and 2C . The crack that opens the two hemispheres often continues all the way around the outer network , resulting in the rip in the outer shell that accompanies the dent . For only one rip to occur , as soon as a crack begins , circumferential tension must relax quickly around the bead before a second crack begins . We can reduce this relaxation around the bead by increasing the strength of attachments with the nucleator ( Figure S16 ) , which prevents the network moving relative to the bead and makes the second crack progressively more prominent . For the experimental conditions in Figure 1 , we had intentionally confined the bead closely between a slide and coverslip to prevent it moving out of focus while we took data . Having seen how the crack propagates around the bead in the simulations , we hypothesized that the lack of a dent seen in the experiments might be a result of this constraint on the network preventing the crack propagating to the rear of the bead . To test this , we ran the same simulation while constraining the network between two planes ( we also excluded nucleation from the very top and bottom 10% of the bead to prevent artifacts caused by this material having nowhere to go ) . Figure 2E and 2F correspond to 2C and 2D , but for this constrained shell ( interactive 3-D representations are included in Figure S4 ) . The constraint creates a toroidal shell that also breaks in a straight-line crack , but unlike the breaking of the spherical shell , the broken toroidal shell relaxes into a much more perfect arc , with the dent much reduced and the shell more closely resembling those seen in the experiments . If our simulations are a valid model for the behavior of the actin network , they predict that if we were to perform the symmetry-breaking experiment in an unconstrained 3-D volume in vitro , it would produce a clamshell break with a dent in the shell opposite the break site as we see in the simulations . To test this , we performed the in vitro experiment using 5-µm diameter ActA-coated beads while controlling the headspace of the reaction with glass spacer beads of either 5 . 1-µm diameter for the constrained condition or 15 . 5 µm for the unconstrained condition . Because the 3-D shell structure is hard to interpret from a single 2-D microscope image , we reconstructed the 3-D shells from confocal z-stacks . We fixed the reaction after symmetry breaking ( see Materials and Methods ) to prevent movement while the z-stack was acquired; so for experiments , we are only able to capture the 3-D geometry at one time point after symmetry breaking has occurred , in contrast to having every time point in the simulations . Figure 2G and 2H show an example of a 2-D projection and 3-D reconstruction of a confocal stack of an unconstrained bead , confirming the distinctive bilobed structure , and V-shaped shell with central dent . Figure 2I and 2J similarly show the constrained condition with the near-perfect arc . ( Beads tend to settle by gravity so that the tail and wide axis of shell are parallel to the coverslip , with shell cracks in the z-direction . ) Figures S5 and S6 contain further examples of 2-D projections and 3-D reconstructions of symmetry breaking . Shell geometry for constrained beads was extremely consistent , always showing the near-perfect arc . Unconstrained beads showed less regularity , but always showed shells with shapes consistent with linear cracks; on one occasion , we observed a shell with a three-way opening ( Figure S6B ) . To confirm that the mechanics of symmetry breaking in our simulations reflect those seen in vitro , we tracked the deformations of the shell during in vitro symmetry breaking using fluorescent speckle microscopy ( Figure 3A , Video S5 ) . Low doping of fluorescent actin produces fiduciary marks that allow us to measure the mechanical deformations of the network [36] . We tracked five parameters: bead displacement , expansion of the crack , circumferential stretching of the inner shell , circumferential stretching of the outer shell , and radial stretching of the shell ( Figure 3B and 3C ) . When symmetry breaks , the crack opens rapidly and then slows as the shell approaches its final shape . As the shell opens , the outer circumference contracts with kinetics that mirror the crack opening , but the inner shell remains approximately the same circumference , merely reducing its curvature . As the shell opens , it also becomes thicker , with the kinetics of radial expansion mirroring the circumferential contraction and crack opening ( magenta and blue lines in the graphs in Figure 3C ) . We plotted similar parameters for a simulation run . We measured the 3-D distance between pairs of points approximately 2 µm apart ( e . g . , in the circumferential direction; Figure 3-D and Videos S6 and S7 ) . The mechanics of the simulations behave like the in vitro experiments , with the crack opening rapidly , the outer circumference of the shell contracting and the shell becoming radially thicker , all with similar kinetics . The values of the Poisson's ratios differ a little , approximately 0 . 2 for the in vitro shell and approximately 0 . 3 for the simulation , likely resulting from simplifications in the functional forms for the link and repulsive forces ( previous theoretical models have assumed a wide range of Poisson ratios , from 0 to 0 . 5 [31] , [33] , [37] ) . Also , the behavior of the inner shell differs slightly between experiment and simulation , with the circumference transiently expanding slightly ( frame 140 ) before returning to its original length , whereas in vitro , the length remains constant . This most likely reflects transient disequilibrium during the most rapid part of the symmetry breaking , which is equilibrated more quickly in vitro than in the simulations . The current model therefore reproduces the qualitative behavior of the experiments but requires calibration in future work before it would be able to match quantitative measures . ( N . B . For convenience , we note that 1 s corresponds to approximately 1 . 4 frames , but stress that this is not extensively kinetically calibrated . ) Our simulations provide detailed information about the mechanism of symmetry breaking , e . g . , the network motion , distribution of forces and ripping of the network ( Figure 4A–4D , Video S8 ) . In the left panels ( Figure 4A ( i ) –4D ( i ) ) , we colored the regions of the network with red stripes to show the trajectory of the network as it moves away from the bead surface . Initially ( frames 1–60 ) , this pattern is radially symmetric—broken links occur randomly around the surface , giving no indication of the future site of symmetry breaking ( link breaks are stochastic , see Video S8 ( ii ) and Video S11 ) . By Frame 62 ( Figure 4A ) , the nodes around the future crack site have begun to diverge ( Figure 4A ( i ) ) , followed by a burst of localized link breaks at the site ( Figure 4B ( ii ) ) . This weakens the network , causing stress in that region to be distributed over fewer remaining links , leading to more breaks by positive feedback ( Figure 4C ( ii ) ) , followed by the bead moving off with links breaking primarily at the front ( Figure 4D ( ii ) , Video S12 ) . To determine the force balance that contributes to shell formation and symmetry breaking , we examined the spatial distribution of stresses within the network . The right-hand graphs ( Figure 4A ( iv ) –4D ( iv ) , Video S8 ( iv ) ) show how the radial and circumferential tensions vary with distance from the surface of the bead ( negative tension corresponds to compression ) , and the center panels ( Figure 4A ( iii ) –4D ( iii ) ) show the spatial distribution of circumferential tension . These are calculated as sums of the link tension forces ( positive ) and the node–node repulsion forces ( negative ) , split into radial and circumferential components ( individual components are graphed in Video S9; we exclude the data point nearest the bead because of surface artifacts caused by the way we deal with nodes that enter the nucleator , see Video S10 for full data ) . Both radial and circumferential tensions are negative at the bead surface , i . e . , the center of the shell is under compression , the inner compressive forces balancing the outer circumferential tension . For small network distortions ( close to the surface ) , the network equilibrates this compressive force primarily through the isotropic node–node repulsions , so the compression is not restricted to the radial component . Close to the bead surface , circumferential tension is lower ( as predicted by the Elastic Gel model ) , so the compressive force is greater than the tension force ( and the overall tensile force is negative ) . Circumferential tension increases rapidly with distance from the bead ( Figure 4C ( iv ) ) , becoming positive at approximately 1 . 0 µm , with the maximum tension approximately 1 . 5 µm from the surface , and tailing off at higher distances as the network becomes sparse . This distribution of forces can be clearly seen when the symmetry break begins ( Figure 4C ( iii ) ) as a red band of maximal circumferential network tension at approximately 1 . 5 µm encloses a blue band of maximal network compression at the bead surface . The distribution remains relatively static over time as forces build up ( Figure 4A ( iv ) –4C ( iv ) ) , although the magnitudes of the forces change , with the maxima occurring when symmetry breaking begins ( Figure 4C ( iv ) ) . These data support the Elastic Gel model for symmetry breaking: as the network is pushed out by nucleation at the center , it expands in the circumferential direction like a balloon , creating circumferential tension . Network compression close to the surface provides the balancing force for this circumferential tension—and because the expanding layers of network pull the network apart circumferentially , but not radially , the resulting radial forces are always compressive ( negative tension in the graphs in Figure 4A ( iv ) –4D ( iv ) ) . The release of tensile energy upon symmetry breaking can be vividly seen between Figure 4C ( iii ) and 4D ( iii ) —the shell opens and pulls back away from the bead , contracting circumferentially and releasing the energy stored in circumferential tension—much of the red region of maximum circumferential tension in Figure 4C ( iii ) turns blue ( compression ) in Figure 4D ( iii ) , Video S8 . Small defects in the outer shell have been proposed to establish the site of symmetry breaking [32] , [33] . We can determine when the symmetry breaking site is established in our simulations relatively easily . In our simulations , we add new network stochastically at the bead surface—this randomness results in a unique network and symmetry-breaking direction for each run . For each run , we save a complete description of the system at each time point , and can resume the run at any point with a different random seed . To discover the time at which the symmetry-breaking direction is determined , we ran a simulation through to symmetry breaking , then rewound and restarted the same simulation from nine different time points , but with a different random seed . We repeated this set of nine runs five times to calculate the mean and standard deviation of the angle between the new symmetry-breaking direction and the original direction ( Figure 4E ) . This produces a high variance in symmetry-breaking direction before the direction is determined , and both very low variance and a close to zero deviance angle afterwards . We find the symmetry-breaking direction is essentially random until frame 80 , at which point the direction becomes the same as the original run . Symmetry-breaking direction is therefore determined between frames 70 and 80 , i . e . , very late—just before symmetry breaks—rather than being determined early by defects in the initial outer network . Our simulations also show that the force balance and pattern of link breaks in the outer network before symmetry breaking define the final curvature of the shell after symmetry has broken . Figure 4F shows that halving the spring constant ( the FL parameter ) causes the shell to double in thickness , and Figure 4G shows that increasing the threshold force for link breakage ( the FBL parameter in the simulation ) causes the shell to become flat ( see also Figures S13 and S12 ) . These results follow from the Elastic Gel model: decreasing the spring constant between links of the network will require that more material be deposited to build up enough circumferential tension for symmetry to break , so the shell is thicker . Also , the final curvature of the shell after recoil is dependent on the number of links that have broken in the outer shell during the earlier stages of shell buildup . Without breaks in the outer shell , the final equilibrium area of the outer shell is still the same as the inner , so the resulting shell is flat . The more links that break in the outer network , the larger its equilibrium area , and the higher the resulting curvature . These parameters and others are more thoroughly explored in Model Robustness , Section S3 of the supporting text ( Protocol S1 ) . Symmetry breaking is a particularly robust behavior of our model . Of the parameters tested , those that do not break symmetry are those that set network link density to extremes ( Figures S10 , S11 , S12 , S13 , S14 , S15 , S16 , S17 , S18 , S19 , and S20 ) . One extreme creates a very strong network that builds a dense shell that never breaks symmetry , by creating conditions in which the network strength increases faster than the network strain , e . g . , when we increase the threshold for link breakage ( Figure S12 ) . The other extreme creates a very weak network in which symmetry does not break because chains of links are too short to communicate tension around the bead , so the network remains unpolarized , seen by decreasing the crosslinking probability , or decreasing the link-breaking threshold ( Figures S11 and S13 ) . Our model network is constructed from nodes and links that are short compared to the size of the bead—to transmit force around the bead , there must be enough links to form chains spanning around the bead . The “mesh size” characterizes the length scale of the network formed from these chains of links , referring to the minimum size of a particle that would be trapped by a network made of these chains . In our case , if the mesh size is greater than the size of the bead , the bead would be able to move through the network , so it would not be possible to build up tension in the shell , and there would not be a clean symmetry break . For our purposes , we define network coherency as the bead size divided by the mesh size , i . e . , high network coherency means that the bead will see the network as an elastic solid , whereas low coherency means the bead would be able to squeeze through the network . We find that even a low level of network coherency is sufficient to support symmetry breaking , the key is that tension is transmitted around the bead . This kind of symmetry breaking does not involve a distinct shell that cracks , but rather a gradual oozing of the bead from a network cloud ( Figures S11 and S13 ) . This oozing demonstrates a qualitative change in behavior that results from the quantitative change in degree of crosslinking . When a sparsely linked network deforms , it undergoes plastic flow as energy is lost by links breaking independently , whereas when a dense network deforms , it builds up elastic energy , as each link stretches slightly while remaining below its breaking strain . Eventually , this dense network undergoes brittle fracture when many links break at once . The initial shell shows a gradient of network density increasing from the outer to the inner surface of the shell both in vitro and in silico . This density gradient emerges spontaneously from the APS model as a result of the increasing circumferential tension in the outer shell compressing the inner shell . The initial outer network is sparse because it is not under compression , so the network has a low density of links ( since links are formed to nearby nodes , and a sparse network means fewer nodes nearby ) . This sparse initial outer network is weak and plastic but does provide enough compression on the inner network to cause an increase in density , hence a greater number of links , and a stronger network , which builds by positive feedback . As demonstrated in Figure 4A–4D , which shows a peak in circumferential tension towards the center at around 1 . 5 µm from the surface , it is this inner brittle network that stores the bulk of the elastic energy , and undergoes brittle fracture during symmetry breaking . In both our experiments and simulations , the bead continues to move after breaking symmetry . To investigate the motility mechanism , we examined network movement by plotting orthogonal views of the network trajectory for a simulation of smooth motion ( Figure 5A ) . To show the network trajectory , we marked the network with a spatiotemporal grid , coloring it red when it originated at evenly spaced locations around the bead ( the parallel lines in the tail ) , and at even time intervals during the run ( the orthogonal shell-like curves ) . During the smooth motion phase , we see a pattern of parallel lines behind the bead , demonstrating that the network does not contract orthogonally as it moves away from the bead surface , which agrees with previous experimental work showing no orthogonal network contraction for motile beads [17] , [38] . So in our simulations , orthogonal contraction of the network does not provide the driving force for motility by squeezing the bead forwards . In Figure 5A , the time-pulse markings highlight regions of network that come from the bead surface within short time windows—in effect demonstrating what happens to the equivalent of “shells” for smooth motion . In the tail , they appear as red lines with curvature much lower than the bead curvature , i . e . , even during smooth motion , the high-curvature network produced at the bead is opening up just like the shell during symmetry breaking . The shape of these smooth-motion shells also match well those produced by physically switching the color of the actin during in vitro experiments [17] , [38] . Even though the bead in this simulations is not constrained , during smooth motion , the network sweeps around the bead primarily in one plane—Figure 5A shows that the tail is much wider in one axis than the other , similar to the shell during symmetry breaking in Figure 2A and 2B . In three dimensions ( Figure 5B and Figure S7 ) , tracking the network trajectory shows ripping in one axis along a sustained straight-line crack at the front of the bead . We confirmed that the trajectories of the network in our simulations match those seen in vitro using fluorescent speckle microscopy ( Figure 5C , Video S13 ) . The composite image is produced by coloring and overlaying successive frames from a video of a motile bead in vitro , registered to the motile bead ( i . e . , lines represent movement relative to the bead ) . The trajectories in vitro mirror those seen in silico , with network expanding away from the bead as it is swept around and incorporated into the tail , and no convergence of trajectories behind the bead . The effect of this sweeping motion on the circumferential tension in the simulated network can be seen in Figure 5D . The network shows a peripheral zone of circumferential tension ( red ) at the outer network surface , and a region of network compression ( blue ) just behind the bead . This tension zone is far from the bead surface except at the thinnest part of the network at the front of the bead . The opening of the “smooth-motion shells” in Figure 5A is reminiscent of how the shell opens during symmetry breaking , and suggests that the network might contract circumferentially and expand radially , as we saw during symmetry breaking in Figure 3 . To test this , we made similar measurements of the network stretching during smooth motion , and because the network is asymmetric during smooth motion , we restricted measurements to the rear of the bead; Figure 5E and 5F show lines used to take circumferential and radial length measurements during the smooth motility phase ( shown in Videos S14 and S15 ) . Figure 5G shows how the network behind the bead stretches as the bead moves , confirming that it stretches circumferentially to approximately 120% before relaxing back to approximately 107% of its original length . As it does so , it expands radially to approximately 112%—similar to the radial expansion of the outer shell during symmetry breaking . This relaxation is complete after approximately 150 frames ( ∼18 µm ) , consistent with previous in vitro photobleaching data showing the network is still undergoing relaxation at approximately one bead diameter and is complete by approximately four bead diameters [17] . Why do the trajectory lines of the network look parallel ( and even diverge slightly ) as they move away from the bead ? Although the network contracts circumferentially , it also rotates around the bead , i . e . , the network on the outer edges of the tail sweeps backwards relative to the inner tail . This rotation allows the points in this smooth-motion equivalent of a shell to contract relative to one another while following the parallel trajectories shown in Figure 5B; i . e . , there is circumferential , but not orthogonal , network contraction . The Soap-Squeezing model proposes that orthogonal elastic contraction of the network drives motility . The lack of orthogonal network contraction rules this out , but could circumferential elastic network contraction play a similar role ? To determine whether circumferential elastic contraction is required for motility , we performed in silico experiments to find out what happens when elastic contraction is reduced or eliminated . Changes in these parameters affect both the bead velocity profile and the stretching of the shell . Figure 5H shows the velocity profile of the bead described above , before reducing elastic contraction . The bead is initially at rest , with a distinct spike in velocity upon the original symmetry-breaking event . ( Note: the smooth motility regime still has small velocity fluctuations , especially just after symmetry breaking . To clearly distinguish between the two regimes , we define smooth motion as having velocity that varies <25% of the mean velocity , and pulsatile motion as having velocity that varies >100% of the mean velocity . ) We first reduced the elastic contraction by tuning network parameters to produce a less elastic network . We based these parameters ( RM = 5 . 0 , FBL = 2 . 0 , FL = 4 . 0 ) on the Model Robustness results , Section S3 in the supporting text ( Protocol S1 ) . Figure 5I shows this less elastic network expands more and contracts less: the network stretches circumferentially to 133% of its original length before relaxing back to only 128% , with a slight radial expansion , to 105% . The velocity profile under these conditions ( Figure 5J ) shows smooth motility , but strikingly lacks the initial spike in velocity compared to Figure 5H , and the onset of motility is delayed . For the elastic network , the initial velocity spike corresponds to the symmetry-breaking event , and Figure 5M shows that for the less elastic network , rather than producing a single shell with its buildup of elastic energy and sudden release and contraction that ejects the bead , the network fractures in multiple places , producing three separate tails . Eventually , the bead squeezes out orthogonal to these tails ( Figure 5N , Videos S16 and S17 ) , with smooth motion and network trajectories that resemble the bead in Figure 5A . In spite of being less elastic , this network still contracts circumferentially , and observation of network motion suggests this contraction is likely driven by network fractures that opened during expansion being closed by the compression forces of material swept around the bead . To abrogate this contraction , we performed the same experiment but allowed network movement only for nodes within a limited range of the bead , permitting the network to expand , but locking it in place before it could contract . This results in similar smooth motility ( and a similar pattern of network tracks ) under these conditions , showing that network recoil is not required for smooth motion ( Figure 5L and 5K , and Video S18 ) . What explains smooth directional motility ? We propose a “Sustained Rip” model: an extension of the symmetry-breaking mechanism combined with a pressure-induced transition from brittle to plastic network behavior . For smooth motility , as during symmetry breaking , network produced at the bead surface tends to be pushed outward , creating circumferential tension ( Figure 5D ) . During motility , however , the existing shell ( or tail ) reinforces the network at the rear , forcing circumferential tension to be relieved by stretching and ripping at the front ( Figure 5B ) . The radial compression that balances the circumferential tension presses on the bead from all sides except where there is little network—at the front ( Figure 5D ) . The imbalance of these compressive forces causes the bead to move forwards , driving it through the rip site . Ripping also means that radial compression does not build up enough to compress the network and cause it to become dense and brittle—it remains sparse and plastic . Direction is maintained because contact with the tail ( or the original shell ) always reinforces the network at the back , leaving tension from the expanding network to be relieved by ripping in the unreinforced zone at the front . The network trajectories in Figure 4D and circumferential tension plot in Figure 5D support this , showing that contact with the original shell restricts the new network from free expansion at the rear—the new network does not expand symmetrically as the original shell did in Figure 4A , but diverges less in the rear region in contact with the shell , and more at the front . This Sustained Rip model predicts that specific changes in network properties will affect the continuity of motion . For example , after symmetry breaking , motility should be smooth only if the newly forming network is sparse and plastic when uncompressed . If the newly forming network has a high enough link density that it behaves like the brittle inner network of the original shell , we should see pulsatile motion—essentially repeated symmetry breaking as new brittle shells form one after another . Changing the probability of forming network links ( PXL ) is a simple way to test this prediction by altering the network link density . ( Note that this is an alternative to , but does not exclude , friction as a contributor to pulsatile motion [20] . ) We ran simulations to see how varying the probability of forming links affects the smoothness of motility . Figure 6A shows the network architecture at regular time intervals , and Figure 6B shows the corresponding bead velocity profiles , for a range of link probability ( PXL ) values . At very low link probabilities ( PXL = 0 . 125 ) , there are so few links that each part of the network behaves independently rather than forming a single coherent network—and a symmetric cloud of material surrounds a stationary bead . Increasing PXL to 0 . 375 , symmetry breaks and the bead moves off . Under these conditions , the shell is barely coherent—it remains together but does not recoil when symmetry breaks; instead a diffuse cloud of material forms , and the bead gradually oozes from it . There are fluctuations in the velocity , but they remain small ( <25% deviation from the mean velocity ) . As we increase PXL to 0 . 625 , a distinct shell forms , the bead undergoes one pulse after the initial symmetry break , and then the motion becomes smooth ( <25% deviation from average velocity ) . As PXL increases further to 0 . 875 , the shell becomes denser , and the motion becomes very strongly pulsatile ( >250% deviation from the mean velocity ) and periodic , as strong shells repeatedly undergo largely independent symmetry-breaking events . Bead velocity rises abruptly when the shell breaks , and tails off slowly as the shell relaxes , leading to an asymmetric velocity profile that closely matches experimental measurements of bead velocity during pulsatile motion [20] . This transition from smooth to pulsatile motion supports the Sustained Rip model for motility: as network coherency increases , the stronger shells formed are more immune to the influence of the previous shell , causing them to undergo essentially independent symmetry breaking . The small influence of the previous tail explains the relatively constant direction of motion . Further supporting the Sustained Rip model , two other parameters of the APS model also control smoothness of motility by affecting the ability of the old network to alter the brittleness of the newly forming network: 1 ) Increasing the node repulsive force makes the network less compressible , reducing the pressure-dependent density increase , and leading to smooth motion ( Figure S15 ) ; and 2 ) lowering the link spring constant FL results in circumferential tension ( and radial compression ) building up more slowly ( i . e . , the network has to get bigger before the dense , brittle shell forms ) causing a much thicker shell when symmetry breaks , thick enough to be beyond the effect of the initial tail , and immune from the sustained rip effect's ability to induce smooth motion ( Figure S14 ) . Friction may also contribute to pulsatile motion: in vitro , increasing surface ActA concentration ( intended to increase the ActA-filament attachment component of friction ) causes a transition from smooth to pulsatile motion [20] . We see a similar effect in our simulations: when we increase friction by increasing the strain limit before node–bead links break , we also see a transition from smooth to pulsatile motion ( Figure S17; note the transition is less clear-cut than those described above ) . However , in the APS model , we can show that friction is unnecessary for pulsatile motion . We can set friction to zero by eliminating node-bead links , but still induce the transition from smooth to pulsatile motion by increasing network coherency , e . g . , by increasing PXL ( Figure S20 ) . We interpret this to mean that the change from smooth to pulsatile motion is directly caused by a change from a plastic to brittle network , and that a dense , brittle network can be caused by increasing its density in two ways , either 1 ) by increasing the coherency of the outer shell , which puts pressure on the inner shell , or 2 ) by increasing the network–bead attachment , which increases the density of the inner shell by holding it close to the bead surface . Our data show how an evenly coated spherical bead can be driven on an actin comet tail , but the original observations of this form of motility were on the intracellular motility of the bacterium , Listeria monocytogenes , which is a different shape ( capsule-shaped rather than spherical ) and has an asymmetric distribution of the actin nucleation factor , rather than symmetric . How important is this asymmetric distribution to the lengthwise motility of Listeria ? To determine the importance of shape and of nucleator distribution on motility , we tested the effect of varying them in silico . When we simulate a capsule-shaped nucleator with nucleation restricted to one half of the capsule , motility is lengthwise and symmetry breaking is unnecessary ( Figure 7A–7D ) . Network tracks with regular spacing and frequency ( Figure 7C ) and 3-D tracks ( Figure 7D , Figure S8 , and Video S19 ) show that the network expands outward from the nucleator , opening up as it moves away from the surface . Similar to the motility of spherical beads , there is no evidence for orthogonal contraction of the network . When we distribute nucleation uniformly over the capsule surface , however , the direction of motion changes: for both symmetry breaking and motility , the capsule moves sideway , as shown in top and side views in Figure 7E–7H and Video S20 . The Elastic Gel model predicts that the higher the surface curvature , the faster the buildup of strain within the network [19] . We therefore anticipated the higher curvature regions at the ends would build up strain faster and that symmetry breaking would occur there ( the ends are higher curvature because although the radii are equal , the curvature is 2-D at the ends but only 1D on the linear section ) . To understand why symmetry breaks sideways , we examined the network motion by plotting network tracks just prior to symmetry breaking ( Figure 7I ) . This shows that as tension builds up , the network on the linear section is drawn towards the ends of the capsule , so relieving the strain and the network tension in this direction remains low ( Figure 7J ) . Around the capsule's cylindrical axis , however , there is no linear section to expand and relieve the strain buildup , so the tension in this direction builds up rapidly ( Figure 7K ) . Symmetry breaking therefore occurs in this direction ( causing sideways motion ) by a similar mechanism to the spherical beads , and the sideways symmetry breaking and motion of this geometry can be explained by the sustained rip mechanism described above , in which the axis of the rip is defined by the long axis of the capsule . We checked our prediction of sideways symmetry breaking and motility by stretching spherical beads to make ellipsoids and comparing their in vitro motion with simulations . Figure 7L ( Video S21 ) shows that simulations of ellipsoids produce the same sideways symmetry breaking seen for the capsules ( subsequent motion is also sideways like the capsules , Video S22 ) . We performed bead motility experiments as above with a 15 . 5-µm headspace ( i . e . , unconstrained ) , and captured 3-D z-stacks of the beads soon after symmetry breaking . Figure 7M and 7N show a 2-D projection and 3-D reconstruction of such an ellipsoidal bead experiment after sideways symmetry breaking , with two density isosurfaces: the green chosen to show the shell , and the semitransparent grey chosen to outline the void space of the ellipsoidal bead to confirm the bead position and orientation . ( Note that it is not possible to determine the direction of motion relative to the bead axis from the 2-D projection in Figure 7M alone . ) More examples of sideways symmetry breaking of ellipsoidal beads are shown in Figure S9 . For ellipsoid aspect ratios >1 . 75∶1 , we almost always see sideways symmetry breaking ( 98% , n = 58 ) and sideways motion ( 95% , n = 55 ) , though we occasionally see beads changing direction or curved bead paths during the subsequent motion .
Our simulations build on the Elastic Gel model of symmetry breaking [19] , [31] , using an Accumulative Particle-Spring ( APS ) model to capture the mesoscopic viscoelastic properties of actin networks . The APS model represents these properties using a series of nodes and springs that allow us adjust a simple set of viscoelastic network parameters that correspond to mechanical properties of the in vitro network . For example , the repulsive force between nodes ( FR ) roughly corresponds to the resistance of the network to compression , and the spring constant ( FL ) roughly corresponds to the resistance to tension . The APS model also captures some network behavior as emergent properties . For example , as the network stretches circumferentially , links reorient circumferentially to result in strain hardening , and compression of the inner network by the outer network increases the node and spring density , resulting in the more brittle behavior necessary to produce the symmetry breaking and transition from smooth to pulsatile motion seen in silico and in vitro . The APS model builds the network from spring-node units that correspond to a particular mesoscopic mechanical behavior of crosslinked actin networks . We know a good deal about the viscoelastic behavior of in vitro actin networks from studies that examine the randomly crosslinked networks produced by mixing crosslinking proteins with stabilized actin filaments . For these networks , crosslinking proteins connect adjacent filaments with one another to form chains with a characteristic mesh size that can resist tension across the sample . The chains of nodes and springs in silico approximate the behavior of these chains of filaments , crosslinks , and friction , to transmit tension around the in silico bead . For Arp2/3-built networks to transmit tension around the bead implies significant friction and entanglement . Activated at the bead surface by ActA , Arp2/3 binds to existing filaments and nucleates new filaments from their sides to form a dendritic branched structure [3] , [5] . Because only new filaments are crosslinked , each dendritic tree cannot crosslink to any other , so there can be no encircling chains of filaments and crosslinks around the bead that could carry tension . Circumferential tension would simply be dispersed by separation of these independent dendritic networks were it not for friction and entanglement . The node-spring links in our APS model , therefore , also implicitly represent these friction and entanglement links between dendritic trees , and just as friction and entanglement would be expected to increase with network density and pressure , so the density of node-spring links in the APS model increase with density and pressure . We create links only at the surface when nodes form , to mimic in vitro filament entanglement , which can only occur when filaments polymerize and insert through gaps in the existing network , and this occurs only at the bead surface . We keep the polymerization rate constant in our simulations in spite of changes in protein concentrations and pressures at the bead surface during shell growth , because previous data show the in vitro rate of deposition of actin to remain essentially constant over this period of the reaction ( Figure S6 from [16] ) . In an expanding shell , the actin network continuously stretches as it is displaced outward by assembly of new actin at the surface . The opening of the shell during symmetry breaking is well explained by the basic assumption of the Elastic Gel model that all network layers tend to relax to their equilibrium area , the area of the surface of the bead where they were created . Since this area is the same for all layers , and since connected layers with equal areas and a non-zero thickness would tend to flatten to a plane , the shell tends to flatten towards a plane once symmetry breaks . For most conditions , we do not see a perfectly flat plane , but we do see the shell relax to a flat plane when we increase the link strength . This is because high link strength reduces the number of links that break in the initial outer shell as it is stretched—high link strength means that links only break during the actual symmetry-breaking event . This explains the curvature of the arc of the symmetry-breaking shell: Before symmetry breaking , as the outer shell is stretched , links break irreversibly , expanding the equilibrium area of the outer shell , so the final shell shape is no longer the relaxation of planes of equal equilibrium areas . The larger equilibrium area of the outer network results in a convex shell . The APS model also shows how the rip that occurs during symmetry breaking brings about the clam-like 3-D geometry of the shell . Since the starting geometry is a sphere , as the shell opens and flattens , large tensile strains occur around the circumference ( Figure 8A ) . Rips relieve these circumferential strains; a single rip will produce a bilobed structure , but multiple cracks are possible ( and observed in silico and in vitro ) as the network strength is increased . We also often see a crack in the outer network opposite the main symmetry-breaking crack . When the bead is unconstrained , this tends to line up with the dent in both the simulation ( Figure 2A and 2C ) and experiment ( Figure 2G ) , but can also be present in constrained beads without the dent ( Figure 2E ) , showing that the dent is not the cause of the rip . In line with a previous experimental observation [35] , our simulations also show linear cracks ( instead of a round-hole opening to release the bead ) . These are linear rather than circular because positive feedback concentrates the strain to regions of high curvature [39] . The resulting cracked-shell geometry is reminiscent of the Mollweide projection of the globe , in which linear cuts in the map allow a 3-D sphere to be flattened to a plane and reduce stretching distortions at the poles . Paradoxically , pulsatile motion is relatively simple—it is essentially repeated symmetry breaking—whereas smooth motion is more complex , involving a transition to a different regime . The very same conditions build an initial rigid brittle shell that cleanly and distinctly breaks symmetry and then builds a more plastic tail on which the bead moves smoothly . How does the presence of the old shell cause adjacent new network to behave in the plastic manner that produces smooth motion ? Our simulations suggest that this switch to plastic behavior rests on the pressure dependence of network plasticity . By reinforcing one side of the newly forming network , the old shell focuses the circumferential tensile strain on a small region of newly forming , uncompressed , and therefore , plastic network on the other side , which rips . Just like inflating a balloon with duct tape on one side—the duct tape not only prevents that side expanding , but it means the other side is stretched twice as much to accommodate and ruptures sooner . In the bead case , this leads to a rip before pressure has built up—so the network remains sparse and plastic , which in turn leads to continued ripping and steady-state smooth motion . If this pressure dependence is disrupted or reduced , the transition to smooth motion is delayed or abolished . In our simulations , increasing PXL increases the number of links and the coherency of the shell , leading to essentially independent shells and pulsatile motion . We expect this mechanism to correspond to the physical mechanisms that produce the switch to smooth motion seen in real actin networks , in this case through pressure-dependent increases in entanglement , friction , and filament orientation effects ( likely to be significantly affected by pressure , as load-directed filaments stall ) . Oblique filaments would tend to entangle and reinforce the network while contributing little to the movement of the bead away from the network , and so this may tip the system into a positive feedback of network stiffening that is relieved by symmetry breaking . We predict a significant alignment of filaments orthogonal to the direction of motion for a pulsatile bead , but less orthogonal alignment for a smoothly motile bead . We can also consider these network behaviors in terms of changes in network mesh size . This refers to the distance between the chains of links that transmit tension through the network , i . e . , the mesh size decreases as crosslink density increases but is always greater than the individual link lengths . When the symmetry-breaking shell forms , the pressure produces a tightly crosslinked network with a small mesh size ( on the order of the link length ) . Because the mesh size is very much smaller than the bead size and the shell , the network behaves as an elastic solid . Decreasing crosslink density increases the mesh size and results in a mesh size that is larger than the bead , but smaller than the shell . This means that the shell can still resist tension , but beads can essentially move through the network , resulting in the oozing symmetry breaking seen in Figures 6 and S11 . Decreasing crosslink density still further produces a mesh size greater than the bead and the shell , so tension is not communicated around the bead , and symmetry does not break . The switch from brittle to plastic behavior can also be seen in terms of mesh size . Although the pressure buildup in the initial shell produces a dense network with small mesh size and elastic-solid behavior , once symmetry breaks and the rip at the front prevents pressure buildup , the sparse network at the front of the bead essentially has a large mesh size that allows the bead to move through unhindered . The repeated shell-breaking mechanism we propose for pulsatile motion does not exclude other proposed models; e . g . , Listeria and motile vesicles have asymmetric nucleator localization during motility [16] , [20] , [38] , [40]–[42] , so are unlikely to build up symmetric shells . This suggests a friction mechanism for pulsatile motion , though pressure buildup still may contribute to periodic variations in friction . In our simulations , we show that a frictionless bead still produces pulsatile motion , suggesting that although friction may contribute to pulsatile motion , it may not be required . In addition to the pulsatile motion whose steps are of the order of the bead size , Listeria can also make steps of approximately 5 . 4 nm [29] , [43] . These “nano-saltations” are very likely to be directly caused by friction because their scale is of the order of actin monomers , much smaller than the characteristic scale of the elastic gel properties of the network . Our prediction that the shell outer network is more flexible and plastic and the inner network more rigid and brittle has implications for the mechanism of symmetry breaking . The driving force behind symmetry breaking is the circumferential stretching of the network as it moves outward , and we initially expected to see a brittle crack in one region of the outer network that would seed the symmetry break as has been previously proposed [32] , [33] . Instead , we find that the symmetry-breaking direction is determined late because the tensile stress is primarily carried , not by the very outer network , but by a dense rigid network relatively close to the bead surface . We stress that this does not mean that the network does not rip at the outside first—it does because this is the most stretched region—but the outer network rips in many places without triggering symmetry breaking; it is the rip of the inner network that determines the symmetry breaking site , and this is not determined by the outer network . If stochastic variations in the density of the initial ( outer ) layers of the network were to determine the symmetry-breaking direction , we would expect the direction to be determined early , when this initial network forms . We show that symmetry-breaking direction is determined late in the simulations , just before the rip occurs , implying that there is no existing vulnerability in the outer network that later seeds the crack , but rather that network density and linking are finely balanced up to the critical point when load becomes too great , and failure occurs stochastically . This fits well with the mechanism proposed above for curved versus flat shells: the balanced stochastic breaking of links in the outer network , not only equilibrates the strain , but results in the even-expansion equilibrium area of the outer shell . When symmetry breaks , shell curvature is determined by the balance of the equilibrium areas of the inner and outer shells—when the outer layer equilibrium area expands , we see curved shells , and when the link strength is increased , the even breaking is eliminated , the outer layer equilibrium area does not expand , and we see flat shells . Our conclusions about site selection are based on our simulations—so do they also hold for the in vitro system ? This depends on where tension is carried , which depends on the network rigidity—if the inner network is more rigid than the outer network in vitro , then our conclusions should hold; if the outer network is more rigid than the inner , then they will not . There are several reasons to think the inner network will be more rigid in vitro: First , the inner network is denser in vitro , as shown in Figures 1 and 3 . Second , we often observe numerous small cracks in the outer network ( Figure 1A and 1B ) prior to symmetry breaking that do not predict symmetry-breaking direction , but rather suggest a general stochastic fracture of the outer network similar to the general breakage of links we observe in the simulations . A third reason follows if the Sustained Rip model is valid , since it predicts that under no compression the network will be plastic , not rigid . Since the initial outer shell is formed under no compression , it should be plastic and therefore not carry significant tension . We show that elastic recoil is not required for smooth motility , but is necessary for the classic “shell-retraction” type of symmetry breaking . At first sight , the lack of orthogonal network contraction during bead motility seems to suggest a lack of elastic recoil during smooth motion , but detailed data from our simulations show elastic retraction circumferentially around the bead and , because of its positive Poisson's ratio , radial expansion , an elastic recoil very similar to symmetry breaking . Although elastic recoil is not required for smooth motility , it is necessary for the shell retraction during symmetry breaking . Without it , the network is unable to expand circumferentially and absorb the energy with elastic stretching , but instead quickly rips , resulting in several tails from which the bead eventually emerges . During smooth motility , the network motion appears dominated by plastic flow around the bead . In previous work , Paluch et al . [17] describe a model for smooth motility that explains the network motion by treating the actin network as an incompressible gel that flows around the bead . Although this model relies on force generation by soap squeezing , which is contradicted by their photobleaching data , the general model of network motion by flow of an incompressible gel is consistent with our findings that network compressibility and retraction are not required for smooth motility . Lacking experimental data , previous models have varied widely in their assumptions about network compressibility [31] , [33] , [37] , though recent work suggests it is a particularly important determinant of stress buildup [44] . In our simulations , the plastic flow we see during smooth motility approximates an incompressible gel regime , not because the gel itself is less compressible , but because the compressive forces are lower—the front rip prevents pressure building up enough to significantly compress the gel . The results of our simulations show how the two processes can be reconciled in one system: Symmetry-breaking behavior is dominated by network compression and elastic recoil because the shell is elastic and brittle because it is built under high pressure , whereas smooth motility is dominated by plastic flow because the tail is built under lower pressure because of tension release at the rip . Our conclusions also agree with previous results showing that actin shells from which a solid bead escapes open wide , straighten , and then go on expanding after the bead has moved out [38] . In that paper , Delatour et al . [38] also suggest that evacuation of the gel by elastic recoil is required for movement by evacuating the actin filaments grown in front of the bead to maintain anisotropy in the system . This is based on the observation that during pulsatile motion , the bead periodically slows down and reinitiates the formation of a quasisymmetric actin shell and repeats the initial symmetry-breaking step over and over . The actin shells in this regime are never perfectly symmetrical , but weaker at the front , so the initial direction of the movement ( defined by the gap in the first shell ) is partially conserved . Our results support Delatour et al . 's interpretation that direction is maintained mechanically by reinforcement by the existing tail , but we differ in our interpretation of the role of elastic recoil . We find that elastic recoil is not necessary for movement ( though its absence prevents pulsatile motion ) ; rather , plastic flow evacuates material from the front of the bead . In our model , direction is also maintained by the tail , which reinforces the network at the rear of the bead , but this works by concentrating circumferential tension at the unreinforced zone at the front , leading to a sustained rip . We find that the Elastic Gel model helps explain the sideways symmetry breaking and motility of capsule-shaped and ellipsoidal nucleators . The network stretches around the long axis to relieve the circumferential tension , so only around the short axis does tension buildup cause symmetry breaking ( and motility ) in the sideways direction . Our experiments using ellipsoidal beads confirm this behavior in vitro , and support the elastic gel mechanism as the determinant of symmetry breaking and motility behavior . We show that for lengthwise symmetry breaking and motility , a capsule geometry requires asymmetric nucleation . Wild-type Listeria is capsule-shaped , moves lengthwise , and has such an asymmetric distribution of its ActA nucleation factor [45] , [46] , but a deletion mutation of ActA has been identified that results in a “skidding” sideways motion of Listeria in vivo [47] . Our data raise the possibility that the effect of this mutation could be to alter the asymmetric distribution of ActA activity . Simple models such as ours have limited scope—e . g . , we do not include filament-specific effects such as filament orientations and elongation by monomer addition—so we cannot evaluate the Brownian ratchet mechanism , nor can we investigate the hollow tails seen for beads coated with VASP [27] , or recreate the nano-saltations observed in vitro [43] . The first 3-D computer simulation of actin-based Listeria motility took a detailed approach , simulating the behavior of large numbers of individual actin filaments and branches [29] . The Alberts-Odell model provided an important insight into the connection between the microscale behavior of individual filaments and larger-scale behavior of motile networks , namely how the buildup and breakage of filament-load attachments can produce nano-saltations in motility similar to those observed experimentally [43] . As with our model , the Alberts-Odell model has limited scope . To make their model computationally tractable , Alberts and Odell modeled actin filaments as inflexible rods , fixed rigidly in space soon after nucleation . Thus , the actin network in their model is an inelastic solid and could not be used to study processes involving elastic energy storage , plastic deformation , or mechanical failure: e . g . , the Alberts-Odell model could not be used to study mechanical symmetry breaking or the role of elastic recoil in sustained motility . Concentrating on different aspects of the system , the two models complement one another and explain a wider range of behaviors . Our approach has been to use a simple model with few parameters that confers strong explanatory power at the risk of oversimplifying the physical mechanisms . One potential oversimplification in our model is the constancy of conditions: e . g . , we assume no changes in polymerization rate over time or spatially over the bead surface . The concentrations of components change during the reaction , and although this does not affect the rate of actin polymerization in the shell in vitro [16] , this does not mean it does not affect more subtle physical characteristics of the network architecture . We also know that Arp2/3-based actin nucleation is autocatalytic [48] , which might bias polymerization to the rear of the bead where there is a higher density of existing actin and help maintain directional motion . Our simulations include the code to implement such processes , but we have deliberately not used them in the current study ( Ockham's razor ) . This allows us to show that we can explain the behavior of the system using viscoelastic mechanical effects alone . The goal of this simulation has been to demonstrate the qualitative mechanisms of symmetry breaking and motility , and we have stressed that our simulations do not produce calibrated physical quantities for force , speed , etc . To do so would require both kinetically tuning the model to a more extensive experimental dataset , and also to include a more sophisticated treatment of internal network friction . The current model treats drag very simply: the system is over-damped , with drag proportional to velocity relative to the reference frame , consistent with a low Reynolds number regime . This explains a significant deviation between our model and our experimental data: that the rapid recoil of the shell in symmetry breaking is slower in our simulations . Would kinetic tuning significantly alter the qualitative behavior of the model ? There are two reasons to think not . First , most of the kinetics are close to observed ( e . g . , the ratio of polymerization rates to rates of shell buildup , relaxation , bead movement , etc . are similar ) , so adjustments should not be major , and therefore , would be unlikely to affect the qualitative behavior . Second , even during the rapid recoil of the shell when the kinetics are dissimilar , the equilibrium states match well—i . e . , the close match in the shapes of the curves shown in Figure 3C and 3E suggest that both the in vitro and in silico systems are relaxing from the same initial to same final states , and therefore , are driven by the same processes . We have aimed to include as few parameters as possible , and although we make no claims that these parameters correspond to calibrated physical units of the in vitro network , an important question is how their values are chosen and how critical these choices are to the behavior . Essentially , we arrived at values that qualitatively reproduce the behaviors of the in vitro system by systematically exploring the effects of varying the model parameters , e . g . , in Model Robustness , Section S3 of the supporting text ( Protocol S1 ) . Some behaviors ( e . g . , symmetry breaking , directional motion ) are extremely robust , whereas others , such as smooth motion , are fragile and are disrupted by varying many different parameters . Working with simulations allows us to refine the hypotheses . Full access to the behavior of the in silico system allows us separate out the gross morphological changes measured in vitro , e . g . , the 2-D shape of the final shell , from the underlying components of the motion , e . g . , circumferential squeezing , but no orthogonal squeezing , to refine our ideas about the underlying mechanisms . Furthermore , simulations allow us to directly test whether the proposed mechanisms are required for the motion or are epiphenomena , for example , by producing networks in silico that do not have elastic recoil effects and seeing that motion is essentially unchanged .
The APS model demonstrates how the simple viscoelastic properties of the in silico reconstituted actin gel can give rise to the observed dynamics of symmetry breaking and steady and pulsatile motility of spherical , capsule-shaped , and ellipsoidal objects coated with actin-nucleation factors . The model demonstrates both explanatory and predictive power in these areas , e . g . , explaining how a pressure-dependent change in gel properties allows for a transition between motility regimes and predicting the 3-D geometry of in vitro shells . In the future , we plan to refine the model , calibrating it with time , length , and force data to allow quantitative estimates of internal actin network parameters that are not directly measureable . For example , excising a cubic “slab” of a calibrated nodes-and-link network , then performing “computer experiments” by compressing , stretching , and shearing this slab in silico and recording the resulting stresses will allow us to compute the effective macroscopic elastic moduli of the in silico network , including Young modulus and Poisson ratio . More experimental data will also allow refinement of the functional forms of the repulsive and link forces , and to determine the extent that polymerization is regulated by force . The APS model also offers a general framework to help investigate other physical cell phenomena that may be dominated by similar , relatively simple viscoelastic behaviors , e . g . , lamellipodia and pseudopodia extension and cell septation , by including the effects of interactions with cell membranes , and simulating the anisotropic networks and contractile proteins found in vivo .
A brief overview of the model is given here ( more details are available in the supporting text ( Protocol S1 ) Sections S1 and S4 , S5 , S6 , S7 , S8 ) . We simulate the network using a discrete-element approach , i . e . , the actin network is represented as network of nodes in 3-D space held together by links ( Figures S1 and S2 ) . This is unlike a finite element approach in which the mesh is a way to reduce the dimensionality of a continuum problem into finite number of equations ( elements ) . Rather , network links and the effective mesh size that results are important properties of the network . Network links also have no direct correspondence to actin filaments , but rather the bulk viscoelastic properties of the network of links and nodes are intended to capture the bulk viscoelastic network properties of the actin network . Under the polymerization conditions used ( i . e . , in the absence of crosslinking proteins ) nodes more properly correspond to entanglement of filaments , and links correspond to the elastic properties of the network . We model these links as simple linear springs with a defined breaking strain and an inverse square repulsive force between nodes that models the compression resistance of the material . We explicitly avoid the unresolved question of how polymerizing filaments behave on a molecular level at the nucleator surface ( according to Brownian ratchet or other models [24] , [49] ) , and model polymerization as the stochastic introduction of material ( nodes ) at constant rate at the nucleator surface . Simulations begin at t = 0 with zero nodes ( and links ) . Once introduced , new nodes form links with their neighbors , with a higher probability of forming links with nearby nodes ( linear tail-off with distance , max probability PXL at zero distance ) , and a limit on the maximum number of links . Nodes at the surface of the bead are also linked to the bead at their last contact point by a link with force proportional to its length . Forces are calculated iteratively ( Figures S3 and S21 ) , and since this is a low Reynolds number regime , there is no inertia ( i . e . , velocity is proportional to force . ) The computational model is implemented in C++ , and run times to symmetry breaking are approximately 1–2 h on a typical desktop computer . The code is designed to use multiple threads to enable large-scale problems to be explored across a number of parameter regimes ( runs typically involve 105 nodes , 106 links , and 106 iterations per simulation ) . The code is open source and made freely available under the GNU General Public License to allow the results to be reproduced , to convey the full details of the model , and to encourage further use of the code by other researchers . A snapshot of the source code together with the parameter control file ( Protocol S2 ) and a compiled executable for Mac OS X ( Protocol S3 ) are provided . A detailed explanation of the code and the parameter control file are included in the supporting text ( Protocol S1 ) and in an online wiki at http://www . dayel . com/comet , where the latest version of the code can also be downloaded . To visualize the results of the simulations in a way comparable to in vitro microscopy images , we calculate the symmetry breaking plane , and create a 2-D projection of the nodes of the network convolved with a Gaussian to represent the point spread function of the microscope . To make visual comparison easier , we rotate the reference frame afterwards so that the bead always appears to move to the right . Measurements of forces in the radial and circumferential directions in Figure 4 are calculated as components in the direction of , or perpendicular to , a vector from the bead center , the magnitudes of which are summed over spherical shells of different radii . “Stretch factor” measures in Figures 4 and 5 are calculated by measuring the distance between particular pairs of nodes over time , normalized to the initial distance then averaged . Bead motility experiments were carried out as previously described [16] , with modifications . Briefly , 5-µm diameter carboxylated polystyrene beads ( Bangs Laboratories ) were covalently coated with ActA . The motility mix contained 0 . 5 mM ATP , 1 mM MgCl2 , 1 mM EGTA , 15 mM TCEP-HCl , 50 mM KOH ( to neutralize TCEP-HCl ) , 20 mM HEPES ( pH 7 . 0 ) , 125 nM Arp2/3 complex , 100 or 120 nM capping protein , and 3 µM actin . To aid microscopic observation , we included 3 mg/ml BSA ( A0281; Sigma-Aldrich ) and 0 . 2% methylcellulose ( M0262; Sigma-Aldrich ) . Initial attempts to define headspace by controlling reaction volume were unsuccessful—the coverslip was not perfectly parallel to the slide , causing the headspace to vary across the sample—so we controlled the headspace by adding 0 . 1% v/v 5 . 1-µm or 15 . 5-µm diameter glass spacer beads ( Duke Scientific ) prior to starting the reaction . For 3-D reconstructions , reactions were stopped before imaging by adding 50% volume of 15 µM phalloidin and 15 µM Latrunculin B ( Sigma-Aldrich ) . Fluorescent speckle microscopy ( Figure 3A ) conditions: 7 . 5 µM actin ( 1/3 , 000 TMR-labeled ) , 3 µM profilin , 40 nM Arp2/3 , and 56 nM capping protein . For the ellipsoidal bead experiments , spherical beads were stretched as previously described [50] with the following modifications: 140 µl of polystyrene bead stock was suspended in 6 ml of 3 . 8% w/v suspension of polyvinyl alcohol ( PVA ) . The PVA/bead suspension was degassed before casting films in a 4 . 5×7 . 0 cm leveled tray . After stretching , the PVA was dissolved by incubating at 90°C for 2 h in distilled water containing 0 . 1% NP-40 . The beads were washed three times in isopropanol and dried in a rotary evaporator . The bead surface was refunctionalized by incubation in 50% ( w/v ) NaOH for 1 h at 90°C and overnight at 42°C , washed once with 20 mM Tris HCl ( pH 8 . 0 ) and 0 . 1% NP-40 , and three times with 0 . 1% NP-40 before coating with ActA .
|
Networks of actin filaments provide the force that drives eukaryotic cell movement . In a model system for this kind of force generation , a spherical bead coated with an actin nucleating protein builds and rockets around on an actin “comet tail , ” much like the tails observed in some cellular systems . How does a spherically symmetric bead break the symmetry of the actin coat and begin to polymerize actin in a directional manner ? A previous theoretical model successfully explained how symmetry breaks , but suggested that the subsequent motion was driven by actin squeezing the bead forwards—a prediction refuted by experiment . To understand how motility occurs , we created a parsimonious computer model that predicted novel experimental behaviors , then performed new experiments inspired by the model and confirmed these predictions . Our model demonstrates how the elastic properties of the actin network explain not only symmetry breaking , but also the details of subsequent motion and how the bead maintains direction .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Conclusions",
"Materials",
"and",
"Methods"
] |
[
"biophysics/theory",
"and",
"simulation",
"cell",
"biology/cytoskeleton"
] |
2009
|
In Silico Reconstitution of Actin-Based Symmetry Breaking and Motility
|
Coxiella burnetii is an intracellular bacterial pathogen that causes human Q fever , an acute flu-like illness that can progress to chronic endocarditis and liver and bone infections . Humans are typically infected by aerosol-mediated transmission , and C . burnetii initially targets alveolar macrophages wherein the pathogen replicates in a phagolysosome-like niche known as the parasitophorous vacuole ( PV ) . C . burnetii manipulates host cAMP-dependent protein kinase ( PKA ) signaling to promote PV formation , cell survival , and bacterial replication . In this study , we identified the actin regulatory protein vasodilator-stimulated phosphoprotein ( VASP ) as a PKA substrate that is increasingly phosphorylated at S157 and S239 during C . burnetii infection . Avirulent and virulent C . burnetii triggered increased levels of phosphorylated VASP in macrophage-like THP-1 cells and primary human alveolar macrophages , and this event required the Cα subunit of PKA . VASP phosphorylation also required bacterial protein synthesis and secretion of effector proteins via a type IV secretion system , indicating the pathogen actively triggers prolonged VASP phosphorylation . Optimal PV formation and intracellular bacterial replication required VASP activity , as siRNA-mediated depletion of VASP reduced PV size and bacterial growth . Interestingly , ectopic expression of a phospho-mimetic VASP ( S239E ) mutant protein prevented optimal PV formation , whereas VASP ( S157E ) mutant expression had no effect . VASP ( S239E ) expression also prevented trafficking of bead-containing phagosomes to the PV , indicating proper VASP activity is critical for heterotypic fusion events that control PV expansion in macrophages . Finally , expression of dominant negative VASP ( S157A ) in C . burnetii-infected cells impaired PV formation , confirming importance of the protein for proper infection . This study provides the first evidence of VASP manipulation by an intravacuolar bacterial pathogen via activation of PKA in human macrophages .
Coxiella burnetii is an intracellular bacterial pathogen that causes the zoonosis human Q fever . C . burnetii infects domestic mammals and livestock , which serve as the primary reservoir for the pathogen in nature . C . burnetii is shed from infected animals in body fluids , particularly during parturition , resulting in human infection by inhalation of contaminated aerosols [1] . Q fever often manifests as a flu-like acute disease with atypical pneumonia , and most individuals recover without medical intervention . However , less than 5% of infected individuals develop chronic disease that largely manifests as endocarditis , and to a lesser extent as bone infection , vascular complications , and granulomatous hepatitis [2] . The fatality rate of patients with Q fever endocarditis approaches 60% if left untreated [1] . Chronic Q fever diagnosis is extremely difficult and treatment requires a prolonged course of antibiotic therapy that is not completely effective . Considering the global distribution of C . burnetii , and recent outbreaks in rural parts of the world , Q fever is now considered an emerging infectious disease [3 , 4] . In aerosol-acquired human infections , C . burnetii preferentially targets macrophages that reside in alveolar spaces . Virulent bacteria enter macrophages by αVβ3 integrin receptor-dependent phagocytosis [5] . Following invasion , organisms reside in tight-fitting phagosomes that decorate with the autophagosome marker LC3 and early endosomal Rab5 [6] . Maturation of nascent phagosomes into unique , replication-permissive parasitophorous vacuoles ( PV ) is achieved by continual heterotypic fusion with autophagosomes , endosomes , and lysosomes [7–9] . Although lysosome fusion creates an acidic , hydrolytic environment , C . burnetii has adapted to resist degradation , and low pH activates bacterial metabolism and subsequent replication [10 , 11] . C . burnetii actively controls infection by directing endosomes , vesicles , autophagosomes , and lysosomes to the PV , delivering nutrients , lipids , and proteins to the expanding vacuole that ultimately occupies most of the host cell cytosol [12 , 13] . Secretion of bacterial effector proteins into the host cytosol via a Dot/Icm type IV secretion system ( T4SS ) is essential for PV formation [14 , 15] . Some C . burnetii effectors contain sequences that resemble eukaryotic motifs and domains , bind host cell proteins , and manipulate host signaling to promote infection [16–18] . To support an intracellular lifestyle , C . burnetii manipulates several host cell signaling pathways . The pathogen activates Akt and Erk1/2 to promote host cell survival and allow completion of a lengthy infectious cycle [19] . C . burnetii also hijacks cyclic adenosine monophosphate ( cAMP ) -dependent protein kinase ( PKA ) signaling to support PV formation and prevent apoptotic cell death . When activated by cAMP , PKA binds and phosphorylates several downstream target proteins that modulate responses including cytokine production , apoptosis , and cytoskeletal remodeling . Differential phosphorylation of PKA substrates has been observed in C . burnetii-infected macrophages , and PKA signaling is indispensable for PV formation and intracellular bacterial replication [20] . Moreover , PKA phosphorylates Bcl-2-associated death promoter ( Bad ) , resulting in sequestration of Bad to the PV and prevention of apoptosis [21] . Although most PKA substrates differentially phosphorylated during C . burnetii infection have not been characterized , they likely play unique roles in PV formation and Q fever pathogenesis . PKA activity is critical for actin-related processes in eukaryotic cells and actin polymerization is involved in the intracellular lifestyle of multiple bacterial pathogens . A subset of intracellular bacteria recruit actin regulatory proteins to the bacterial cell surface to form actin tails that propel the pathogen through the cytosol and facilitate cell-to-cell movement without exposure to host immune cells . For example , Listeria monocytogenes produces ActA that recruits Arp2/3 and the PKA target protein vasodilator-stimulated phosphoprotein ( VASP ) to polymerize actin and form actin comets [22 , 23] . Manipulation of VASP by bacteria that do not use actin-based motility , such as C . burnetii , has not been reported . However , actin reorganization is essential for PV expansion in HeLa cells infected with avirulent C . burnetii [24] . In this study , we identified VASP as a PKA substrate that is preferentially activated during C . burnetii growth in human macrophages . Because PKA signaling is manipulated by C . burnetii [20 , 21] and VASP is a PKA substrate that regulates actin polymerization , we predicted that VASP is involved in PV formation and/or maintenance . Our findings indicate VASP is phosphorylated by PKA during infection in a T4SS-dependent fashion , and VASP activity is required for PV formation and C . burnetii replication in human macrophages . Additionally , we identified VASP residues ( S157 and S239 ) required for PV formation and heterotypic fusion with other compartments . This study provides the first evidence of an intravacuolar bacterial pathogen usurping VASP function to promote replication vacuole formation .
We previously showed that PKA substrates are differentially phosphorylated during C . burnetii infection of human macrophages [20] . Moreover , PKA activation and phosphorylation of downstream targets occurs throughout intracellular growth ( 24–96 h post-infection ( hpi ) ) . To identify PKA target proteins with increased phosphorylation during infection , we immunoprecipitated PKA substrates from C . burnetii-infected THP-1 macrophage-like cells using a PKA phospho substrate-specific antibody . This antibody specifically recognizes proteins with phosphorylated Ser/Thr residues with arginine at the -3 and -2 positions ( RRXS/T ) . At 24 and 96 hpi , immunoprecipitated proteins were analyzed by Coomassie blue staining . We consistently observed an increase in the levels of a ~47 kDa protein in C . burnetii-infected cells ( Fig 1A ) . Mass spectrometry analysis identified this protein as eukaryotic vasodilator-stimulated phosphoprotein ( VASP ) , a known PKA target [25] . VASP is involved in actin polymerization , contains distinct protein-protein interaction domains , and is regulated by phosphorylation at five residues ( Fig 1B ) . Immunoblot analysis using a VASP-specific antibody confirmed increased VASP levels in proteins immunoprecipitated from infected cells ( Fig 1C ) . PKA commonly phosphorylates VASP at S157 [26] . Therefore , we further assessed whether phosphorylated VASP ( S157 ) was immunoprecipitated with the PKA phospho substrate-specific antibody . As shown in Fig 1C , immunoblot and densitometry analysis revealed an increase in phosphorylated VASP ( S157 ) levels in immunoprecipitated samples collected from infected cells , indicating VASP is targeted by PKA during C . burnetii intracellular growth . VASP is involved in remodeling actin and is regulated by phosphorylation at Y39 , S157 , S239 , S322 , and T278 . PKA , which plays a significant role in PV biogenesis [20] , is known to phosphorylate VASP at S157 , S239 , and T278 . Actin remodeling is a dynamic process involved in PV formation and/or maintenance [24] . Therefore , we predicted the kinetics of VASP phosphorylation may differ during early and late stages of infection . To test this hypothesis , we infected THP-1 cells with C . burnetii and collected cell lysates at 2–96 hpi . Immunoblot analysis using a specific antibody directed against phosphorylated VASP ( S157 ) revealed low levels of phosphorylated protein in uninfected cells ( Fig 2A ) . Cells infected with C . burnetii for 2 and 6 h also did not show a significant increase in levels of phosphorylated VASP , suggesting VASP activity is not required for early infection events , such as cell adherence , phagocytosis , and nascent phagosome formation . However , VASP phosphorylation levels increased significantly at 24 hpi and were maintained through 96 hpi ( Fig 2A ) . Levels of total VASP remained unaltered in C . burnetii-infected cells , indicating VASP expression and turnover were not substantially altered by the pathogen . Furthermore , VASP can be phosphorylated at S239 by PKA or protein kinase G , and at T278 by adenosine monophosphate-activated protein kinase [25 , 26] . Although C . burnetii infection did not trigger increased levels of phosphorylation at T278 or S322 in THP-1 cells , significantly increased levels of VASP phosphorylated at S239 were evident with similar timing to S157 phosphorylation ( Fig 2A ) . Although VASP phosphorylation increased prior to vacuole expansion ( 24 hpi ) , we sought to confirm that phosphorylation was not simply a consequence of the presence of a large vacuole . Cells were infected with C . burnetii deficient in CpeD or CpeE , which are both T4SS-secreted proteins . CpeD and CpeE mutant bacteria entered cells and were maintained in a vacuole much smaller than a typical PV ( Fig 2B ) . However , increased VASP phosphorylation was apparent at 72 hpi with either CpeD- or CpeE-deficient C . burnetii , demonstrating that phosphorylation does not require a pre-formed large vacuole . Relative to uninfected cells , CpeD mutant-infected macrophages showed an ~ 1 . 9 fold increase in S157 and S239 phosphorylation . CpeE mutant-infected macrophages showed an ~ 8-fold and ~ 2 . 1-fold increase in S157 and S239 phosphorylation , respectively . C . burnetii secretes effector molecules into the host cell cytosol using a Dot/Icm T4SS . PKA activation is triggered by secreted bacterial effectors , suggesting T4SS-defective C . burnetii should not trigger VASP phosphorylation . Therefore , we anticipated that inhibiting C . burnetii protein synthesis or antagonizing PKA signaling would prevent increased VASP phosphorylation . To test these predictions , we examined the status of VASP phosphorylation in the presence or absence of chloramphenicol or the PKA inhibitor H89 at 72 hpi , a time when robust VASP phosphorylation occurs . As shown in Fig 3A , treatment with H89 significantly reduced VASP phosphorylation , confirming VASP is a downstream target of PKA during infection . Treatment with the bacterial protein synthesis inhibitor chloramphenicol also abrogated increased VASP phosphorylation , supporting previous observations that induction of PKA signaling requires metabolically active C . burnetii . Similar to cells treated with H89 or chloramphenicol , IcmD mutant C . burnetii , which lacks a functional T4SS and cannot secrete effectors , did not alter VASP phosphorylation , indicating a secreted effector ( s ) promotes this signaling event ( Fig 3A ) . As expected , treatment with the PKA activator forskolin triggered significant VASP phosphorylation at S157 and S239 . The abundance of total VASP was not altered during infection or any inhibitor treatment . The requirement of PKA for VASP phosphorylation was further confirmed by silencing expression of the Cα subunit of PKA using a siRNA approach . This method depleted > 90% of PKACα relative to THP-1 cells transfected with non-targeting siRNA ( Fig 3B ) . As expected , C . burnetii triggered increased VASP S157 and S239 phosphorylation in non-targeting siRNA-transfected cells . However , decreased expression of PKACα abrogated increased VASP phosphorylation without altering expression of total VASP . Relative to non-targeting siRNA-transfected cells , PKACα-silenced , C . burnetii-infected cells showed an ~ 66% and ~ 52% decrease in S157 phosphorylation and S239 phosphorylation , respectively . Together , these results indicate that C . burnetii triggers PKA- and T4SS-dependent phosphorylation of VASP at S157 and S239 in macrophage-like cells . Based on the results above , we predicted that VASP is required for C . burnetii PV expansion and intracellular growth . To assess the importance of VASP function during infection , THP-1 cells were transfected with VASP-specific siRNA or non-targeting siRNA . Cell lysates were collected from 24–144 h post-transfection and analyzed by immunoblot to confirm VASP knockdown . VASP production decreased >80% using this approach ( Fig 4A ) and phosphorylated VASP ( S157 ) was barely detectable at 24 and 96 h post-transfection . Importantly , VASP silencing did not alter viability of transfected cells at any time point tested ( Fig 4B ) . siRNA-transfected cells were then infected with mCherry-expressing C . burnetii and bacterial growth was monitored by measuring fluorescence for six days . In VASP-depleted cells , a significant reduction in C . burnetii growth was observed relative to non-targeting siRNA-transfected cells ( Fig 4C ) . We next confirmed that reduction of mCherry fluorescence correlated with genome equivalents representing bacterial numbers . THP-1 cells were infected with wild type C . burnetii , total DNA isolated at 24 or 96 hpi , and genome equivalents determined as previously described [27] . As shown in Fig 4D , no significant difference in bacterial numbers was observed at 24 hpi between control and VASP-depleted cells , indicating VASP is not required for bacterial uptake by THP-1 cells . However , the number of bacterial genomes at 96 hpi was ~ 40% lower in VASP-depleted cells , corresponding to mCherry fluorescence results and confirming VASP is necessary for optimal bacterial growth in macrophages . Next , we confirmed the growth curve analyses using confocal microscopy to monitor PV formation . The lysosomal marker CD63 was used to label the PV membrane . As shown in Fig 5 , large ( diameter > 6 μm ) , CD63-decorated PV were present in ~ 50% of non-targeting siRNA-transfected cells and only 7% of PV were smaller than 2 μm . In contrast , the number of vacuoles larger than 6 μm was reduced to 13% in VASP-silenced cells and 40% of PV were smaller than 2 μm ( Fig 5 ) . CD63 presence on small vacuoles in VASP-silenced cells indicated that VASP knockdown did not alter invasion events to prevent C . burnetii entry into the host cell or prevent phagolysosomal maturation . However , VASP silencing prevented vacuole expansion , potentially limiting available space and nutrients for replicating C . burnetii . Additionally , the impact of decreased VASP expression on PV formation was confirmed using a second set of VASP siRNA constructs . Together , these results indicate that VASP activity is required for optimal PV expansion and C . burnetii growth in human macrophages . Avirulent and virulent C . burnetii produce different LPS structures , with virulent phase I LPS masking cell wall antigens that activate Toll-like receptors , preventing activation of the innate immune response [28] . Avirulent phase II C . burnetii is widely used for in vitro studies to characterize pathogen-host cell interactions [29] . However , results obtained from avirulent studies should be validated using virulent C . burnetii , as human disease is caused by phase I organisms . Similar to avirulent bacteria , THP-1 cells infected with virulent C . burnetii contained increased levels of phosphorylated VASP ( S157 and S239 ) from 24–96 hpi without altering levels of total VASP ( Fig 6A , left panel ) . siRNA-mediated knockdown of VASP expression substantially reduced PV expansion and resulted in smaller vacuoles ( Fig 6A , middle and right panels ) . These results indicate that virulent C . burnetii uses host VASP for optimal PV formation . In addition to obtaining efficient silencing using individual siRNA , we observed similar PV formation results when cells were transfected with a mixture of VASP siRNA constructs . Although THP-1 macrophage-like cells are commonly used as an in vitro cellular model [17 , 30 , 31] , C . burnetii preferentially infects and replicates within human alveolar macrophages ( hAMs ) during natural infection [32] . Additionally , alveolar macrophages express substantial levels of VASP [33] . Therefore , we isolated cells from human lungs post-mortem and infected primary hAMs with virulent C . burnetii . As shown in Fig 6B ( left panel ) , infection triggered increased VASP phosphorylation ( S157 and S239 ) from 24–96 hpi similar to THP-1 cells . Knockdown of VASP expression by siRNA interfered with PV expansion and cells contained smaller vacuoles relative to non-targeting siRNA-transfected cells ( Fig 6B , middle and right panels ) . These results demonstrate the importance of VASP activity in a natural C . burnetii disease scenario . Phosphorylation of VASP at S157 and S239 regulates protein localization and actin polymerization , respectively [26] . We specifically tested whether motifs containing S157 or S239 are required for PV formation by expressing GFP-VASP constructs with individual residues mutated ( S157E and S239E ) in C . burnetii-infected THP-1 cells . A serine to glutamic acid mutation mimics the conformation of a phosphorylated serine with respect to negative charge , maintaining the structure of a phosphorylated protein [34] . However , the ability to bind target proteins that directly recognize phospho-serine motifs is prevented by this mutation . To examine functional effects of individual VASP mutants , we assessed VASP localization , actin arrangement , and PV formation in infected cells . As shown in Fig 7 , large typical PV were present at 72 hpi in GFP-VASP and GFP-VASP ( S157E ) -expressing cells . Additionally , wild type and VASP ( S157E ) co-localized with actin around the PV . In contrast , expression of GFP-VASP ( S239E ) resulted in substantially smaller PV containing fewer C . burnetii . Furthermore , actin levels were reduced around the PV in GFP-VASP ( S239E ) -expressing cells . These results indicate proper S239 phosphorylation and regulation of distinct downstream target proteins is critical for infection-specific VASP functions potentially through interactions with a host or bacterial protein ( s ) . Reduced actin polymerization may also destabilize the PV , resulting in smaller vacuoles . PV generation involves numerous heterotypic fusion events with phagosomes , autophagosomes , and lysosomes [29] . To determine if VASP function is required for phagosome trafficking to the PV , THP-1 cells were infected with C . burnetii for 72 h , then incubated with fluorescent beads , a common technique for assessing PV heterotypic fusion with cellular compartments [35] . As shown in Fig 8 , beads were delivered to the PV in cells expressing GFP-VASP or GFP-VASP ( S157E ) . In contrast , beads were sequestered away from PV in cells expressing GFP-VASP ( S239E ) , corresponding to the PV formation results observed above . These results indicate proper VASP S239 activity is critical for heterotypic fusion events during C . burnetii infection . To further assess the requirement of VASP phosphorylation for PV development , cells were transfected with constructs encoding phosphorylation-defective VASP mutants ( S157A and S239A ) . In contrast to the phospho-mimetic mutants above , these proteins serve as dominant negative mutants and are not phosphorylated . As shown in Fig 9 , cells expressing GFP-VASP or GFP-VASP ( S239A ) contained expanded PV while cells expressing GFP-VASP ( S157A ) harbored multiple small atypical PV . These results indicate that S157 phosphorylation is critical for PV expansion .
Here , we demonstrate , for the first time , that eukaryotic VASP can be exploited by an intracellular bacterial pathogen to promote replication vacuole formation . Successful PV formation by C . burnetii is necessary for development of Q fever and the pathogen actively manipulates host signaling using effector proteins secreted into the host cell cytosol via a Dot/Icm T4SS . Previous studies in our laboratory demonstrated that PKA is hijacked by C . burnetii to facilitate PV formation and prevent apoptotic cell death [20 , 21] . In this study , we identified VASP as a PKA substrate that is activated during infection by T4SS-proficient C . burnetii . VASP is an essential protein for actin remodeling; therefore , we hypothesize that VASP activity is required for actin-dependent PV expansion and/or maintenance in human macrophages . As a member of the Ena/VASP protein family , VASP has an N-terminal EVH1 domain , C-terminal EVH2 domain , and proline-rich region ( Fig 1B ) . The EVH1 domain binds to focal adhesion proteins that anchor VASP to the integrin complex of cell membranes [36 , 37] . The proline-rich region binds to SH3 and WW domain-containing proteins and profilin , which catalyze actin monomer formation and fast actin polymerization [38] . The EVH2 domain mediates F- and G-actin binding to regulate actin polymerization [39 , 40] . Anti-capping is one mechanism proposed for VASP-dependent actin polymerization where VASP binds to actin filament barbed ends and prevents recruitment of capping proteins , resulting in long actin filaments [41] . Additionally , VASP can inhibit Arp2/3-dependent actin filament branching; however , a molecular mechanism has not been characterized [23] . VASP dysfunction has been linked to many diseases including cancer , atherosclerosis and thrombosis [42] . VASP contains three major phosphorylation sites , S157 , S239 , and T278 that control numerous cellular functions . For example , phosphorylation by PKA at S157 facilitates localization to the cell periphery into focal adhesions [43] , whereas phosphorylation at S239 and T278 regulates F-actin assembly [26] . During C . burnetii infection , prolonged VASP phosphorylation ( 24–96 hpi ) occurs at S157 and S239 . S157 is located adjacent to a proline-rich region and controls VASP translocation to the cell membrane , while exerting a minimal effect on F-actin polymerization [25 , 26 , 43–45] . S157 phosphorylation also blocks VASP interaction with Abl , SH3 domains , and Src proteins [26 , 46] . Phosphorylation of S239 , which is located adjacent to a G-actin binding motif in the EVH2 domain , negatively regulates VASP anti-capping activity , resulting in shortened F-actin filaments , reduced F-actin accumulation , and filopodia formation [26 , 45 , 47 , 48] . Phospho-mimetic mutation of S239 reportedly decreases F-actin accumulation , does not efficiently co-localize with actin , and antagonizes adhesion and spreading of smooth muscle cells [49] . The timing of increased VASP phosphorylation during C . burnetii infection correlates with PV biogenesis and expansion , events that are regulated by T4SS effectors . Moreover , C . burnetii actively replicates during this time , and bacterial protein synthesis and T4SS activity are required to trigger increased VASP phosphorylation levels . Therefore , we predict a distinct secreted effector ( s ) hijacks PKA signaling to promote VASP phosphorylation . PKA and protein kinase C ( PKC ) can phosphorylate VASP at S157 in human platelets [50] and protein kinase G preferentially phosphorylates the protein at S239 [45 , 50] . However , PKA phosphorylates S157 and S239 during C . burnetii infection as demonstrated by specific inhibitor and siRNA treatments . Four different PKA catalytic subunits , Cα , Cβ , Cγ , and X chromosome encoded protein kinase X ( PRKX ) , have been identified in humans [51] . Using a confirmatory siRNA knockdown approach , it is clear that the PKACα subunit is required for S157 and S239 phosphorylation during infection . PKACα promotes breast cancer cell viability by inactivating the pro-apoptotic BCl-2-associated death promoter protein [52] . We previously showed that PKA promotes macrophage survival during C . burnetii infection , although the specific subunit responsible has not been defined . Considering an important role in cell survival , we anticipate PKACα is critical for PKA pro-survival signaling and VASP phosphorylation during infection . A subset of intracellular pathogens , such as Listeria monocytogenes , recruit VASP to facilitate actin tail formation [53] . Resulting actin tails propel L . monocytogenes through the cytosol and to the cell surface , facilitating spread to other cells . Thus , hijacking VASP for actin-based motility is a critical part of the L . monocytogenes pathogenic life cycle . In contrast , some intracellular bacteria , such as Shigella spp . , do not require VASP for actin-based mobility [54] . To our knowledge , the current study provides the first evidence of VASP manipulation by a non-motile , intravacuolar bacterium to control replication vacuole formation and promote intracellular replication . Indeed , siRNA-mediated VASP knockdown prevents typical C . burnetii growth and PV formation . Additionally , VASP activity is required during virulent C . burnetii infection of primary hAMs , suggesting the protein is necessary for optimal pathogen growth in the human lung . We predict that VASP-dependent actin rearrangements around the PV are required for vacuole stability and expansion . This prediction is supported by co-localization of GFP-VASP and actin at the PV membrane . Our macrophage infection results are consistent with reports suggesting actin is required for avirulent C . burnetii PV formation in HeLa cells [24] . Using F-actin depolymerizing chemical agents , Aguilera et al . showed that F-actin polymerization facilitates fusion between the PV and bead-containing phagosomes . Autophagosomes , lysosomes , and phagosomes continuously fuse with the PV , providing nutrients and membrane , and VASP may facilitate these actin-dependent vesicular fusion events . Additionally , actin may provide structural support for the maturing PV , allowing controlled expansion within the cytosol . A role for actin in structural support has been reported for Chlamydia trachomatis replication vacuole formation [55] . During growth in epithelial cells , actin cages form around the C . trachomatis inclusion , and disruption of actin polymerization diminishes vacuole membrane integrity . Therefore , it is possible that VASP is involved in growth of C . trachomatis and other intravacuolar pathogens by regulating formation of a nucleating complex at the replication vacuole membrane . Phosphorylated VASP motifs recruit adaptor proteins that assemble protein complexes . For example , members of the 14-3-3 family , WD40 domain-containing F-box proteins , and WW domain-containing proteins form multimolecular signaling complexes through specific interactions with phospho-Ser/Thr motifs [56] . It is not known if similar proteins bind to VASP motifs to regulate macrophage responses to C . burnetii . However , ectopic expression of VASP ( S239E ) negatively impacts PV formation and reduces actin accumulation around the PV . In contrast , expression of VASP ( S157E ) does not alter PV formation or actin accumulation at the vacuole , consistent with reports that S157 does not significantly impact F-actin polymerization . During infection , VASP phosphorylated at S239 may localize to specific regions around the PV where controlled actin depolymerization could occur to provide space for vacuole expansion . Although VASP ( S239E ) has a negative charge , is functionally active , and negatively impacts actin polymerization , the mutant protein may not assemble phospho-Ser/Thr motif-binding protein complexes needed for PV formation . Moreover , VASP-mediated arrangement of actin ultrastructure around the PV may facilitate fusion of incoming vesicles and phagosomes with the PV . Indeed , bead trafficking results indicate S239 is critical for phagosome movement to the PV . These results do not prove that VASP plays a direct role in fusion events , but rather allows incoming phagosomes access to the PV along the cytoskeletal network . Expression of VASP ( S157A ) severely impairs PV expansion , while VASP ( S239A ) expression allows normal vacuole development , demonstrating S157 phosphorylation is critical for infection . VASP phosphorylation at S157 prevents interactions with the SH3 domains of Abl , alpha II spectrin , and Src [46 , 57] , and also regulates VASP cellular distribution . The dominant negative S157A mutant , which is not phosphorylated , may facilitate interaction with SH3 domain-containing proteins , preventing typical PV formation . The finding that expression of the phosphomimetic S157E mutant , which mimics phosphorylated VASP and provides conformational similarity , does not adversely impact PV formation supports the importance of S157 phosphorylation for optimal PV formation . Additionally , S157 phosphorylation minimally impacts actin filament formation [26] . Therefore , S157 phosphorylation may primarily act via regulation of other targets that contribute to optimal PV formation , and not by direct actin rearrangement . In contrast , expression of the dominant negative S239A mutant , has no significant impact on PV formation . Under physiological conditions , S239 is phosphorylated in small quantities and the level of phosphorylation is tightly regulated [26 , 47] . Therefore , PV-localized S239-phosphorylated VASP may reduce local F-actin levels , providing the necessary space for incoming cargo-laden vesicles and facilitating fusion events . It is also possible that infected macrophages regulate actin rearrangement via redundant mechanisms in the absence of S239-phosphorylated VASP . In conclusion , C . burnetii hijacks host PKA signaling to phosphorylate VASP and facilitate PV formation . VASP phosphorylation requires bacterial protein synthesis and an active T4SS , indicating the pathogen actively targets this pathway . VASP may directly regulate actin polymerization dynamics , providing structural stability and physical space for the expanding PV needed for bacterial replication . Supporting this prediction , depletion of VASP negatively impacts intracellular bacterial growth and PV size . Phosphorylation at S157 and S239 is critical for VASP promotion of PV formation . Overall , this study provides the first evidence of a non-motile intravacuolar bacterial pathogen manipulating eukaryotic VASP to facilitate intracellular growth .
Avirulent C . burnetii Nine Mile phase II ( NMII; RSA439 , clone 4 ) , virulent Nine Mile phase I ( NMI; RSA493 ) , ΔCpeD , and ΔCpeE bacteria were cultured in acidified citrate cysteine medium ( ACCM ) at 37°C with 5% CO2 and 2 . 5% O2 for 7 days . Cultures were then washed and resuspended in sucrose phosphate buffer ( pH 7 . 4 ) . mCherry-expressing C . burnetii was grown in ACCM supplemented with chloramphenicol ( 3 μg/ml; Sigma ) . IcmD mutant C . burnetii ( icmD::Tn ) [14] was grown in ACCM supplemented with kanamycin ( 350 μg/ml ) . Construction of CpeD- and CpeE-deficient strains is described in S1 Text . All work with virulent C . burnetii was performed in the Centers for Disease Control and Prevention-approved biosafety level-3 facility at the University of Arkansas for Medical Sciences . Human THP-1 monocytes ( TIB-202; ATCC ) were cultured in RPMI1640 medium ( Invitrogen ) supplemented with 10% fetal bovine serum ( FBS; Invitrogen ) . Before infection , THP-1 cells were treated with phorbol 12-myristate 13-acetate ( PMA; 200 nM ) overnight for differentiation into macrophage-like cells . After replacing PMA-containing medium with fresh medium , THP-1 cells were infected with C . burnetii at a multiplicity of infection ( MOI ) of 10 . To facilitate infection , plates were centrifuged at 900 rpm for 5 minutes , then incubated for 2 h at 37°C . Cells were washed with medium to remove excess C . burnetii and supplemented with fresh medium for the duration of the infection . Primary human alveolar macrophages ( hAMs ) were collected from lung tissue obtained post-mortem from the National Disease Research Interchange and maintained in DMEM/F12 containing 10% FBS as previously described [36] . Use of primary hAMs was assessed by the University of Arkansas for Medical Sciences institutional review board and deemed to not be human subjects research ( #87788 ) . THP-1 cells were infected with C . burnetii and harvested at 72 hpi in non-denaturing buffer ( 20 mM Tris-HCl pH 7 . 4 , 0 . 1% Triton X-100 , 150 mM NaCl , 2 mM NaF , 0 . 1% glycerol , and a protease and phosphatase inhibitor cocktail ) . Uninfected cells were processed as controls . Total protein was quantified using a DC protein assay ( BioRad ) and confirmed by immunoblot to detect β-tubulin . Immunoprecipitations ( IPs ) were performed using a Classical IP kit ( Pierce ) . Lysates were pre-cleared by incubation with agarose resin for 1 h at 4°C . For immune complex preparation , 1 mg of total protein was incubated overnight with 10 μg of anti-PKA phospho substrate antibody at 4°C . Immune complexes were captured by incubation with protein A/G agarose beads for 1h at 4°C , then eluted in elution buffer ( 0 . 12 M Tris-HCl , 2% SDS , 20% glycerol , pH 6 . 8 ) . Eluted PKA phospho substrate antibody was assessed by immunoblot and probing with anti-IgG to confirm equal amounts of capture antibody in control and infected samples . THP-1 cells in 6-well plates were harvested in lysis buffer ( 50 mM Tris , 1% sodium dodecyl sulfate ( SDS ) , 5 mM EDTA , 5 mM EGTA , 1 mM sodium vanadate , protease and phosphatase inhibitor cocktails ) . Lysates were passed through a 26 gauge needle 10 times and quantified using the DC protein assay . Equal amounts of total protein were separated using 4–15% Mini-PROTEAN TGX gels ( BioRad ) . Proteins were then transferred onto a polyvinylidene fluoride membrane ( BioRad ) , and membranes blocked with 5% milk in Tris-buffered saline containing 0 . 1% Tween 20 ( TBS-T ) . Membranes were probed with mouse anti-human VASP ( BD Biosciences ) , rabbit anti-human p-VASP S157 ( Cell Signaling ) , rabbit anti-human p-VASP S239 ( Cell Signaling ) , rabbit anti-human p-VASP T278 ( Sigma ) , rabbit anti-human p-VASP S322 , or mouse anti-human β-tubulin ( Sigma ) primary antibodies diluted in TBS-T with 5% BSA . Anti-mouse or anti-rabbit IgG secondary antibodies conjugated to horseradish peroxidase ( Cell Signaling ) were used for chemiluminescence-based detection . Secondary antibodies were diluted in 5% milk-containing TBS-T and membranes incubated 1 h at room temperature . Reacting proteins were visualized using a WesternBright ECL kit ( Advansta ) and exposure to film . Previously optimized protocols were used for chemical inhibitor treatments [20] . THP-1 cells in 6-well plates were infected with C . burnetii as described above . After 2 h , infectious inoculum was replaced with fresh medium and cells were treated with specific inhibitors or inducers . Media was replaced daily and chemicals were present throughout . Whole cell lysates were collected at the indicated times post-infection and processed for immunoblot analysis . H89 ( 10 μM; Sigma ) was used to inhibit PKA activity and forskolin ( 10 μM; Sigma ) was used to trigger PKA signaling . To inhibit intracellular C . burnetii protein synthesis , infected cells were treated with chloramphenicol ( 10 μg/ml ) . Validated human VASP siRNA ( 5′- GGACCUACAGAGGGUGAAAdTdT -3' ) was used for VASP silencing [58] , and non-targeting siRNA ( 5′-UGGUUUACAUGUCGACUAA-3' ) was used as a control for transfection experiments . THP-1 cells were nucleofected with VASP or non-targeting siRNA as previously described [59] with some modifications . THP-1 monocytes ( 3 X 106 ) were resuspended in Human Monocyte Nucleofector Solution ( Lonza ) . siRNA ( 1 μg ) was mixed with cells , and cells were transfected using a Nucleofector 2b and program Y001 ( Lonza ) . Following nucleofection , cells were transferred into fresh culture medium , then incubated 4 h at 37°C for recovery . Cells were treated overnight with PMA for differentiation into macrophage-like cells . After removing PMA-containing medium and replacing with fresh medium , monolayers were infected with C . burnetii as described above . For siRNA-mediated knockdown of VASP expression in hAMs , DharmaFECT 1 transfection reagent ( Thermo Scientific ) was used according to the manufacturer’s instructions . hAMs were cultured on coverslips in 24 well plates . Transfection complex was formed using VASP siRNA ( 50 nM ) and DharmaFECT 1 reagent , then added to cells drop wise and incubated overnight for siRNA uptake . Cells were washed once , supplemented with fresh medium , and infected with C . burnetii as indicated above . For ectopic expression , full length human VASP , S157A , S239A , or S157E mutant cDNA cloned into pEGFP-N1 , which were previously characterized for expression of GFP-VASP and GFP-VASP ( S157E ) [60] , were used . Expression of GFP-VASP and GFP-VASP ( S239E ) was achieved using full length VASP or S239E mutant cDNA cloned into pCMV6-AC-GFP ( OriGene ) . THP-1 cells were nucleofected with 1 μg of each plasmid as described above . Viability was determined using a Cell Counting Kit-8 ( Dojindo Laboratories ) according to the manufacturer’s instructions . siRNA-transfected cells were left uninfected or infected for the indicated times in 96-well plates . At each time point , WST-8 ( 2- ( 2-methoxy-4-nitrophenyl ) -3- ( 4-nitrophenyl ) -5- ( 2 , 4-disulfophenyl ) -2H-tetrazolium , monosodium salt ) reagent was added to wells and incubation continued for 4 h at 37°C . Following measurement of the A450 of cultures , viability was calculated using the following formula: { ( Atest-Abackground ) / ( Acontrol-Abackground ) x 100} , where A = absorbance at A450 , background = media alone , and control = non-transfected cells . THP-1 cells were plated on 12 mm diameter circular cover glasses ( Fisher ) in 24-well plates . After treatments and infections , cells were fixed and permeabilized with 100% ice-cold methanol for 3 min and blocked with 0 . 5% bovine serum albumin ( BSA ) in PBS ( pH 7 . 4 ) for 1 h at room temperature . For actin detection , cells were fixed with 4% formaldehyde for 15 min and blocked with 0 . 5% BSA containing 0 . 3% Triton X-100 for 1 h . Cells were incubated with mouse anti-CD63 ( BD Biosciences ) and rabbit anti-C . burnetii primary antibodies in 0 . 5% BSA for 1 h at room temperature . After washing with PBS , cells were incubated with Alexa Fluor 488-conjugated anti-mouse IgG and Alexa Fluor 594-conjugated anti-rabbit secondary antibodies ( Invitrogen ) . Indicated samples were treated with Alexa Fluor 594-labeled phalloidin ( Invitrogen ) for 30 min at room temperature to detect actin . In phalloidin-treated samples , CD63 primary antibody was detected with Alexa Fluor 633-conjugated anti-rabbit secondary antibody . Cells were incubated with DAPI for 5 min at room temperature and mounted with MOWIOL ( Sigma ) . Confocal imaging was performed with a Nikon C2si microscope and data were analyzed using NIS-Elements software ( Nikon ) . PV diameter measurements were taken from at least 15 randomly selected fields . When multiple vacuoles were visible , the two largest vacuoles were measured in a single cell and average vacuole size calculated for all fields . THP-1 cells were transfected with constructs encoding GFP-VASP or serine mutants ( S157E or S239E ) , then infected with C . burnetii as described above . 0 . 5 million fluorescent 1 . 0 μm polystyrene microsphere beads ( Invitrogen ) were added to cells on coverslips and incubated overnight at 37°C . Cells were washed with media to remove excess beads . At 72 hpi , cells were fixed with 4% paraformaldehyde and processed for confocal microscopy as described above . For fluorescence-based growth curves , THP-1 cells were transfected with non-targeting or VASP siRNA as described above . Cells were then cultured in glass flat bottom 96 well black plates . After PMA treatment and removal , cells were infected with mCherry-expressing C . burnetii overnight ( MOI = 10 ) . Medium was then replaced with fresh medium and intracellular C . burnetii growth monitored by fluorescence intensity using a BioTek Synergy H1 microplate reader ( 585 nm excitation/620 nm emission ) for 6 days post-infection . For determination of genome equivalents , infected THP-1 cells were harvested in media by centrifugation ( 10 , 000 x g ) . Cells were disrupted by vortexing with microbeads and total DNA was extracted using an UltraClean Microbial DNA kit ( MoBio Laboratories ) . 10 ng of total DNA was used for quantitative PCR . Previously optimized primers [27] designed to amplify C . burnetii dotA were used with a Power SYBR Green PCR master mix ( Applied Biosystems ) . pCR2 . 1-dotA was serially diluted and used to generate a standard curve and calculate C . burnetii genome copies . Immunoblots were scanned in gray scale with a resolution of 300 dpi and band intensities quantified using ImageJ software ( version 1 . 48v ) . Band intensities were normalized to β-tubulin levels . All experiments were performed at least in triplicate . Statistical significance between experimental and control groups was calculated using a Students t test and Prism 6 software ( GraphPad ) . Results were considered statistically significant when p < 0 . 05 .
|
Q fever , caused by the intracellular bacterial pathogen Coxiella burnetii , is an aerosol-transmitted infection that can develop into life-threatening chronic infections such as endocarditis . The pathogen preferentially grows within alveolar macrophages in a phagolysosome-like compartment termed the parasitophorous vacuole ( PV ) . C . burnetii actively manipulates host cAMP-dependent protein kinase ( PKA ) signaling to promote PV formation and cell survival . Identification of bacterial effector proteins that manipulate PKA and downstream target proteins is critical to fully understand pathogen-mediated signaling circuits and develop new therapeutic strategies . Here , we found that PKA controls vasodilator-stimulated phosphoprotein ( VASP ) activity to promote PV formation and bacterial replication . VASP regulates actin-based motility used by a subset of intracellular bacteria for propulsion through the host cell cytosol and into bystander cells . However , C . burnetii does not use actin-based motility and replicates throughout its life cycle within a membrane bound vacuole . Thus , this study provides the first evidence of VASP manipulation by an intravacuolar bacterial pathogen . Characterization of VASP function in PV formation and identification of additional PKA substrates that promote infection will provide new insight into host-pathogen interactions during Q fever .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"vacuoles",
"intracellular",
"pathogens",
"gene",
"regulation",
"pathogens",
"microbiology",
"immunoprecipitation",
"cellular",
"structures",
"and",
"organelles",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"coxiella",
"burnetii",
"contractile",
"proteins",
"small",
"interfering",
"rnas",
"actins",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"precipitation",
"techniques",
"biochemistry",
"cytoskeletal",
"proteins",
"rna",
"cell",
"biology",
"post-translational",
"modification",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna"
] |
2016
|
Vasodilator-Stimulated Phosphoprotein Activity Is Required for Coxiella burnetii Growth in Human Macrophages
|
Approximately 200 million people worldwide harbour parasitic flatworm infections that cause schistosomiasis . A single drug—praziquantel ( PZQ ) —has served as the mainstay pharmacotherapy for schistosome infections since the 1980s . However , the relevant in vivo target ( s ) of praziquantel remain undefined . Here , we provide fresh perspective on the molecular basis of praziquantel efficacy in vivo consequent to the discovery of a remarkable action of PZQ on regeneration in a species of free-living flatworm ( Dugesia japonica ) . Specifically , PZQ caused a robust ( 100% penetrance ) and complete duplication of the entire anterior-posterior axis during flatworm regeneration to yield two-headed organisms with duplicated , integrated central nervous and organ systems . Exploiting this phenotype as a readout for proteins impacting praziquantel efficacy , we demonstrate that PZQ-evoked bipolarity was selectively ablated by in vivo RNAi of voltage-operated calcium channel ( VOCC ) β subunits , but not by knockdown of a VOCC α subunit . At higher doses of PZQ , knockdown of VOCC β subunits also conferred resistance to PZQ in lethality assays . This study identifies a new biological activity of the antischistosomal drug praziquantel on regenerative polarity in a species of free-living flatworm . Ablation of the bipolar regenerative phenotype evoked by PZQ via in vivo RNAi of VOCC β subunits provides the first genetic evidence implicating a molecular target crucial for in vivo PZQ activity and supports the ‘VOCC hypothesis’ of PZQ efficacy . Further , in terms of regenerative biology and Ca2+ signaling , these data highlight a novel role for voltage-operated Ca2+ entry in regulating in vivo stem cell differentiation and regenerative patterning .
Flatworms ( ‘platyhelminths’ ) comprise a diverse grouping of ∼25 , 000 species representing some of the simplest organisms that are triploblastic and bilaterally symmetric . The majority of flatworms are parasitic ( tapeworms , flukes and skin/gill ectoparasites ) and several are associated with infections in humans and farmed livestock or fish . The most clinically important of these is Schistosomiasis ( Bilharzia ) caused by infection with trematode flukes of the Schistosoma genus that infects ∼200 million people worldwide [1] , [2] . With a high morbidity rate associated with chronic infection , it remains one of the most burdensome tropical diseases . Praziquantel ( PZQ ) has remained the drug of choice for treating Schistosomiasis ( and other cestode infections ) for over 30 years and remains the focus of country-wide treatment regimens . As the mainstay of pharmacotherapy , the fact that the relevant in vivo targets of PZQ remain to be identified prevents rational design of the next generation of antischistosomal chemotherapeutics and is clearly a precarious scenario relative to the potential emergence of drug resistance [3] , [4] . A variety of hypotheses have been advanced concerning possible target ( s ) that mediate PZQ toxicity in schistosomes , encompassing effects on nucleoside uptake [5] , phosphoinositide metabolism [6] , actin [7] , myosin light chain [8] , inhibition of glutathione S-transferase [9] , and stimulation of Ca2+ entry through voltage-operated Ca2+ channels ( VOCCs , [10] , [11] ) . However , no single target has received unequivocal experimental support , and the relevant in vivo molecule ( s ) /pathway ( s ) targeted by PZQ remain elusive [2] , [5] . A smaller grouping of flatworms ( ∼10% of species ) are free-living planarians ( ‘turbellarians’ ) . These organisms have a long history of experimental usage owing to their developmental plasticity and remarkable regenerative abilities . For example , small fragments excised from a planarian have the ability to reform a complete body plan [12]–[14] . This ability is driven by a totipotent population of stem cells , called ‘neoblasts’ that populate the planarian mesenchyme . If a cut fragment contains neoblasts , these cells will migrate toward the wounds and replace appropriate cell types from a regenerative structure ( ‘blastema’ ) formed at the site ( s ) of injury . For example , if a trunk fragment is cut from an intact worm , a new ‘head’ will regenerate at the anterior blastema , a ‘tail’ will regenerate from the posterior blastema and other structures will differentiate in a position-dependent manner , thereby reestablishing the anterior-posterior ( AP ) polarity of the original body plan . Understanding the cellular signaling events which regulate in vivo neoblast differentiation to form the <30 planarian cell types in a robust , positionally correct manner has proved to be a problem that has fascinated biologists for almost 200 years [15] . The utility of planarians as a simple model for studying regenerative biology has stimulated optimization of experimental methods , including in vivo RNAi [16] , [17] , to investigate molecular events involved in neoblast maintenance and differentiation . This experimental tractability to in vivo RNAi , coupled with our discovery of a simple and striking phenotype elicited by PZQ – anteriorization of regeneration to yield two-headed worms – provided opportunity to bring a fresh perspective to the problem of resolving molecules relevant to in vivo PZQ activity . Here , we demonstrate the ability of exogenous PZQ to produce bipolar organisms was ( i ) phenocopied by modulators of calcium ( Ca2+ ) homeostasis , ( ii ) enhanced by a variety of depolarizing stimuli that activate VOCCs and was ( iii ) selectively ablated by in vivo RNAi of VOCC β subunits [18] , [19] , but not by a VOCC α subunit . In lethality assays , at higher PZQ doses and exposure intervals , similar resistance to PZQ was induced in worms where VOCC β were targeted for knockdown by RNAi . Consequently , in terms of the long standing problem of identifying biological target ( s ) of PZQ , these data provide the first in vivo genetic support for the ‘Ca2+ hypothesis’ of PZQ efficacy [10] , [11] , albeit from a free-living flatworm species . Furthermore , this novel activity associated with PZQ establishes voltage-operated Ca2+ influx as a regulator of stem cell differentiation and patterning of the anterior-posterior axis during flatworm regeneration .
An asexual clonal GI strain ( Gifu , Iruma river ) of Dugesia japonica was used in this study [20] . This strain exhibits a robust growth rate ( 1 , 000-fold colony expansion over 3 years ) and broad drug responsiveness [20]–[22] . Planaria were maintained ( ∼10 , 000 worms in 15 L of water in 4 containers ) at room temperature ( 20–23°C ) and fed strained chicken liver puree ( ∼10 ml ) once a week . Regenerative assays were performed using 5 day-starved worms ( n>20 ) in pH-buffered artificial water at 22°C ( 1×Montjuïch salts: 1 . 6 mM NaCl , 1 . 0 mM CaCl2 , 1 . 0 mM MgSO4 , 0 . 1 mM MgCl2 , 0 . 1 mM KCl , 1 . 2 mM NaHCO3 , pH 7 . 0 buffered with 1 . 5 mM HEPES ) . Drugs were sourced as detailed in Table S1 . Isoquinolinone derivatives ( Figure S1 ) were obtained from ChemBridge ( San Diego , CA ) . Phenotypes were scored and archived using a Leica MZ16F stereomicroscope and a QiCAM 12-bit cooled color CCD camera . Drug effects were examined using paired t-tests , with differences considered significant at P<0 . 05 . 45Ca2+ uptake assays ( ∼53 mCi/ml , PerkinElmer ) were performed using trunk fragments incubated in the absence or presence of PZQ ( 70 µM ) for 24 hr and incorporated radioactivity determined after filtration ( GF/C , Whatman ) by liquid scintillation counting [23] . Whole-mount in situ hybridization was performed at 55°C in hybridization solution ( 50% formamide , 5×SSC , 100 µg/ml yeast tRNA , 100 µg/ml heparin sodium salt , 0 . 1% Tween-20 , 10 mM DTT , 5% dextran sulfate sodium salt ) incorporating digoxygenin ( DIG ) -labeled antisense riboprobe ( 40 ng/ml ) denatured at 72°C for 15 min prior to use [20] . A standard mixture of BCIP/NBT in chromogenic reaction solution was used for color development , followed by paraformaldehyde fixation . DIG-labeled antisense riboprobe was synthesized by RNA polymerase ( Roche ) from using either linearized cDNA plasmid or a PCR fragment as the template . Probe regions , and accession numbers for related gene products , were as follows: PC2 ( 1–2285 bp ) ; Inx7 ( 1–1528 bp; AB189256 ) ; Opsin ( 1–475 bp; AJ421264 ) ; Myosin heavy chain ( 4879–5905 bp; AB015484 ) ; Inx3 ( 1–1 , 810 bp; AB189253 ) ; Hox9 ( 1–1 , 491 bp; AB049972 ) ; ndk ( 122–1692 bp; AB071948 ) ; Cavβ1 ( 51–1 , 762 bp; FJ483940 ) ; Cavβ2 ( 1–2 , 017 bp; FJ483939 ) . Total RNA was isolated from 20 intact worms using TRIzol and cDNA subsequently synthesized using the SuperScript III First-Strand Synthesis System ( Invitrogen ) . PCR amplification was performed using degenerative primers ( forward 5′-AAYMANGAYTGGTGGAT-3′; reverse , 5′-GCYTTYTGCATCATRTC-3′ ) for VOCC β subunits and products were cloned into pGEM-T Easy vector ( Promega ) for sequencing . Full length clones were isolated by step-wise screening a cDNA library previously prepared from regenerating fragments of D . japonica [20] . Cavβ1 ( 1653 bp ) , Cavβ2 ( 1421 bp ) and Cavα1 . 1 ( 780 bp ) sequences were amplified using gene specific primers incorporating Kozak sequence and cloned into the IPTG-inducible vector pDONRdT7 [17] using Gateway BP Clonase ( Invitrogen ) . The Djsix-1 clone ( AJ312218 ) was from [24] . Sequence within a Xenopus clone ( IMAGE:4406813 ) with minimal BLAST homology in the planarian genome database was used as a negative control . In vivo RNAi was performed as described previously [17] , [25] with minor modifications . Worms were fed a mix of chicken liver and bovine red blood cells containing transformed HT115 bacteria induced to express individual dsRNA constructs over several feeding/regeneration cycles ( see below ) . To assess the efficiency of knock-down , quantitative real-time PCR ( qPCR ) was performed using a ABI 7500 real-time PCR system ( Applied Biosystems ) and SYBR GreenER qPCR SuperMix Universal ( Invitrogen ) . cDNA ( not containing the RNAi targeted sequence ) for each gene was cloned into pGEM-Teasy vector ( Promega ) and used as a template to create gene-specific standard curves for assessing mRNA levels in samples isolated at equivalent regenerative timepoints from different worms . The mRNA levels of specific genes were compared with controls using planarian β-actin to normalize RNA input . Primers were: Cavβ1: 5′-AGTATTCAGATTACCCGCCTGACAAT-3′ , 5′-CACCAAGATGGATTATCACATGAGA-GA-3′; Cavβ2: 5′-AGACACATACTGGACAGCTACTCATCCT-3′ , 5′-AGCTGAGCTTGTAT-CTGTATTTTTGTTG-3′; β-actin: 5′-GGTAAT-GAACGATTTAGATGTCCAGAAG-3′ , 5′-TCTGCATACGATCAGCAATACCTGGAT-3′; six-1: 5′-CATTTAGTACAAGTGCCACCA-ACATCCA-3′ , 5′-GTTGGATGTTCGGATTTTGATGAGTTCA-3′ . As a further calibration of qPCR results were compared to those from semi-quantitative RT-PCR using the same sample ( data not shown ) .
Figure 1A depicts a simple , manual screen focused on anterior-posterior ( AP ) regenerative polarity in the planarian Dugesia japonica . In this assay , trunk fragments were cut ( heads & tails of worms were amputated ) and incubated in drug-containing solution ( ≤48 hrs ) , after which the media was exchanged for drug-free solution . Regenerative phenotype was scored after at least 5 days later ( i . e . a total of 7 days post-amputation ) , by quantifying ‘normal’ regeneration ( i . e . head structures regenerating at the anterior blastema , and a tail from the posterior blastema ) or abnormal phenotypes . The assay was robust and no defects in regenerative polarity were caused by surgery alone in the absence of drug exposure ( >1 , 000 fragments ) . Serendipitously , we discovered that exposure to praziquantel ( PZQ ) in these assays invariably produced two-headed ( ‘bipolar’ ) worms from regenerating trunk-fragments ( Figure 1B , 87±11% of fragments were bipolar at 70 µM PZQ , n = 5 trials , 285 worms ) . Bipolarity was first evident after ∼3 to 3 . 5 days when developing posterior eyespots became apparent and duplicated pharyngeal structures were observed by 5–6 days ( Figure 1C ) . Bipolar worms were viable , able to move , feed and reproduce asexually by splitting ( Video S1 ) . This effect of PZQ on regenerative polarity was dose-dependent ( EC50 = 35±7 µM , Figure 1D ) and structure-activity studies confirmed that the ability to evoke bipolarity was retained , albeit with lower potency , by an isoquinoline derivative with high structural similarity to PZQ ( Figure S1 ) . However , if PZQ was removed prior to cutting ( i . e . samples were washed prior to trunk fragment isolation ) , bipolarity was not observed . Maximal bipolarity was evoked by exposure to PZQ for only 18 hrs after cutting ( Figure 1E ) . Shifting the time window of PZQ exposure to start 6 , 12 or 18 hrs after cutting markedly decreased the anteriorization phenotype ( ∼50% for 6 hrs , >90% for longer delays , Figure 1E ) , suggesting that PZQ impacted an early regenerative event . Finally , PZQ also anteriorized regeneration from varied types of fragments cut from varied locations in both asexual and sexualized planarians ( Figure S2 ) . There is no prior example of a drug-evoked bipolarity with such robust effectiveness [20] , [26]–[31] . How complete was the axis repatterning evoked by PZQ ? In situ hybridization of tissue-specific mRNAs in PZQ-exposed flatworms demonstrated that PZQ evoked a complete AP axis anteriorization , manifest as a duplication of internal structures along the entire AP axis . CNS markers ( prohormone convertase-2 [PC2] , innexin-3 [Inx3] ) , a gut marker ( innexin-7 ) , a head edge marker ( polycystin-2 [Pkd2] ) , an optic nerve marker ( opsin ) and a pharynx marker ( myosin ) all revealed AP axis duplication in PZQ-treated samples , whereas a tail enriched marker ( Hox9 ) was lost within 1 day ( Figure 2 ) . Finally , the early brain marker ndk , normally localized in the predicted brain region within the anterior blastema , was resolved in the posterior blastema in PZQ-treated samples within 18 hrs post-amputation . These data underscore that PZQ acted early in the regenerative process to dysregulate expression of the earliest known polarity markers ( Figure 1E & 2 ) and that external application of a drug to a living organism induced differentiation of a second set of integrated organs/organ systems , including a dual functional CNS . Praziquantel serves as the mainstay pharmacotherapy for schistosomiasis and other cestode infections [2] . Although the relevant in vivo molecular target ( s ) of praziquantel remain undefined [5]–[11] , in vitro evidence from heterologous expression systems demonstrates that PZQ acts acutely to potentiate voltage-operated Ca2+ entry [10] , [11] . Four pieces of data suggest that PZQ also acts as an activator of Ca2+ influx in planarians . First , increased media Ca2+ concentrations ( [Ca2+]out ) potentiated the efficacy of submaximal concentrations of PZQ at promoting anteriorization ( Figure 3A ) , demonstrating the bipolar phenotype produced by PZQ is facilitated by an increasing gradient for Ca2+ entry . Second , net 45Ca2+ accumulation was higher in regenerating trunk fragments incubated with PZQ compared to regenerating controls ( 132±13% of controls after 24 hrs , n = 3 ) . Third , although significant increases in [K+]out proved toxic , increased [K+]out potentiated the ability of submaximal PZQ ( 25 µM ) to produce bipolarity when coincubated with 30 mM K+ gluconate ( Figure 3B ) . The maximal extent ( ∼4 . 5-fold ) of potentiation occurred over an 18 hour period post-cutting ( Figure 3B ) . Finally , nicarpidine , an l-type voltage-operated Ca2+ channel antagonist with proven efficacy against voltage-operated Ca2+ channels ( VOCCs ) in flatworms [32] , attenuated the ability of PZQ to evoke bipolarity in co-incubation experiments . Bipolarity evoked by 50 µM PZQ decreased by 79±15% in the presence of 5 µM nicarpidine ( Figure 3B , n = 3 ) . However , it was also noted from these assays that manipulations that antagonized Ca2+ entry – notably , chelation of media Ca2+ with EGTA ( Figure 3A&C ) or addition of higher concentrations of nicarpidine ( >40 µM ) in the absence of PZQ ( Figure 3C ) also produced a small proportion of two-headed flatworms . The penetrance of these phenotypes was much lower ( typically ∼10% for pharmacological VOCC inhibition , Figure 3C ) than that observed with maximal concentrations of PZQ ( ∼100% peak effect ) . The observation that biphasic modulation of Ca2+ entry produced bipolarity likely implicates a role for a macroscopic ( anterior-posterior ) Ca2+ gradient in the readout of positional information during regeneration . Such a Ca2+ gradient could be flattened by either activating ( PZQ ) or inhibiting Ca2+ entry ( see Discussion ) , but is dampened more effectively by activation ( PZQ , ≤100% penetrance ) rather than inhibition of a subset of Ca2+ entry pathways ( ∼10% penetrance ) . On the basis of these data , we evaluated a broader panel of pharmacological agents known to act as modulators of Ca2+ entry and downstream Ca2+-dependent effectors in the regenerative assay ( Figure 3C , Table S1 ) . Unsurprisingly , the majority of drugs resulted in no regenerative polarity defect: this negative cohort encompasses ‘true’ negatives as well as ‘false’ negatives ( drugs that fail to accumulate , or those that lack affinity for invertebrate channels/transporters ) . However , consistent with the previous data ( Figure 3A&B ) , other depolarizing agents ( ivermectin , [33] ) , or agents that indirectly activate VOCCs ( donepezil , imidacloprid , [34] ) also evoked bipolarity in a significant proportion of worms ( 20–60% , Figure 3C ) . Drugs with low observed incidences of bipolarity ( ≤10% ) comprised inhibitors of molecules involved in Ca2+ signaling . In summary , these data show again that either activation ( PZQ ) or inhibition of Ca2+ signaling can miscue regenerative polarity , albeit with different penetrance . On the basis of these results implicating Ca2+ homeostasis ( Figure 3 ) , we proceeded to investigate the role of voltage-operated Ca2+ channels ( VOCCs ) , and notably Cavβ subunits in PZQ efficacy [10] , [11] , using a chemical genetic in vivo RNAi approach [17] . Degenerate PCR revealed the presence of two VOCC β subunits in D . japonica , from which we proceeded to clone full length sequences ( Cavβ1 , 551 amino acids , ∼62 kDa; Cavβ2 , 652 amino acids , ∼74 kDa; GENBANK Accession Numbers FJ483939/40 ) . Both these Cavβ subunits displayed conservation of key β core domains ( SH3 , HOOK , guanylate kinase ) , regulatory motifs and residues crucial for α subunit interaction defined from vertebrate Cavβ crystal structures ( Figure S3 , [18] , [19] , [35] ) and the subunits exhibited ∼47% ( Cavβ1 ) and ∼33% ( Cavβ2 ) overall identity to human CACNB1/2 ( Table S2 ) . D . japonica Cavβ1 displayed ∼57% sequence identity to Schistosoma mansoni ( Sm ) Cavβ , while sequence identity was lower between Cavβ2 and Sm Cavβvar ( 34% , Table S2 ) . In situ hybridization revealed widespread distribution of β subunit mRNA in mesenchyme , brain ( Cavβ1 & Cavβ2 ) and pharyngeal muscle ( Cavβ1>Cavβ2 , Figure S3C ) . Similarly , RT . PCR screening for individual Cavβ subunits in different cut sections of worms ( head , trunk and tail ) confirmed mRNA for both Cavβ subunits was present in trunk fragments used in the regenerative assay ( Figure S3D ) . To investigate the in vivo role of each Cavβ subunit , worms were fed bacteria expressing dsRNA against Cavβ1 or Cavβ2 , as well as constructs serving as positive ( six-1 , a transcription factor essential for to eye regeneration [24]; PC2 , a enzyme needed for photoaversion [36] ) and negative RNAi controls ( Xen , see Materials and Methods ) over multiple feeding/regenerative cycles ( Figure 4A ) . After the second regenerative cycle , phenotypic effects were recorded ( Figure 4B&C ) and compared to results from real time PCR analysis from the same cohort of worms to quantify the effectiveness and specificity of in vivo RNAi ( Figure 4D ) . Finally , a third regenerative cycle was performed during which half the worms were exposed to PZQ ( with the other half remaining untreated ) for the purpose of assaying whether knockdown of specific mRNAs impacted the ability of PZQ to evoke bipolarity . In total , from first feeding to final phenotypic scoring , each independent assay was ∼1 month in duration . In this assay , we observed no phenotypic effect in worms fed the negative control ( Xen ) construct ( n = 4 independent cycles , as per Figure 4A ) . As expected , worms fed the positive control ( six-1 ) construct failed to develop eyespots after regeneration ( ∼90% of worms lacked eyespots after the second regenerative cycle , Figure 4B ) . Similarly , after the second regenerative cycle , worms subject to PC2 RNAi display impaired mobility ( Figure 4B , Figure S4 ) . In vivo RNAi of Cavβ1 and Cavβ2 subunits yielded distinct phenotypic outcomes . Knockdown of Cavβ1 disrupted worm motility and feeding , leading to first ‘corkscrewing’ ( Video S2 ) and then ‘curled’ immobilized worms ( Figure 4B , Video S3 ) . These defects in motility , observed in Cavβ1 and PC2 RNAi worms , did not however prevent regeneration . Knockdown of Cavβ2 did not produce any motility defect ( Video S4 ) . The most apparent phenotype in Cavβ2 RNAi worms was a rounded ( as opposed to normally ‘arrowed’ ) head morphology ( Figure 4B ) . Finally , in double VOCC β subunit knockdown worms ( Cavβ1&2 ) , both the mobility and morphological phenotypes were apparent . Importantly , in vivo RNAi of VOCC β subunits modulated AP polarity . Knockdown of Cavβ1 , either alone or in combination with Cavβ2 , yielded a small proportion of bipolar worms in the absence of PZQ exposure ( 4 . 8±2 . 8% in Cavβ1; 8 . 0±4 . 0% in Cavβ1 and Cavβ2 , n = 4; Figure 4C ) . The magnitude of this effect was similar to the small proportion of bipolar worms resulting from pharmacological blockage of VOCCs ( ∼10% , Figure 3C ) , confirming that either genetic or pharmacological inhibition of voltage-operated Ca2+ entry miscued regenerative polarity , albeit with a low peak effect by either approach . To determine the specificity and effectiveness of in vivo RNAi , we performed real time PCR analysis ( Figure 4D ) . Quantification by real-time PCR revealed effective knockdown of the targeted mRNAs , ranging from a decrease of ∼40% ( six-1 ) to ∼90% ( Cavβ2 ) in single knockdowns . The double knockdown ( Cavβ1 & Cavβ2 ) was also effective , although the magnitude of knockdown of individual β subunits was lower compared to the levels achieved in worms fed either construct alone . Finally , knockdown was selective: there was no significant change in β subunit mRNA in worms fed a control construct ( six-1 ) and despite the identity ( ∼37% ) between the two VOCC β subunits , Cavβ1/2 mRNA levels changed <15% in the reciprocal ( Cavβ2/1 ) single β subunit knockdown ( Figure 4D ) . To test the effectiveness of PZQ in evoking bipolarity in the different RNAi cohorts following manipulation of β subunit levels , trunk fragments were cut and allowed to regenerate either in the presence or absence of PZQ . In both the negative and positive RNAi controls , PZQ ( 70 µM , 24 hrs ) induced two-headed ( Xen ) or two-headed , no-eyed worms ( six-1 ) worms with similar effectiveness to that observed in the naïve ‘no-RNAi’ cohort ( Figure 5A ) . No bipolarity was observed in six-1 or Xen worms ( >100 trunk fragments ) in the absence of PZQ exposure ( Figure 5B ) . Similarly , immobilization of worms via PC2 RNAi failed to prevent PZQ-evoked bipolarity , and genetic bipolarity was not observed in this cohort in the absence of PZQ exposure ( Figure 5A&B ) . In contrast , Cavβ1 knockdown ( alone or in combination with Cavβ2 ) antagonized the ability of PZQ to produce bipolar worms . Only a small percentage of bipolar worms were evoked by PZQ in Cavβ1 ( 10 . 1±3 . 5% ) and double ( Cavβ1 and Cavβ2 , 3 . 9±1 . 4% ) knockdown worms , and this residual number was similar to the percentage of ‘genetic’ two-heads ( Figure 4C ) . These results are consistent with earlier pharmacological and physiological evidence ( Figure 2 ) that PZQ stimulates Ca2+ entry through VOCCs , as this activity was ablated by knockdown of VOCC subunits . Knockdown of Cavβ2 also attenuated PZQ evoked-bipolarity ( Figure 5B ) but the extent of the inhibition was less pronounced than with Cavβ1 ( 36 . 7±3 . 8% vs 10 . 1±3 . 5% bipolarity for Cavβ2 or Cavβ1 knockdown , respectively ) , despite a near complete loss of Cavβ2 mRNA ( Figure 4D ) . These data suggest that Cavβ2 is a less effective mediator of PZQ-evoked bipolarity , although we cannot exclude the possibility that Cavβ2 is irrelevant to PZQ-evoked bipolarity and the partial attenuation simply results from the small decrease in Cavβ1 levels ( ∼12% ) seen in Cavβ2 worms ( Figure 4D ) . Finally , to assess whether PZQ-evoked bipolarity could be prevented by knockdown of other Ca2+ channels subunits , we developed a further RNAi construct ( see Methods ) targeting one of the several pore-forming VOCC α subunits expressed in D . japonica ( Cav1 . 1 , Zhang et al . unpublished data ) . In worms treated with this constructs , PZQ was still effective at evoking bipolarity ( Figure 5B&C ) . Therefore , the ability of VOCC β subunit ablation to attenuate PZQ-evoked bipolarity was specific to the manipulation of certain VOCC subunits , and notably Cavβ1 . Conservatively , these in vivo RNAi data collectively demonstrate that knockdown of VOCC β subunits attenuated PZQ-evoked bipolarity . Finally , we tested whether in vivo RNAi of Cavβ subunits provided resistance to PZQ-evoked lethality in intact worms . First , we performed toxicity testing to evaluate the concentration range of maintained exposure to PZQ that resulted in lethality in D . japonica . Figure 6 shows that intact worms , subjected to multiple feeding cycles with the negative ( ‘Xen’ ) RNAi control , began to die within a few days when continually incubated in 100 µM PZQ ( LD50 = 7 . 0±2 . 4 days , n = 4 ) . In parallel assays , the double Cavβ1 and Cavβ2 RNAi cohort exhibited heightened resistance to PZQ exposure , surviving for almost twice as long ( LD50 = 13 . 3±2 . 3 days , n = 3 ) . Analysis of survival curves in single knockdown cohorts ( i . e . Cavβ1 or Cavβ2 ) , resolved lethality over a timeframe between the control and double knockdown ( LD50 = 12 . 7±1 . 1 for Cavβ1 , 11 . 7±3 . 1 for Cavβ2 respectively , n = 3 ) . RT . PCR analysis of intact worms fed individual Cavβ constructs confirmed knockdown was selective in the cohort of worms used for this survival time assay ( Fig . 6B ) . Therefore , these data demonstrate that manipulation of Cavβ levels also afforded protection against PZQ toxicity in intact worms .
Here , we report a novel biological activity of the antischistosomal praziquantel by illustrating an unexpected activity of this drug , and close structural mimetics , to anteriorize regeneration of fragments cut from a species of free-living flatworm . By chemical genetic analysis we show that PZQ-evoked bipolarity , as well as PZQ-evoked toxicity , was attenuated by in vivo RNAi of VOCC β subunits . While such data do not identify VOCC β subunits as the direct target of PZQ –only as gene products epistatic to PZQ action – when considered in conjunction with previous whole cell current analysis of VOCC properties in the presence of heterologously overexpressed schistosome β subunits [10] , [11] , [37] , they add further , and crucially , in vivo genetic support for the ‘Ca2+ channel hypothesis’ of PZQ action . Therefore , we believe these results bring a fresh , albeit unorthodox , perspective to the problem of defining molecules crucial for PZQ efficacy in vivo . By taking a phylogenetic side-step to a different class of platyhelminths , relevance to schistosomal physiology is certainly an extrapolation: however , we note that Cavβ subunits are well conserved across between different species ( trematodes , cestodes as well as turbellarians ) that show sensitivity to PZQ [37] . A couple of differences between these data and the previous heterologous expression experiments in Xenopus oocytes deserve comment [10] , [11] . Prior molecular evidence demonstrated that PZQ acted to potentiate Ca2+ entry through heterologously expressed non-flatworm VOCCs ( <2-fold change in peak current ) contingent on overexpression of a ‘variant’ ( Cavβvar , [10] , [11] , [37] ) schistosome VOCC β subunit . This ‘variant’ β subunit lacks two consensus PKC phosphorylation sites shown by site-directed mutagenesis as being crucial for conferring PZQ sensitivity , such that when consensus phosphorylation motifs are reintroduced , PZQ sensitivity is lost [10] , [11] , [37] . Although D . japonica Cavβ2 also lacks the consensus PKC phosphorylation sites defined as being critical for conferring PZQ sensitivity in heterologous systems ( Figure S3 ) , in our experiments PZQ efficacy was more critically dependent on Cavβ1 levels ( where the consensus PKC phosphorylation sites are present ) . A second difference relates to timescale: the heterologous current recordings reveal an acute action of PZQ ( secs ) , while our in vivo data reveal an effect that is manifest chronically ( Figure 1E , 18 hrs for peak exposure effect ) . It is interesting to note that PZQ , originally developed in a synthesis of novel anti-anxiety compounds , shares some structural similarity to benzodiazepines , several of which upregulate the expression of specific VOCC subtypes in neurons [38] . Therefore , both acute ( potentiation of existing currents ) as well as chronic effects of PZQ ( upregulation of VOCC expression ) merit further examination . A second area of significance of these studies relates to regenerative patterning . As a remarkably penetrant effect on the polarity of regeneration in D . japonica , these data provide novel impetus to define the epigenetic role played by Ca2+ signals in regulating in vivo stem cell differentiation and regenerative specification , expanding the versatility of differently sourced Ca2+ signals in regulating the patterning of different body axes in different organisms [39]–[41] . Particularly curious is the observation from both the RNAi and pharmacological screens demonstrating that AP polarity is miscued by both activation and inhibition of VOCCs , albeit far more effectively by activating ( PZQ ) than inhibiting Ca2+ influx . One speculative model for AP fate consistent with the experimental data is shown in Figure 7 . It has been demonstrated recently that blastema polarity within the excised trunk fragment is determined by β-catenin-1 [25] , [42] , such that tail specification occurs above a critical local ( nuclear ) threshold of β-catenin-1 . If Ca2+ regulates both the gradient of positional identity and is antagonistic to β-catenin-1 stability [43] , then changes in Ca2+ influx will impact both the distribution of positional cues as well as the overall concentration of β-catenin-1 . Flattening the macroscopic Ca2+ gradient by inhibiting Ca2+ influx through VOCCs flattens the gradient of positional identity , leading to less robust posterior fate decisions ( ∼10% misspecification ) . Praziquantel , by activating Ca2+ influx through VOCCs , also flattens the Ca2+ gradient controlling positional identity , but crucially also decreases β-catenin-1 levels thereby leading to consistent anteriorization outcomes . The model is consistent with the antagonistic role of Ca2+ signals on canonical Wnt signaling established for patterning events in other systems [43] . However , direct validation of this model would require Ca2+ imaging experiments , which would be facilitated by future optimization of transgenic methods in this system to enable global expression of genetically-coded Ca2+ indicators competent to resolve changes in resting free Ca2+ concentration and Ca2+ influx during regeneration . Indeed , planaria harbor a broad diversity of neurotransmitters that could regionally stimulate VOCCs during regeneration [44] . Finally , as a safe , clinically approved drug , PZQ treatment provides a cheap and facile method for generating bipolar organisms for laboratory experiments and teaching demonstrations . Planarians are commonly used model organisms in the classroom to showcase the phenomenon of regeneration [45] . Useful for studies spanning molecular regeneration of the CNS [46] to whole animal behavioral responses [47] , the ability to induce dual , integrated central nervous systems by exogenous drug application is a striking visual outcome .
|
Praziquantel is the major drug used to treat people infected with parasitic worms that cause the neglected tropical disease schistosomiasis . Despite being in widespread clinical use , it is surprising that scientists have not identified how praziquantel works to kill pathogenic schistosomes . This lack of pathobiological insight is a major roadblock to the directed design of new drugs to treat schistosomiasis , as the relevant in vivo target molecule/pathway of praziquantel remains undefined . In this report , we have discovered a new biological activity of praziquantel that enables us to bring a unique chemical genetic perspective to the problem of identifying molecules needed for in vivo praziquantel efficacy . Specifically , we show that praziquantel miscues regenerative patterning in a species of free-living flatworm to yield bipolar ( two-headed ) organisms . By using this phenotype to screen for molecules underpinning this activity , we provide in vivo support for the ‘Ca2+ channel hypothesis’ of PZQ efficacy , and show that manipulation of specific subunits of voltage-gated Ca2+ channels prevent this effect , and lessen praziquantel-mediated toxicity . These data provide further impetus to studying the role of these proteins in schistosome pharmacotherapy .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"developmental",
"biology/stem",
"cells",
"cell",
"biology/cell",
"signaling",
"physiology/pattern",
"formation",
"developmental",
"biology/pattern",
"formation",
"physiology/neurodevelopment",
"physiology/cell",
"signaling",
"pharmacology"
] |
2009
|
A Novel Biological Activity of Praziquantel Requiring Voltage-Operated Ca2+ Channel β Subunits: Subversion of Flatworm Regenerative Polarity
|
Recent evidence suggests that many malignancies , including breast cancer , are driven by a cellular subcomponent that displays stem cell-like properties . The protein phosphatase and tensin homolog ( PTEN ) is inactivated in a wide range of human cancers , an alteration that is associated with a poor prognosis . Because PTEN has been reported to play a role in the maintenance of embryonic and tissue-specific stem cells , we investigated the role of the PTEN/Akt pathway in the regulation of normal and malignant mammary stem/progenitor cell populations . We demonstrate that activation of this pathway , via PTEN knockdown , enriches for normal and malignant human mammary stem/progenitor cells in vitro and in vivo . Knockdown of PTEN in normal human mammary epithelial cells enriches for the stem/progenitor cell compartment , generating atypical hyperplastic lesions in humanized NOD/SCID mice . Akt-driven stem/progenitor cell enrichment is mediated by activation of the Wnt/β-catenin pathway through the phosphorylation of GSK3-β . In contrast to chemotherapy , the Akt inhibitor perifosine is able to target the tumorigenic cell population in breast tumor xenografts . These studies demonstrate an important role for the PTEN/PI3-K/Akt/β-catenin pathway in the regulation of normal and malignant stem/progenitor cell populations and suggest that agents that inhibit this pathway are able to effectively target tumorigenic breast cancer cells .
There is increasing evidence that a variety of cancers , including those of the breast , may be driven by a component of tumor-initiating cells that retain stem cell-like properties . These properties include self-renewal , which drives carcinogenesis , as well as differentiation , which contributes to tumor cellular heterogeneity [1] . A number of signaling pathways have been found to play a role in mammary stem cell self-renewal , including Wnt , Notch , and Hedgehog [2]–[4] . In addition , the PTEN ( phosphatase and tensin homolog deleted on chromosome 10 ) tumor suppressor gene , one of the most frequently mutated genes in human malignancies , has also been suggested to play a role in stem cell self-renewal [5] . PTEN acts as a lipid phosphatase to dephosphorylate phosphatidylinositol ( 3-5 ) -trisphosphate ( PIP3 ) , antagonizing the PI3-K/Akt pathway . Deletion of PTEN results in increased activation of the PI3-K/Akt pathway , which correlates with poor prognosis in breast cancer patients [6] . Furthermore , deletion or reduced expression of PTEN in a wide variety of human tumors is associated with resistance to conventional therapeutic agents and relapse following initial treatment [7] , [8] . In prostate tumors , loss of PTEN expression predicts progression to invasive and metastatic disease [9] . Deletion of PTEN in murine models of prostate cancer results in expansion of the prostate stem/progenitor cell population and initiation of prostate tumors resembling those in humans [10] . In the hematopoietic system , PTEN deletion induces excessive proliferation of hematopoietic stem cells with subsequent depletion of this cell population in the bone marrow [11] , [12] . This PTEN deficiency also results in the induction of myeloproliferative disorders that progress to leukemia [11] , [12] . Recent studies have suggested that cancer stem cells , by virtue of their resistance to chemotherapy and radiation therapy , may contribute to tumor resistance and relapse [13] , [14] . The PTEN/PI3-K/Akt pathway has been described as a major pathway conferring resistance to conventional therapies in multiple tumor types [15]–[17] . Using a large-scale RNA interference genetic screen , Berns et al . identified PTEN as the modulator of drug resistance in breast cancer [17] . Patients with HER2 amplified breast tumors that also contain PTEN deletions are resistant to Trastuzumab treatment [8] . Because cancer stem cells have been found to be resistant to radiation and chemotherapy , we postulated that the PTEN/Akt pathway may play a role in the regulation of mammary stem/progenitor cells . Thus , we examined the PI3-K/Akt pathway and characterized its downstream signaling components for their role in regulating mammary stem/progenitor cells . In the present study , we demonstrate that the PI3-K/Akt pathway plays an important role in regulating the Aldefluor-positive cell population , which is enriched in mammary stem/progenitor cells , by mediating Wnt/β-catenin signaling through phosphorylation of GSK3-β . Furthermore , we demonstrate that the Akt inhibitor perifosine is able to target normal and malignant Aldefluor-positive mammary epithelial cells in vitro and in mouse xenograft models .
We have previously demonstrated that primitive mammary stem/progenitor cells are enriched in vitro in floating spherical colonies termed mammospheres . Mammospheres are composed of a small number of cells with stem cell-like properties including the ability to form secondary mammospheres as well as the ability to undergo multilineage differentiation [18] . In addition , we recently reported the enrichment of mammary stem/progenitor cells within the aldehyde dehydrogenase ( ALDH ) -expressing cell population as assessed by the Aldefluor assay [19] . When these primitive mammary cells are cultured in the presence of serum on an adhesive substratum , they lose these primitive properties and undergo differentiation . Recent studies suggest that signal transduction pathways including the PTEN/PI3-K/Akt play a role in embryonic and tissue-specific stem cell self-renewal [10] , [20] , [21] . To determine whether this pathway was activated in primitive mammary cells , we compared the levels of PTEN and Akt phosphorylation and its downstream targets in normal mammary stem and progenitor cells in mammospheres compared with those in cells induced to undergo differentiation in monolayer cultures . Activation of the PTEN/PI3-K/Akt pathway was assessed by Western blotting using phospho-specific antibodies . As compared to adherent cultures , normal mammary epithelial cells ( NMECs ) in mammosphere cultures expressed increased Ser380 phosphorylation of PTEN ( Figure 1A ) , which results in its conformational changes masking the PDZ binding domain [22] . PTEN , through its lipid phosphatase activity , antagonizes PI3-K/Akt signaling . We detected increased Akt Ser473 phosphorylation in mammospheres as compared with monolayer cultures , suggesting that inactivation of PTEN results in increased Akt phosphorylation in more primitive cells ( Figure 1A ) . Akt has a number of known downstream targets including GSK3-β , which regulates the Wnt/β-catenin pathway . As compared to differentiated cells , cells in mammospheres displayed increased levels of GSK3-β phosphorylation and β-catenin activation ( Figure 1A ) . β-catenin has been demonstrated to play an important role in the development of mammary stem cells in mouse models [23] , suggesting that this pathway may also be active in human mammary stem/progenitor cells in mammospheres . We have previously reported an enrichment of mammary stem/progenitor cells within the cell population expressing of ALDH , which can be detected by the enzymatic Aldefluor assay [19] , [24] . Consistent with this , we found significantly higher ALDH expression in mammospheres as compared to attached mammary epithelial cell cultures ( Figure 1A ) . We also analyzed activation of the Wnt/β-catenin pathway in ALDH-expressing cells . Aldefluor-positive cells showed significantly higher β-catenin activation as well as increased GSK3-β phosphorylation as compared with Aldefluor-negative cells ( Figure S1 ) , suggesting that mammary stem/progenitor cells display activation of the Wnt/β-catenin pathway . To examine the functional role of the PI3-K/Akt/GSK3-β/β-catenin pathway , we used both gain-of-function and loss-of-function strategies . To activate this pathway , PTEN levels were decreased by using a PTEN shRNA DsRed-labeled lentivirus . Inclusion of the DsRed label allowed us to eliminate noninfected cells by flow cytometry . As shown in Figure 1B , we achieved greater than an 80% reduction in PTEN protein expression in mammospheres as assessed by Western blotting . Knockdown of PTEN in these cells resulted in increased levels of Akt phosphorylation , GSK3-β phosphorylation , and β-catenin activation ( Figure 1B ) . In addition , activation of this pathway further increased expression of ALDH1 ( Figure 1B ) . To examine the functional role of the PTEN/PI3-K/Akt pathway in human mammary stem/progenitor cell fate , we measured the effect of PTEN knockdown on mammosphere formation . Knockdown of PTEN resulted in an increase in the number of primary and secondary mammospheres as compared to control , p<0 . 01 ( Figure 1C and 1D ) . Furthermore , this increase was maintained upon serial passage to tertiary mammospheres ( Figure 1D ) . To provide additional evidence that this pathway enriches for mammary stem/progenitor cells , we determined the effect of PTEN knockdown on the percentage of cells expressing ALDH as assessed by the Aldefluor assay . Primary NMECs from reduction mammoplasties contain between 4% and 9% Aldefluor-positive cells , which increases to 14%–19% in primary mammospheres , consistent with the enrichment of stem/progenitor cells when grown in suspension cultures ( Figure S2A ) . PTEN knockdown increased the proportion of Aldefluor-positive cells ( p<0 . 01 ) in mammospheres more than 2-fold to 37%–41% ( Figure 1E ) . Thus , knockdown of PTEN resulted in enrichment of mammary stem/progenitor cells in vitro as determined by both mammosphere and Aldefluor assays . We previously observed that while mammary stem/progenitor cells are enriched in mammosphere cultures , they are depleted in attachment cultures ( Figure S2B ) . Cells with PTEN knockdown maintained a higher percentage of Aldelfuor-positive cells in attachment culture ( Figure S2B ) as well as in suspension culture ( Figure S2A ) . Furthermore , PTEN knockdown increased phospho-Akt expression in cells grown either in attachment or suspension cultures ( Figure S2C ) . These results suggest that PTEN knockdown is able to enrich for the stem/progenitor cell population independent of culture conditions . We previously used a mouse model described by Proia and Kuperwasser [25] in which NMECs form outgrowths in NOD/SCID mice whose mammary fat pads have been humanized by the introduction of both irradiated and non-irradiated human mammary fibroblasts . We used this system to examine the effects of PTEN knockdown on mammary development . Serial dilutions of flow cytometry-sorted cells were introduced into the humanized fat pads of NOD/SCID mice . As indicated in Table S1 , at all dilutions , NMECs with PTEN knockdown were more efficient in generating outgrowths than DsRed control cells . While at least 10 , 000 control cells were required for efficient outgrowth formation , as few as 250 PTEN knockdown cells generated outgrowths in 50% of the mice , indicating that PTEN knockdown increased the frequency of multipotent mammary stem/progenitors . In addition , we observed significant morphological alterations in structures generated by PTEN knockdown in NMECs compared to DsRed controls ( Figure 2A ) . PTEN knockdown cells produced much larger structures that displayed significant morphologic alterations . Knockdown of PTEN was confirmed by the lack of PTEN expression in PTEN knockdown outgrowths compared to controls ( Figure 2B , a and h ) . We used immunohistochemical staining for markers of myoepithelial , basal , and luminal epithelial cells to ascertain the effects of PTEN knockdown on cellular differentiation . Outgrowths generated by control-infected NMECs consisted of ductal structures that were characterized by a single layer of myoepithelial cells , which expressed smooth muscle actin recapitulating the architecture of normal mammary ducts in humans . In contrast , structures generated by PTEN shRNA-infected cells were characterized by gross disorganization with increased numbers of smooth muscle actin expressing myoepithelial cells distributed throughout the gland ( Figure 2B , b and i ) . Glands produced by control cells contained only a small number of cells expressing the primitive cytokeratins 5/6 , whereas the frequency of these cells was greatly increased in PTEN knockdown structures ( Figure 2B , c and j ) . Examination of epithelial markers also revealed significant differences between structures derived from PTEN knockdown and control cells . In control structures , the majority of the luminal epithelial cells expressed the luminal marker CK18 , whereas expression of this marker occurred only in a subfraction of PTEN knockdown cells ( Figure 2B , d and k ) . Estrogen receptor ( ER ) was expressed in luminal epithelial cells in structures generated from DsRed control cells , but not in structures derived from PTEN knockdown cells ( Figure 2B , e and l ) . Furthermore , structures with PTEN knockdown displayed significant increases in proliferating cells as determined by Ki67 expression ( Figure 2B , f and m ) . Consistent with the in vitro experiments , PTEN knockdown also increased the proportion of cells expressing ALDH1 ( Figure 2B , g and n ) . These experiments confirm and extend the in vitro findings and suggest that in addition to resulting in enrichment of the stem/progenitor cell pool , activation of the PTEN/PI3-K/Akt pathway affects cellular growth and differentiation . This results in the generation of cells displaying increased proliferation , with aberrant differentiation resulting in increased expression of primitive and basal markers and decreased expression of luminal epithelial markers . All of these histopathologic features are characteristic of atypical ductal hyperplasia , a premalignant lesion that may progress to invasive breast cancer . To further characterize the pathways regulating mammary stem/progenitor cell self-renewal , we used inhibitors of PI3-K/Akt signaling as well as its downstream targets , GSK3-β and mammalian target of rapamycin ( mTOR ) . As demonstrated in Figure 3A , treatment of NMECs with the PI3-K inhibitor LY294002 or the Akt inhibitor IV or perifosine reduced the number of both primary ( unpublished data ) and secondary mammospheres ( p<0 . 001 ) . Furthermore , primary mammospheres treated with inhibitors of PI3-K or Akt completely failed to form tertiary mammospheres ( unpublished data ) . To determine whether the mTOR pathway , which is downstream of Akt signaling , plays a role in this process , we examined the effect of rapamycin , an inhibitor of this pathway , on mammosphere formation . As shown in Figure 3A and Figure S2 , rapamycin had little effect on secondary mammosphere formation , suggesting that mTOR was not responsible for mediating the effects of Akt signaling . In contrast to rapamycin , PI3-K and Akt inhibitors suppressed the formation of secondary mammospheres in both control and PTEN knockdown NMECs , supporting the importance of this pathway in mammary stem/progenitor cell self-renewal ( Figure 3A ) . To determine whether PTEN knockdown affected cellular sensitivity to Akt inhibition , we performed dose response studies with perifosine . As shown in Figure 3B , 2 µM perifosine inhibited Akt phosphorylation by more than 50% in PTEN knockdown cells while having no demonstrable effect on control cells . Consistent with the effects on mammosphere formation , the Akt inhibitor perifosine significantly reduced the proportion of Aldefluor-positive cells in mammospheres ( Figure 3C ) . The effect of perifosine on PI3-K/Akt signaling has previously been reported [26] , [27] . In that study , the authors found that 10 µM perifosine did not have any inhibitory effect on the MAPK pathway [27] . We also failed to detect an effect of perifosine on MAPK activity in our system ( unpublished data ) . To determine whether the PI3-K/Akt pathway was also critical for mammary development in vivo , we examined the effect of the Akt inhibitor perifosine on the development of human mammary structures generated by control DsRed infected or PTEN knockdown NMECs . NOD/SCID mice implanted with control or PTEN knockdown NMECs were treated with intraperitoneal injections of perifosine ( 30 mg/kg ) 4 days a week for five weeks or a saline control . Consistent with previous experiments , control DsRed NMECs generated mammary ductal structures with normal morphology , PTEN knockdown NMECs generated ductal hyperplasias in saline-treated control mice . In contrast , administration of perifosine completely inhibited outgrowth formation by both control and PTEN knockdown NMECs ( Figure 3D ) . These results further support the in vitro experiments by demonstrating a critical role for Akt signaling in normal mammary development . Activated Akt has been demonstrated to be capable of phosphorylating and inactivating GSK3-β , leading to nuclear translocation and activation of β-catenin [28] , [29] . In addition , Akt can directly phosphorylate β-catenin , an event which further facilitates its nuclear translocation [21] . Since Wnt signaling through β-catenin has been shown to play a role in mammary development and stem cell self-renewal , we examined whether Akt effects on mammary stem/progenitor cells were mediated by this pathway . To examine the role of Wnt/β-catenin signaling , we used the GSK3-β inhibitor 6-bromoindirubin-3′-oxime ( Bio ) , which has been shown to be able to maintain pluripotency of human and mouse embryonic stem cells through activation of β-catenin signaling [30] . We first confirmed that addition of Bio to NMECs resulted in phosphorylation of GSK3-β with subsequent activation of β-catenin ( Figure 4A ) . To determine the biological relevance of this pathway , we examined the effects of Bio on mammosphere formation . As shown in Figure 4B , the addition of Bio increased mammosphere formation 2-fold . In addition , the reduction of mammosphere formation induced by perifosine was reversed by the addition of Bio ( Figure 4B ) . Reversal of inhibition occurred despite the persistent inhibition of Akt activity in the presence of both compounds ( Figure 4A ) . To confirm the importance of β-catenin signaling in stem/progenitor cell self-renewal , we knocked down β-catenin expression by using an shRNA lentivirus ( Figure S3 ) . Lentiviral shRNA-mediated down-regulation of β-catenin expression in NMECs reduced the number of mammospheres by approximately 70% ( Figure 4B ) . These experiments suggest that the effects of Akt on mammary stem/progenitor cell self-renewal are mediated through GSK3-β/β-catenin signaling . To provide further evidence for cross-talk between Akt signaling and the Wnt pathway , we used a LEF-1/TCF reporter system to monitor β-catenin transcriptional activity in PTEN knockdown or Bio-treated NMECs . NMECs were infected with a LEF-1/TCF reporter driving GFP ( TOP-GFP ) or a control reporter with mutated LEF-1/TCF binding sites ( FOP-GFP ) . Approximately 2–5% of cells in mammospheres expressed GFP ( Figure 4C ) , whereas FOP-GFP-infected NMECs did not express GFP ( Figure 4D ) . Bio treatment increased the percentage of GFP-positive cells in mammospheres more than 2-fold , confirming that GSK3-β plays a role in nuclear translocation and activation of β-catenin ( Figure 4C and 4D ) . In addition , when cells were co-infected with lentiviral PTEN shRNA and TOP-GFP , the percentage of GFP-expressing cells increased more than 2-fold ( Figure 4C ) . To confirm that Akt regulates mammary stem/progenitor cells by activating β-catenin signaling , TOP-GFP infected cells were treated with perifosine or Bio alone or in combination . Perifosine treatment reduced the proportion of GFP-positive cells by more than 50% , whereas Bio treatment increased the number of GFP-expressing cells more than 2-fold as assessed by flow cytometry ( Figure 4D ) . Bio reversed the effect of perifosine when mammospheres were treated with a combination of perifosine and Bio ( Figure 4D ) . The role of β-catenin signaling in mammary stem/progenitor cells was further investigated by examining the β-catenin activity and nuclear localization in Aldefluor-positive and -negative populations . As shown in Figure S1 , Aldefluor-positive but not Aldefluor-negative cells displayed GSK3-β phosphorylation and nuclear β-catenin localization . To extend these observations to the mouse models , we examined the expression and localization of phospho-Akt and β-catenin in mammary outgrowths generated from control and PTEN knockdown NMECs . As shown in Figure 4E , control outgrowths contained cells with low levels of phospho-Akt and membranous β-catenin staining . In contrast , outgrowths generated from PTEN shRNA lentivirus-infected cells demonstrated increased phospho-Akt expression and nuclear β-catenin localization ( Figure 4E ) . Together , these experiments demonstrate that Akt effects on mammary stem/progenitor cells are mediated by GSK3-β/β-catenin signaling . By using primary mammary carcinoma xenografts , we previously demonstrated that cancer cells with stem cell-like properties were contained within the Aldefluor-positive population [19] . Recent studies have suggested that established breast cancer cell lines also contain subpopulations with stem cell-like properties . In MCF7 and SUM159 cell lines , we previously demonstrated that only the Aldefluor-positive populations were able to form tumors capable of serial passaging in NOD/SCID mice [31] . To determine whether the PTEN/PI3-K/Akt pathway played a similar role in the regulation of malignant mammary stem/progenitor cells to that of normal mammary stem/progenitor cells , we determined the effect of knocking down PTEN expression on the cancer stem/progenitor cell populations in these cell lines . As shown by Western blotting ( Figure 5A ) , control MCF7 cells expressed PTEN , but not phospho-Akt . Knockdown of PTEN with an shRNA lentivirus resulted in Akt activation ( Figure 5A ) . SUM159 cells display a baseline level of phospho-Akt expression , which was further enhanced by PTEN knockdown ( Figure 5A ) . To determine whether reduction of PTEN expression affected the mammary cancer stem/progenitor cell components of these cell lines , we used tumorsphere and Aldefluor assays . As shown in Figure 5B , knockdown of PTEN increased tumorsphere formation in both MCF7 and SUM159 breast carcinoma cells . Furthermore , this knockdown resulted in more than a 2-fold increase in the Aldefluor-positive population ( p<0 . 01 ) in these cell lines ( Figure 5C ) . We confirmed that PTEN knockdown in MCF7 xenographs resulted in increased Akt phosphorylation ( Figure S4B ) . Subsequently we determined whether the increase in the cancer stem/progenitor cell population resulting from PTEN knockdown resulted in increased tumorigenicity . As shown in Figure S4A and S4C , PTEN knockdown in MCF7 or SUM159 cells increased their tumorigenicity in NOD/SCID mice . Since Zhou et al . previously reported that the side population ( SP ) in MCF7 contained tumor-initiating cells , we examined the overlap between the SP and Aldefluor-positive populations . As demonstrated in Figure S4 , we found a 2-fold enrichment of Aldefluor-positive population in the MCF7 SP population as compared to non-SP cells . This suggests that the Aldefluor and SP assays detect distinct , but partially overlapping , cell populations . We previously reported that tumorigenicity in these cell lines is mediated by the Aldefluor-positive population , which is enriched for tumorigenic cancer stem/progenitor cells [31] . Thus , these results suggest that Akt activation increases tumorigenicity through effects on the cancer stem/progenitor cell population . We next used the TOP-GFP reporter system to determine whether Akt signals through the Wnt/β-catenin pathway in breast carcinoma cells . Perifosine treatment of TOP-GFP-infected SUM159 breast cancer cells resulted in a significant decrease in GFP-positive cells ( p<0 . 01 ) , whereas treatment of these cells with Bio increased the proportion of GFP-positive cells 3-fold , p<0 . 01 ( Figure 5D ) . As was the case for normal mammary cells , Bio treatment was able to reverse the effect of perifosine in these mammary carcinoma cells ( Figure 5D ) . To determine whether Akt also drives the tumorigenic Aldefluor-positive cell population in vivo , we examined the effects of the Akt inhibitor perifosine on the growth of SUM159 breast cancer cells in NOD/SCID mouse xenografts . SUM159 cells are more tumorigenic than MCF7 cells , which may relate to their higher level of Akt phosphorylation . In addition , to extend this study to primary tumor xenografts , we examined the effects of this inhibitor on two different breast cancer xenografts , MC1 and UM2 , in which we previously demonstrated that the tumorigenic cells are contained within the Aldefluor-positive population [19] . We compared the effects of perifosine alone or with docetaxel , a chemotherapeutic agent that is commonly used to treat human breast cancers . As shown in Figure 6 , both perifosine and docetaxel significantly inhibited tumor growth in all three xenografts compared with saline-treated controls . Furthermore , the combination of perifosine and docetaxel reduced tumor size compared with either treatment alone ( Figure 6A–6C ) . After 4 wk of treatment , animals were killed , and cells in the residual tumors were analyzed by the Aldefluor assay . Treatment of mice with docetaxel had no significant effect on the percentage of Aldefluor-positive cells . In contrast , perifosine either with or without docetaxel significantly ( p = 0 . 001 ) reduced the Aldefluor-positive population by over 75% in Sum159 and MCI ( Figure 6A and 6B ) and over 90% in UM2 xenografts ( Figure 6C ) . ( Flow cytometry data are presented in Figure S6A–S6C ) . To determine whether these treatments affected the Wnt/β-catenin pathway , we examined their effects on β-catenin expression and cellular localization . Nuclear β-catenin was detected in the majority of cells from control and docetaxel-treated animals . In contrast , tumors treated with perifosine or the combination of perifosine and docetaxel showed a significant reduction in β-catenin expression that was largely localized in the cytoplasm or plasma membrane rather than nucleus ( Figure 6D ) . Although cancer stem cells are enriched within the Aldefluor-positive cell population , a more definitive assay for breast cancer stem cells involves their ability to self-renew as determined by tumorigenicity in NOD/SCID mice . We therefore determined the ability of serial dilutions of cells obtained from control or treated tumors to form secondary tumors in NOD/SCID mice . Figure 7A and 7B demonstrate that tumor cells derived from docetaxel-treated or control animals showed similar tumor regrowth at all dilutions in secondary NOD/SCID mice . In contrast , perifosine treatment with or without chemotherapy significantly reduced tumor growth in secondary recipients ( Figure 7A and 7B ) . When equal numbers of cells were injected , those from perifosine treated animals showed a 2–3-fold reduction ( For 5×104 cells; p<0 . 01 , for 1×104 cells; p = 0 . 05 , and for 1×103 cells; p = 0 . 01 ) in tumor growth compared to cells from chemotherapy treatment alone or control animals ( Figure 7A and 7B and Figure S6 ) . Furthermore , in contrast to cells obtained from docetaxel treatment or control animals , 1 , 000 cells from perifosine or the combination of perifosine- and docetaxel-treated animals failed to produce tumors in secondary mouse xenografts ( Figure 7A and 7B ) . The kinetics of secondary tumor growth at different inoculation dilutions are shown in Figure S7A and S7B . Thus , the Aldefluor assay as well as tumor regrowth experiments indicate that perifosine is able to target and reduce the Aldefluor-positive population , whereas one of the most commonly used chemotherapeutic agents , docetaxel , failed to do so . To determine whether perifosine treatment selected for a resistant population of tumor stem/progenitor cells , cells from tumors remaining after perifosine or docetaxel treatment were sorted by the Aldefluor assay and reimplanted in secondary NOD/SCID mice . As we have previously reported [19] , only Aldefluor-positive cells were tumorigenic in NOD/SCID mice ( Figure 7C ) . Furthermore secondary tumors generated from perifosine- or docetaxel-treated tumors were as sensitive to perifosine as were primary tumors ( Figure 7C ) . Together these studies indicate that perifosine is able to significantly reduce the tumorigenic cell population without selecting for resistant cells .
The cancer stem cell hypothesis holds that cancer arises in tissue stem or progenitor cells through dysregulation of the self-renewal process [32] , [33] . This process generates tumors organized in a cellular hierarchy that are driven by “cancer stem cells , ” which are capable of self-renewal as well as differentiation , generating the bulk of the tumor . Although considerable progress has been made in identifying “cancer stem cells” in a variety of hematologic and solid human malignancies , the pathways that drive transformation of these cells are poorly understood . We and others have suggested that carcinogenesis may involve dysregulation of the normally tightly regulated process of stem cell self-renewal coupled to aberrant differentiation of progeny cells [1] . The PTEN tumor suppressor gene is one of the most frequently dysregulated genes in breast cancer . Mouse genetic studies reveal that PTEN is essential for embryonic development [5] , with heterozygous mice developing tumors in several organs including the breast [5] , [34] . Germline mutations of PTEN cause cancer predisposition and a rare developmental disease called Cowden syndrome , which is associated with an increased incidence of breast cancer [35] . Humans with germline BRCA1 mutations may also develop microdeletions in the PTEN gene [36] . There is also accumulating evidence that PTEN may play a role in stem cell self-renewal [37] . We have used both in vitro systems and mouse models to demonstrate an important role for the PTEN/PI3-K/Akt/β-catenin pathway in regulating both normal and malignant mammary stem/progenitor cells . Compared with differentiated normal mammary epithelial cells in monolayer cultures , mammospheres displayed significantly higher levels of Akt phosphorylation . This was accompanied by increased GSK3-β phosphorylation and β-catenin activation , suggesting that the β-catenin signaling pathway may have a role in maintaining mammary stem/progenitor cell population . The Wnt/β-catenin pathway has previously been shown to play a role in mammary stem cell function in transgenic mouse models [38] . To examine the role of these signaling pathways in mammary stem cell function , we used both gain-of-function and loss-of-function approaches . Knockdown of PTEN using a PTEN shRNA lentiviruses resulted in an enrichment of mammary stem/progenitor cells in vitro as evidenced by mammosphere formation and by the expression of ALDH as assessed by the Aldefluor assay [19] . Increases in percentage of Aldelfuor-positive cells and mammosphere formation were seen in PTEN knockdown cells even when they were initially cultured in attached conditions , indicating that PTEN knockdown maintains a higher proportion of mammary stem/progenitors . We have previously demonstrated an enrichment of stem/progenitor cells within the Aldefluor-positive cell population as well as in mammospheres [19] , [39] . We next examined the effects of Akt activation via PTEN knockdown on the formation of mammary outgrowths in NOD/SCID mice whose mammary glands were humanized by the introduction of human mammary fibroblasts [19] , [25] . Control structures generated from DsRed-infected NMECs were composed of CK18+ epithelial cells , a portion of which also expressed ERα , surrounded by a single layer of smooth muscle actin expressing myoepithelial cells . These structures closely resembled those found in normal human mammary ducts . In contrast , PTEN shRNA lentivirus-infected cells generated hyperplastic structures exhibiting gross tissue disorganization . This was characterized by an increase in cells expressing primitive cytokeratins 5 and 6 and a decrease in cells expressing the luminal marker CK18 , indicating an expansion of primitive cells . This was further evidenced by a lack of ERα expression . Furthermore , these disorganized structures contained an increased proportion of proliferating cells , as evidenced by Ki67 expression and an expansion of ALDH1-expressing cells . All of these histologic characteristics resembled those of atypical ductal hyperplasia , one of the most common premalignant lesions in humans that is believed to be a precursor of DCIS ( ductal carcinoma in situ ) and invasive ductal carcinoma [40] . Interestingly , inactivating mutations of PTEN have been reported to be present in these preneoplastic lesions in humans [41] . Our studies provide a potential molecular explanation for these findings by suggesting that PTEN knockdown and subsequent Akt activation may regulate mammary stem cell self-renewal with an accompanying alteration in cellular differentiation , events which may be important during the initiating stages of carcinogenesis . Utilizing the SP from MCF7 cells , Zhou et al . , recently suggested that the PTEN/mTOR/STAT3 pathway is required for the maintenance of breast cancer stem cells [42] . We detected a 2-fold enrichment of Aldefluor-positive cells in the SP population , suggesting that these assays detect distinct , although partially overlapping , cell populations . We found no evidence for mTOR regulation of normal or malignant mammary stem/progenitor cells . There might be several explanations for this discrepancy . A number of reports have indicated that inhibition of mTOR results in the activation of Akt through a positive-feedback loop [43]–[46] . Furthermore , recent studies have shown that mammary stem cells are not contained within the SP population [47] . In contrast to the report by Zhou et al . , we used both pharmacologic and genetic approaches to elucidate the downstream signaling pathways of Akt in both normal and malignant mammary stem/progenitor cells . Treatment of NMECs with the PI3-K inhibitor Ly294002 , or the Akt inhibitors IV and perifosine , reduced mammosphere formation and the Aldefluor-positive cell population . The effect of perifosine on the PI3-K/Akt pathway has previously been described [26] , [48] . It has also been reported that Akt may regulate Wnt signaling through Akt phosphorylation and inactivation of GSK3-β , which in turn mediates β-catenin degradation [28] or , directly by phosphorylating β-catenin on serine 552 , promoting the nuclear translocation of β-catenin [21] . Thus , by these two mechanisms , Akt activation promotes the activation and accumulation of nuclear β-catenin [49] . We demonstrated in mammospheres that Akt phosphorylation was associated with increased phosphorylation of GSK3-β and activation of β-catenin . GSK3-β targets β-catenin for ubiquitin-mediated degradation through phosphorylation of its N-terminal serine and threonine residues [50] . To demonstrate the importance of β-catenin in mammosphere formation , we used a β-catenin shRNA lentivirus to knock down β-catenin expression . This resulted in significantly reduced mammosphere formation . Furthermore , we demonstrated that the GSK3-β inhibitor Bio rescued the effects of Akt inhibition on mammosphere formation . To examine more directly the activation of the Wnt pathway , we used a TOP-GFP reporter system that is activated by β-catenin signaling . The Akt inhibitor perifosine significantly reduced the proportion of TOP-GFP expressing cells , an effect that was reversed by Bio . The relevance of these findings to mammary development in vivo was determined by examining the expression and cellular localization of phospho-Akt and β-catenin in mammary outgrowths generated from control and PTEN knockdown NMECs . While cells from control outgrowths showed minimal phospho-Akt expression and membranous β-catenin localization , outgrowths from PTEN shRNA lentivirus-infected cells demonstrated increased phospho-Akt expression and nuclear β-catenin localization . The importance of this pathway was demonstrated by the ability of the Akt inhibitor perifosine to completely block mammary development in the mouse model . Together these in vitro and mouse experiments suggest that the effects of Akt on mammary stem/progenitor cells are mediated by GSK3-β phosphorylation and β-catenin activation . To determine whether the PTEN/Akt/β-catenin signaling pathway also plays a role in the regulation of malignant mammary stem/progenitor cells , we used both breast cancer cell lines and a primary tumor xenograft . Although Hollestelle et al . reported an activating PIK3CA mutation in MCF7 cells , they did not determine whether this mutation resulted in activation of the PI3K/Akt pathway [51] . In contrast , Neve et al . screened 38 breast cancer cell lines including MCF7 for various pathways and found that MCF7 cells displayed significantly low levels of the PI3/Akt activation when compared with other cell lines [52] . This was confirmed in a recent report demonstrating that despite harboring a PIK3CA helical mutation , MCF7 cells displayed a low level of AKT phosphorylation [53] . In addition , these authors found no correlation between similar activating mutations of PIK3CA in primary tumors and patient survival or the chemoresistance [53] . Most importantly , they found no evidence that these mutations in PIK3CA resulted in activation of Akt pathway as compared with the wild-type PIK3CA , but there was a strong correlation between PTEN inactivating mutation and activation of Akt signaling [53] . In agreement with latter reports , we found that PTEN knockdown in MCF7 or SUM159 resulted in activation of PI3-K/Akt pathway . We demonstrated that activation of this pathway through knockdown of PTEN significantly increased tumorsphere formation and the ALDH-expressing cell population , indicating an enrichment of cancer stem/progenitor cells . Moreover , enrichment of the cancer stem/progenitor cell population directly correlated with increased tumorigenicity . Since this population mediates tumorigenesis [31] , it suggests that Akt activation enhances tumorigenesis through effects on the cancer stem/progenitor cell population . Furthermore , as was the case for NMECs , we demonstrated that Akt effects on mammary carcinoma stem/progenitor cells are mediated by Wnt/β-catenin signaling . By using the TOP-GFP reporter system , we found that inhibition of Akt signaling significantly reduced the number of GFP-positive cells , an effect that was rescued with the GSK3-β inhibitor Bio . These findings are in agreement with the findings of Li et al . , who reported that Wnt-induced mouse mammary tumors show expansion of stem/progenitor population as characterized by increased stem cell markers [54] . Furthermore the Wnt/β-catenin pathway has been shown to play a role in mediating the radiation resistance of mouse mammary progenitor cells [55] . In addition to its effects on the stem/progenitor cell population , Akt has been demonstrated to play a role in chemoresistance [15]–[17] . Recent evidence utilizing in vitro systems [56] , animal models [13] , and clinical trials have suggested that breast cancer stem cells are relatively resistant to both radiation and chemotherapy [57] . Improved clinical outcomes may require the development of strategies that are able to target this cancer stem cell population . The demonstration that Akt signaling plays a prominent role in stem cell self-renewal makes it an attractive target for such strategies . We used the Akt inhibitor perifosine , an orally bioactive alkylphospholipid , to determine its effects on Aldefluor-positive cell population . We demonstrate both in in vitro and in mouse xenograft models that perifosine is able to target the Aldefluor-positive tumorigenic cell population as determined by the Aldefluor assay and by reduced tumorigenicity upon serial transplantation . Furthermore , tumorigenic Aldefluor-positive cells remaining after perifosine treatment generated secondary tumors that were still sensitive to perifosine . In contrast , the chemotherapeutic agent docetaxel , although capable of causing tumor regression , failed to target the Aldefluor-positive cell population . If cancer stem cells indeed contribute to tumor resistance and relapse , then the addition of agents that are capable of targeting these cells may increase the clinical efficacy of current therapies . Our studies identify inhibitors of the PI3-K/Akt/Wnt signaling pathway as potential agents for therapeutic targeting of cancer stem cells .
The MCF7 cell line was maintained in RPMI supplemented with 5% fetal bovine serum ( FBS ) , 5 µg/ml insulin , and antibiotic-antimycotic . The SUM159 cell line was maintained in Ham's F12 medium supplemented ( with 5% FBS , 5 µg/ml insulin , 1 µg/ml hydrocortisone , and antibiotic/antimycotic 10 , 000 units/ml penicillin G sodium , 10 , 000 µg/ml streptomycin sulfate , and 25 µg/ml amphotericin B ) . LY294002 , Akt inhibitor IV , and Bio were purchased from Millipore; perifosine was obtained from Keryx Biopharmaceuticals , and docetaxel ( Taxotere ) was from Sanofi Aventis . Antibodies to phospho-Akt , Akt , β-catenin , GSK3-β , phospho-GSK3-β Ser9 , PTEN , and phospho-PTEN Ser380 were purchased from Cell Signaling Technology . The α-tubulin antibody was from Santa Cruz Biotechnology , the ALDH1 antibody was from BD Transduction Laboratory , and the active-β-catenin ( anti-ABC ) , clone 8E7 mouse monoclonal antibody was from Millipore Corporation . The primary antibodies for human smooth muscle actin , Ki67 , CK5/6 , and CK18 were purchased from Zymed laboratories ( Invitrogen ) . Mammary tissue from reduction mammoplasties was dissociated as previously described [19] . The single-cell suspension was used for various experiments including sphere formation , flow cytometry analyses , and lentiviral infections . At least three different patient samples ( reduction mammoplasties ) were used for each experiment involving NMECs . Single NMECs were plated on 1% agarose coated plates at a density of 1×105 cells/ml and grown for 7–10 d . Subsequent cultures after dissociation of primary spheres were plated on ultra-low attachment plates at a density of 5×103–1×104 cells/ml . Mammosphere cultures were grown in a serum-free mammary epithelium basal medium as previously described [18] . Tumorspheres were cultured under identical conditions to mammospheres . All lentiviral constructs were prepared by the University of Michigan Vector or shRNA core facilities . The primers targeting the human PTEN short hairpin sequences were purchased from Integrated DNA Technologies . The forward primer TCCCAGGTGAAGGTATGTTCCTCCAATCTAAAGGATTGGAGGAATATATCTTCACCTGGGATTTTTTC and the reverse primer TCGAGAAAAAATCCCAGGTGAAGATATATTCCTCCAATCCTTTAGATTGGAGGAACATACCTTCACCTGGAC were digested with the Xho1 enzyme , annealed , and cloned into the pLentilox 3 . 7 vector . After confirmation of DNA sequences , the vectors were infected into 293 host cells to produce viruses at the University of Michigan Vector Core facility . Resulting lentiviral PTEN shRNA has been used to transfect NMECs and breast cancer cell lines . Two different CTNNB1 shRNA lentiviral vectors were purchased from University of Michigan shRNA core facility . TOP-GFP and FOP-GFP lentiviral reporter vectors were kindly provided by Irving L . Weissman ( Stanford University School of Medicine , Stanford , California ) . To measure and isolate cells with high ALDH activity , the Aldefluor assay was carried out according to manufacturer's ( Stemcell Technologies ) guidelines . Briefly , dissociated single cells from cell lines or from primary mammospheres were suspended in Aldefluor assay buffer containing an ALDH substrate , bodipy-aminoacetaldehyde ( BAAA ) at 1 . 5 µM , and incubated for 40 min at 37°C . To distinguish between ALDH-positive and -negative cells , a fraction of cells was incubated under identical condition in the presence of a 10-fold molar excess of the ALDH inhibitor , diethylamino benzaldehyde ( DEAB ) . This results in a significant decrease in the fluorescent intensity of ALDH-positive cells and was used to compensate the flow cytometer . NMECs , SUM159 , and MCF7 cell lines were infected with the lentivirus expressing human PTEN shRNA or control lentiviruses . The pLentilox 3 . 7 vector contains DsRed sequences in its backbone for selection . Following 12–16 h of incubation in suspension cultures of NMECs or monolayer cultures of breast cancer cell lines , the viruses were replaced with fresh medium . All mice were housed in the AAALAC-accredited specific pathogen-free rodent facilities at the University of Michigan . Mice were housed on sterilized , ventilated racks and supplied with commercial chow and sterile water both previously autoclaved . All experimentation involving live mice were conducted in accordance with standard operating procedures approved by the University Committee on the Use and Care of Animals at the University of Michigan . The fat pads of three week old NOD/SCID mice ( NOD . CB17-Prkdcscid/J , stock number 001303 ) purchased from the Jackson Laboratories were cleared and replaced with 1:1 mixture of irradiated and nonirradiated human fibroblasts obtained from reduction mammoplasties . Within 2–3 wk , human fibroblasts generated a humanized mammary fat pad in mice . These resulting fat pads were injected with NMECs infected with DsRed or PTEN shRNA lentiviral vectors . A 60-d release 0 . 18-µg estrogen pellet ( Innovative Research of America ) was implanted subcutaneously in each mouse . The mice were killed after 4–8 wk and the fat pads were analyzed for outgrowths . An identical protocol was followed for injection of breast cancer cell lines and xenografts , except that they were injected directly into the fat pads of NOD/SCID mice . Indicated dilutions of cell lines or xenografts were directly injected into fat pads of NOD/SCID mice . We used five mice for each experiment . Cells were lysed in Laemmli buffer , boiled , and subjected to SDS-PAGE . The proteins were transferred onto Nitrocellulose Membranes ( Pierce ) using semi dry Trans-Blot ( Bio Rad Laboratories ) . Blots were first incubated in TBS blocking buffer containing either 2% milk or 2% BSA ( for phospho-specific antibodies ) for 1 or 2 h at room temperature and then with the respective primary antibodies diluted in TBST ( containing 0 . 1% Tween20 and 2% BSA ) either overnight at 4°C , or 2 h at room temperature . Subsequently , blots were washed and incubated with appropriate secondary antibodies ( GE Healthcare ) in TBST and detected using SuperSignal West Pico Chemiluminescent Substrate ( Pierce ) . For immunohistochemistry , paraffin-embedded sections were deparaffinized in xylene and rehydrated in graded alcohol . Antigen enhancement was done by incubating the sections in citrate buffer pH6 ( Dakocytomation ) as recommended . Staining was done using peroxidase histostain-Plus Kit ( Zymed ) according to the manufacturer's protocol . Sections were incubated with primary antibodies for 1 h . Following the incubation with broad-spectrum secondary antibody and the HRP-conjugated streptavidin , AEC ( Zymed ) was used as substrate for peroxidase . Slides were counter-stained with hematoxylin and coverslipped using glycerin . For fluorescent staining , cells were fixed with 95% methanol at −20°C for 10 min . After rehydrating in PBS , cells were incubated with respective antibodies at room temperature for one hour , washed and incubated 30 min with FITC-conjugated secondary antibodies . The nuclei were stained with DAPI/antifade ( Invitrogen ) and coverslipped . Sections were examined with a fluorescent microscope ( Leica ) . Statistical differences for the number of mammospheres , GFP-positive cells , Aldefluor assays and tumor growths were determined using Students t-test .
|
Healthy adult tissues are maintained through the regulated proliferation of a subset of cells known as tissue stem and progenitor cells . Many cancers , including breast cancer , also are thought to arise from and be maintained by a small population of cells that display stem cell-like properties . These so-called “cancer stem cells” may also contribute to tumor spread ( metastasis ) , resistance to treatment , and disease relapse . Effective , long-lasting cancer treatments likely will need to target and eliminate these cancer stem cells specifically . Regulatory pathways responsible for maintaining cancer stem cells therefore may be promising targets for treatment . Breast cancers in humans frequently display abnormalities in the PTEN/PI3K/Akt pathway . We demonstrate using cell culture and a mouse model of breast cancer that stem or progenitor cells in both normal breast tissue and breast tumors are dependent for their continued growth on this pathway and on the Wnt/β-catenin pathway . We further show that the drug perifosine , which inhibits the kinase Akt , is able specifically to reduce the population of breast cancer stem or progenitor cells growing in mice . Our findings support the idea that drugs that selectively target breast cancer stem cells through the PTEN/PI3K/Akt pathway may reduce tumor growth and metastasis and result in improved patient outcomes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"molecular",
"biology",
"oncology/breast",
"cancer",
"cell",
"biology/cell",
"signaling"
] |
2009
|
Regulation of Mammary Stem/Progenitor Cells by PTEN/Akt/β-Catenin Signaling
|
Sporulation is an ancient developmental process that involves the formation of a highly resistant endospore within a larger mother cell . In the model organism Bacillus subtilis , sporulation-specific sigma factors activate compartment-specific transcriptional programs that drive spore morphogenesis . σG activity in the forespore depends on the formation of a secretion complex , known as the “feeding tube , ” that bridges the mother cell and forespore and maintains forespore integrity . Even though these channel components are conserved in all spore formers , recent studies in the major nosocomial pathogen Clostridium difficile suggested that these components are dispensable for σG activity . In this study , we investigated the requirements of the SpoIIQ and SpoIIIA proteins during C . difficile sporulation . C . difficile spoIIQ , spoIIIA , and spoIIIAH mutants exhibited defects in engulfment , tethering of coat to the forespore , and heat-resistant spore formation , even though they activate σG at wildtype levels . Although the spoIIQ , spoIIIA , and spoIIIAH mutants were defective in engulfment , metabolic labeling studies revealed that they nevertheless actively transformed the peptidoglycan at the leading edge of engulfment . In vitro pull-down assays further demonstrated that C . difficile SpoIIQ directly interacts with SpoIIIAH . Interestingly , mutation of the conserved Walker A ATP binding motif , but not the Walker B ATP hydrolysis motif , disrupted SpoIIIAA function during C . difficile spore formation . This finding contrasts with B . subtilis , which requires both Walker A and B motifs for SpoIIIAA function . Taken together , our findings suggest that inhibiting SpoIIQ , SpoIIIAA , or SpoIIIAH function could prevent the formation of infectious C . difficile spores and thus disease transmission .
A small subset of bacteria can survive adverse environmental conditions by forming a metabolically dormant cell-type known as an endospore ( referred to as a “spore” hereafter ) [1–3] . Spore formation allows bacteria to survive harsh environmental conditions , such as heat , desiccation , oxygen-rich environments , disinfectants , and antibiotic treatment , since they can “reawaken” when favorable conditions return [1–4] . While spore formation is an ancient and adaptive mechanism for members of the Firmicutes , this developmental process is essential for the survival of many obligate anaerobes that inhabit or transiently live in the gut [5 , 6] . Clostridium difficile is a spore-forming obligate anaerobe that is a leading cause of nosocomial diarrhea and a major threat to healthcare systems around the world [7–10] . When C . difficile spores are ingested by susceptible hosts , they germinate in the gut and outgrow to form toxin-secreting vegetative cells [7 , 11 , 12] . While the toxins produced by C . difficile are responsible for the disease infection symptoms , spores are essential for this obligate anaerobe to transmit disease [6] . Accordingly , during growth in the gastrointestinal tract , C . difficile strongly induces sporulation in order to survive exit from the host [6 , 13] . Spores complicate C . difficile infection clearance because they are resistant to many disinfectants and inert to antibiotics [4] . As a result , they can persist in the environment for long periods of time and facilitate C . difficile disease recurrence [12 , 14 , 15] . Recurrent C . difficile infections are particularly problematic because they can lead to severe complications such as pseudomembranous colitis , toxic megacolon , and death [14–16] . However , despite the importance of spores to the pathogenesis of C . difficile , the molecular mechanisms underlying infectious spore formation remain largely uncharacterized . Transmission electron microscopy analyses of several spore-forming organisms including C . difficile have shown that sporulation is defined by a series of morphological events starting with the formation of a polar septum , which generates a larger mother cell and smaller forespore [1–3 , 17] . The mother cell engulfs the forespore to create a protoplast surrounded by two lipid bilayer membranes suspended within the mother cell cytosol . The germ cell wall between the two membranes serves as the template for the synthesis of a thick protective layer of modified peptidoglycan called the cortex , while a series of protective proteinaceous shells called the spore coat is deposited on the outer forespore membrane [2 , 18] . Once forespore maturation is complete , the mother cell lyses to liberate a highly resistant spore . Our knowledge of how these morphological events occur derives primarily from studies of the organism Bacillus subtilis . These analyses have revealed that morphological changes during sporulation are coupled to compartment-specific transcriptional changes [1–3] . In particular , the sequential and compartment-specific activation of four conserved sporulation-specific sigma factors , σF , σE , σG , and σK , leads to the activation of transcriptional programs that allow key morphological stages to be completed [1–3 , 17] . Following asymmetric division , σF and σE are activated early in the forespore and mother cell , respectively; following forespore engulfment , σG and σK are activated in the forespore and mother cell , respectively . These activation events depend upon coordinated intercompartmental signaling events . σF- and σG-dependent signaling in the forespore activates σE and σK in the mother cell , respectively , via regulated intramembrane proteolysis . σF and σE control σG activation in the forespore following engulfment completion by inducing the formation of a channel , also known as the “feeding tube” [19 , 20] . While the precise composition of this channel has not been determined , “feeding tube” components are thought to physically connect the mother cell to the forespore and transport unknown substrates that are required for σG activity in the forespore [19–21] . The “feeding tube” also controls forespore integrity [20] , since the forespore collapses and eventually lyses in mutants lacking channel components [20 , 22] . σG activity may be further regulated by its apparent dependence on engulfment completion [23–25] . Analyses of sporulation-specific sigma factor function in C . difficile have revealed important differences in the regulatory architecture controlling sporulation [17 , 26] . While the sigma factors are controlled in a similar compartment-specific manner , σE activation only partially depends on σF; σG activation does not require σE; and σK activation does not depend on σG [27–29] . Since a C . difficile sigE mutant , which is stalled at asymmetric division , still activates σG in the forespore [28] , C . difficile σG activity does not appear to be coupled to engulfment completion , in contrast with B . subtilis [23] . In general , activation of C . difficile sporulation-specific sigma factors appears to depend less on intercompartmental signaling and morphological changes than B . subtilis [17 , 26] . Since genome-wide transcriptional profiling has shown that σG regulon genes are expressed at wildtype levels in a C . difficile sigE−mutant [27 , 29] , the mother cell-to-forespore channel shown to regulate B . subtilis σG activity appears to be dispensable for C . difficile σG activity , at least at early stages of sporulation [30] . Intriguingly , however , the genes encoding B . subtilis channel components , spoIIQ and the eight gene spoIIIA operon , are conserved across all spore-forming bacteria [31 , 32] and are induced during sporulation in a manner analogous to B . subtilis , with σF activating spoIIQ transcription and σE activating spoIIIA transcription [27 , 29] . These observations suggest that the mother cell-to-forespore channel may play important but possibly distinct roles during C . difficile spore formation relative to B . subtilis [30] . SpoIIQ has homology to Zn2+-dependent M23 peptidases ( LytM domain , [33 , 34] ) and forms a multimeric ring in the inner forespore membrane of B . subtilis [35–38] . The SpoIIIA proteins , SpoIIIAA-SpoIIIAH [39] , have homology to secretion system components [19–21 , 40] . SpoIIIAA appears to function as an ATPase that likely powers the transport of metabolites across the “feeding tube” during B . subtilis sporulation [20] . SpoIIIAH forms a multimeric ring in the mother cell-derived outer forespore membrane that directly binds the SpoIIQ multimeric ring formed in the inner forespore membrane [34 , 35 , 37 , 38] . The SpoIIQ-SpoIIIAH complex alone can drive “zipper-like” engulfment in sporulating B . subtilis lacking a cell wall [41] . Based on these observations , this complex has been proposed to function as a Brownian “ratchet” that helps power engulfment . Consistent with this model , a spoIIQ mutant fails to complete engulfment [33] when sporulation is induced by nutrient starvation , even though “feeding tube” mutants can complete engulfment when sporulation is induced by resuspension [20] . C . difficile SpoIIIAA and SpoIIIAH exhibit 57% and 38% similarity , respectively , to their orthologs in B . subtilis ( S1 and S2 Figs ) , while C . difficile SpoIIQ ( CD0125 ) exhibits only 28% similarity despite also encoding a C-terminal LytM domain ( [32] , S3 Fig ) . In contrast with the degenerate active site of B . subtilis SpoIIQ [34] , C . difficile SpoIIQ has an intact active site ( [30] , S3 Fig ) , suggesting that it may have peptidoglycan endopeptidase activity and thus function differently in C . difficile relative to B . subtilis . Furthermore , residues that directly mediate binding between B . subtilis SpoIIQ and SpoIIIAH are not well conserved in C . difficile ( S1 and S2 Figs ) , raising the question as to whether these proteins interact in C . difficile . Indeed , whether SpoIIQ and/or SpoIIIA proteins regulate forespore integrity and/or have additional functions during C . difficile sporulation remain unknown [30] . To address these questions , we constructed gene disruptions of C . difficile spoIIQ , spoIIIAA , and spoIIIAH and determined their effects on spore formation using microscopic and cell biological assays . We also tested whether C . difficile SpoIIQ and SpoIIIAH interact and whether the predicted ATPase and endopeptidase activities of C . difficile SpoIIIAA and SpoIIQ are required for spore formation . These analyses revealed that SpoIIQ , SpoIIIAA , and SpoIIIAH regulate multiple stages of C . difficile spore formation , including engulfment , proper coat localization around the forespore , and maintenance of the forespore .
To first determine if C . difficile spore development depends on the SpoIIQ and SpoIIIA proteins , we constructed targeted gene disruptions in the σF-regulated spoIIQ and σE-regulated spoIIIAA and spoIIIAH genes using the ClosTron gene knockout system ( S4 Fig , [42] ) . Since targetron insertion into the spoIIIAA gene likely causes polar effects on the spoIIIA operon , which includes spoIIIAA-spoIIIAH ( S4 Fig ) , the spoIIIAA mutation will be referred to as a spoIIIA mutant from hereon . However , since a second promoter within the spoIIIA operon has been shown to drive expression of spoIIIAG and spoIIIAH ( [29] , S4 Fig ) , the spoIIIA mutant likely still produces SpoIIIAH . Microscopic analysis of spoIIQ , spoIIIA , and spoIIIAH mutants during sporulation using the membrane dye FM4-64 and the nucleoid dye Hoechst revealed that these mutants are defective in engulfment ( Fig 1 ) . The percentage of cells captured at ( i ) asymmetric division , ( ii ) pre-engulfment with FM4-64 staining and Hoechst staining , ( iii ) post engulfment with FM4-64 staining and Hoechst staining , ( iv ) post engulfment with FM4-64 staining and Hoechst exclusion , ( v ) post engulfment with DIC-bright , FM4-64 exclusion and Hoechst exclusion , and ( vi ) free spore was quantified based on analyses of 100 sporulating cells . Whereas uniform staining of FM4-64 around the entire forespore or the presence of DIC-bright spore compartments was observed in wildtype sporulating cells 73% of the time ( blue , green , white , and pink arrows ) , FM4-64 staining of sporulating spoIIQ and spoIIIAH mutants was restricted to the curved membrane at the mother cell-forespore interface ( yellow arrows ) , and no DIC-bright forespore compartments were observed in these mutants , indicative of an engulfment defect . While the spoIIQ and spoIIIAH mutants both failed to complete engulfment , the spoIIIA and sigG mutants completed engulfment 10% and 24% of the time , respectively , although they did not mature to a stage that excluded Hoechst or FM4-64 ( Fig 1 ) . Taken together , these results suggest that SpoIIQ and SpoIIIAH are required for C . difficile forespore engulfment , while the SpoIIIAA-AF complex may be only partially required for engulfment . Disruption of C . difficile spoIIQ , spoIIIA , and spoIIIAH resulted in a significant decrease in heat-resistant spore formation relative to wild type . Interestingly , while the C . difficile spoIIQ mutant did not show evidence of mature spore formation by fluorescence microscopy , we observed only a 200-fold defect in heat resistance relative to wild type ( Fig 1 ) . Since this defect was not as severe as the ~4–6 log defect reported for B . subtilis spoIIQ mutants [19] , we investigated the possibility that the heat-resistant C . difficile spoIIQ mutant cells might arise from a heritable change by testing the heat resistance of subcultured spoIIQ−colonies that arose following heat treatment . The same frequency of heat resistance was observed , indicating that the production of heat-resistant spoIIQ−spores is a stochastic event . In contrast with the C . difficile spoIIQ mutant , no heat-resistant spores were observed for the C . difficile spoIIIA−and spoIIIAH−strains within the limits of detection of our assay ( <10−6 , Fig 1 ) . The heat resistance defect of the C . difficile spoIIIA−mutant was similar to the defect reported for a B . subtilis ΔspoIIIAA mutant [20] , although the C . difficile spoIIIAH−mutant was at least 4-logs more severe than the defect of a B . subtilis ΔspoIIIAH mutant [20] . To confirm that the spoIIQ , spoIIIA , and spoIIIAH gene disruptions abrogated protein production , we analyzed sporulating cell lysates prepared from spoIIQ , spoIIIA , and spoIIIAH mutants by Western blotting using antibodies raised against SpoIIQ and SpoIIIAH . As expected , SpoIIQ and SpoIIIAH were not detected in the spoIIQ and spoIIIAH mutants , respectively , and both proteins were absent in the spo0A mutant , which fails to initiate sporulation altogether ( Fig 2A ) . Consistent with spoIIQ and spoIIIAH regulation by σF and σE , respectively [27 , 29] , SpoIIQ was absent from the sigF mutant , and SpoIIIAH was absent in the sigE mutant . SpoIIIAH was nevertheless detected at wildtype levels in the spoIIIA mutant , since an internal promoter drives expression of spoIIIAG-AH ( [29] , S4A Fig ) similar to the regulation of the B . subtilis spoIIIA operon [43] . Wildtype levels of SpoIIQ were observed in the spoIIIAH mutant , and vice versa , suggesting that loss of the predicted interaction between SpoIIQ and SpoIIIAH did not affect their steady state levels . Wildtype levels of SpoIIQ and SpoIIIAH were also observed in the sigG mutant , suggesting that this mutant’s engulfment defect did not result from the absence of these components . The small amount of SpoIIIAH that was detected in a sigF mutant ( Fig 2A ) is consistent with the partial activation of σE in a sigF mutant [27 , 29] . Since we were unable to generate a working antibody for detecting SpoIIIAA , we measured spoIIIAA transcript levels in the same strains . Consistent with the previously reported regulation of spoIIIAA by σE [27 , 29] , statistically significant decreased levels of spoIIIAA transcripts were observed in spo0A , sigF , and sigE mutants relative to wild type ( Fig 2B , p < 0 . 01 ) . spoIIIAA transcript levels were unaffected in the spoIIQ , spoIIIAH , and sigG mutants ( Fig 2B ) , indicating that loss of SpoIIQ or SpoIIIAH does not alter spoIIIAA expression . spoIIIAA transcripts could not be accurately measured in the spoIIIA mutant , since the amplification product is downstream of the targetron insertion . To validate that the observed mutant phenotypes were due to the targeted insertions , we attempted to complement the mutant strains with a wildtype copy of the disrupted gene ( s ) expressed from their native promoter using the pMTL83151 multicopy plasmid [44] . The spoIIIA mutant was complemented with the full spoIIIA operon , and the spoIIIAH mutant was complemented with either the full spoIIIA operon , or the spoIIIAH gene alone ( S5 Fig ) . The spoIIQ and spoIIIA complementation constructs all restored production of heat-resistant , DIC-bright spores to their respective mutant backgrounds ( S5 Fig ) . While complementation of the spoIIIAH mutant with either the spoIIIA operon or spoIIIAH gene under the control of the spoIIIA promoter restored heat-resistant spore production , the spoIIIA operon conferred ~8-fold higher heat-resistance to the spoIIIAH mutant relative to complementation with the spoIIIAH gene alone . Western blot analysis revealed that SpoIIIAH levels were elevated in the spoIIIA operon complementation strain relative to spoIIIAH complementation strain and wildtype carrying empty vector ( S6 Fig ) . Complementation of spoIIQ−resulted in ~4-fold greater heat-resistant spore formation than wildtype carrying empty vector . Western blot analysis indicated that SpoIIQ levels were slightly elevated in the spoIIQ complementation strain relative to wildtype carrying empty vector ( S6 Fig ) . Fluorescence microscopy analyses of the spoIIQ , spoIIIA , and spoIIIAH strains carrying empty vector confirmed that the majority of mutant cells failed to complete engulfment ( S5 Fig , yellow arrows ) . On rare occasions , we observed that the spoIIIA mutant carrying empty vector completed engulfment ( S5 Fig , blue arrow ) , similar to our observations with the spoIIIA mutant alone ( Fig 1 ) . Regardless , these results indicate that the gene disruptions in the spoIIQ , spoIIIAA and spoIIIAH genes are responsible for the observed engulfment and heat-resistance defects . To gain further insight into the nature of the engulfment defect in the spoIIQ , spoIIIA , and spoIIIAH mutants , we analyzed each mutant using transmission electron microscopy ( TEM ) . We failed to observe engulfment of spoIIQ and spoIIIAH mutants based on analyses of over 50 cells that had progressed beyond asymmetric division for each mutant , with engulfment being defined as the mother cell-derived membrane surrounding the entire forespore ( Fig 3 ) . The spoIIIA mutant was observed to complete engulfment in ~20% of cells analyzed by TEM , even though this mutant failed to produce heat-resistant spores ( Fig 1 ) . In contrast , none of the spoIIQ and spoIIIAH mutant cells strains completed engulfment . However , since heat-resistant spoIIQ−spores could be detected at a frequency of 1 in 200 ( Fig 1 ) , we extensively analyzed the TEM grids and identified a single spoIIQ−cell that had completed engulfment ( S7A Fig ) . Taken together , these results confirm the live cell microscopy analyses ( Fig 1 ) : loss of SpoIIQ and SpoIIIAH causes a severe defect in forespore engulfment , while the apparent loss of SpoIIIAA-AF in the spoIIIA mutant still permits engulfment in 10–20% of cells . In addition to the engulfment defects observed in the spoIIQ , spoIIIA , and spoIIIAH mutants by TEM , a second “compartment” was often observed to extend from the forespore of the spoIIQ , spoIIIA , and spoIIIAH mutants ( Fig 3 , white arrows ) . Closer inspection of these extensions revealed multiple striated lines that were consistent with coat . The mutant coat-like structures appeared to anchor to the leading edge of the engulfing membrane but were not adhered to the mother cell-forespore interface in the majority of cells with engulfment defects . Coat-like structures were present 100% of the time in spoIIQ– , spoIIIA– , and spoIIIAH−strains that had begun engulfment; these structures appeared anchored to the leading edge of the engulfing membrane 94% , 96% , and 66% of the time in spoIIQ– , spoIIIA– , and spoIIIAH−strains , respectively . A similar phenotype was observed in the sigG−strain ( Fig 3 , [27] ) . In some instances , the coat-like structures were not associated with the forespore at all and were instead mislocalized to the mother cell cytosol , similar to the previously described coat mislocalization phenotype of a spoIVA mutant ( [45] , Fig 3 , yellow arrows ) . In particular , mislocalized cytosolic coat was observed with high frequency in the spoIIIAH mutant ( 51% ) , regardless of whether the coat was anchored to the leading edge . For the spoIIQ– , spoIIIA– , and spoIIIAH−cells that had coat anchored to the leading edge of the engulfing membrane , 98% , 89% , and 94% , respectively , did not have coat intimately associated with the forespore interface ( Fig 3 , white arrows ) . Notably , the rare spoIIIA and spoIIQ mutants that completed engulfment had visible coat surrounding the forespore compartment by TEM ( S7A Fig , black arrows ) . Furthermore , in wildtype cells , coat was only observed after engulfment was complete . To confirm that the coat-like assemblages observed in the spoIIQ , spoIIIA , spoIIIAH , and sigG mutants were indeed coat , we analyzed the localization of a known coat protein in these mutant backgrounds . In particular , we correlated the localization of the previously reported surface-exposed coat protein CotE fused to a SNAP imaging tag [28] with FM4-64 and Hoechst staining . The CotE-SNAP protein fusion was detected concentrated at both poles of the developing forespore in wild type , with a weaker signal surrounding the forespore ( Fig 4 ) similar to the previously reported localization of this protein fusion around the forespore [28] . Faint CotE-SNAP staining was observed around free spores of wild type , consistent with the surface localization of CotE [46 , 47] . In contrast , in the spoIIQ , spoIIIA , and spoIIIAH mutants , CotE-SNAP signal was frequently offset from FM4-64 staining of the forespore membrane . The CotE-SNAP signal was also observed mislocalized to the mother cell cytosol in these mutants ( Fig 4 ) , similar to the displacement of the CotE-SNAP signal to the mother cell cytosol of the spoIVA mutant , which has previously been shown to mislocalize coat [45] . While the FM4-64 readily stained forespore membranes , it also appeared to associate with mislocalized coat in the spoIVA , spoIIQ , spoIIIA , and spoIIIAH mutants ( Fig 4 , yellow arrows ) , making it difficult to assess by light microscopy whether CotE-SNAP was adhered to the forespore membrane . Nevertheless , combined with our TEM data , the CotE-SNAP localization experiments strongly suggest that coat detaches from the forespore and/or completely mislocalizes to the mother cell cytosol in the absence of SpoIIQ and SpoIIIA proteins . While the dominant phenotypes observed by TEM for spoIIQ , spoIIIA , and spoIIIAH mutants are shown in Fig 3 , 13% , 14% , and 27% of spoIIQ , spoIIIA , and spoIIIAH mutant cells , respectively , harbored forespores that were undergoing forespore collapse . In particular , large invaginations of the forespore membrane were observed in these mutants ( S7B Fig , blue arrows ) , similar to the phenotypes previously described for B . subtilis mutants lacking SpoIIQ or SpoIIIA complex components [20] . These results indicate that these proteins in C . difficile are also required to maintain forespore integrity , similar to B . subtilis [20] . In addition to maintaining forespore integrity in B . subtilis , the feeding tube is required to sustain transcription in the forespore [19] and thus is necessary for σG activity [19 , 20 , 25 , 40] . However , previous transcriptional analyses in C . difficile suggested that the “feeding tube” components were dispensable for σG activity , since a C . difficile σG-dependent transcriptional reporter is produced in the forespore of a sigE mutant [28] , and the σG regulon is expressed at wildtype levels in the sigE mutant [27 , 29] . To test whether C . difficile spoIIQ , spoIIIA , and spoIIIAH are required for σG activity , we measured σG-dependent transcript levels in wild type , spoIIQ , spoIIIA and sporulation sigma factor mutants using quantitative RT-PCR . As expected , no statistically significant difference in σG-dependent transcripts spoVT , spoVAD , and CD1430 [27 , 29] , were observed in the feeding tube mutants ( S8 Fig ) . Consistent with the dependence of σG activity on Spo0A and σF , spoVT was significantly decreased in spo0A , sigF , and sigG mutants ( p < 0 . 0005 ) , CD1430 was significantly decreased in spo0A , sigF , and sigG mutants ( p < 0 . 05 ) and spoVAD was significantly decreased in spo0A , sigF , and sigG mutants ( p < 0 . 01 , 0 . 01 , and 0 . 05 , respectively ) . Since it is possible that the wildtype levels of σG activity detected in spoIIQ and spoIIIA mutants by qRT-PCR may derive from improper σG activation in the mother cell , we used the σG-dependent SNAP-tag transcriptional reporter to visualize σG activation in spoIIQ and spoIIIA mutants . The promoter region of the σG-dependent sspA gene previously described by Pereira et al . [28] was fused to a codon-optimized SNAP gene and conjugated into wildtype , sigF– , spoIIQ– , sigE– , spoIIIA– , spoIIIAH– , and sigG−strains . Similar to the previous reports [28] , SNAP labeling with the TMR-Star substrate ( i . e . σG activity ) was restricted to the forespore of cells that had completed asymmetric division or engulfment ( Fig 5A ) . No SNAP signal was detectable in either the sigF or sigG mutants ( Fig 5A , black arrows ) , as expected . In contrast , σG-dependent SNAP labeling was detectable in the forespores of cells undergoing sporulation in the spoIIQ– , sigE– , spoIIIA– , and spoIIIAH−strains ( Fig 5A , yellow arrows ) . The prevalence of σG-dependent transcription in cells undergoing sporulation ( based on the presence of an asymmetric septum , engulfment initiated , or engulfment completed phenotype ) was determined by counting the number of cells exhibiting SNAP labeling . The σG-dependent transcriptional reporter was produced in the forespore of spoIIQ– , spoIIIA– , and spoIIIAH−cells 58% , 57% , and 50% , respectively , of sporulating cells , which was similar to the frequency observed in wild type ( 57% , Fig 5A ) . The SNAP signal was observed in the forespore of the sigE mutant in 28% of sporulating cells . Western blot analyses confirmed that wildtype levels of SNAP protein were observed in spoIIQ , sigE , spoIIIA , and spoIIIAH mutants ( Fig 5B ) . Taken together , these results demonstrate that the C . difficile SpoIIQ and SpoIIIA components are dispensable for maintaining transcription in the forespore , in contrast with B . subtilis [19] . Since these analyses indicated that the C . difficile SpoIIQ and SpoIIIA proteins regulate different cellular processes during sporulation than in B . subtilis , namely σG activity and forespore engulfment , we next sought to investigate how these components regulated these processes . We first tested whether the C . difficile feeding tube components assemble into a complex as has been shown in B . subtilis [34] , since the interaction between SpoIIQ and SpoIIIAH is necessary for feeding tube function in B . subtilis [35] . Using a co-affinity purification assay , we determined whether C . difficile SpoIIQ and SpoIIIAH directly interact through their extracellular domains . To this end , we co-expressed His6-tagged SpoIIQ and HA-tagged SpoIIIAH , both lacking their transmembrane domains , in E . coli . Affinity purification of His6-tagged SpoIIQ resulted in the co-purification of HA-tagged SpoIIIAH , whereas HA-tagged SpoVT , which was used as a specificity control , did not co-purify with His6-tagged SpoIIQ when co-expressed ( Fig 6 ) . Thus , despite the low degree of sequence homology between B . subtilis and C . difficile SpoIIQ orthologs ( S3 Fig ) , C . difficile SpoIIQ and SpoIIIAH directly interact in vitro through their extracellular domains , consistent with the hypothesis that these proteins form a complex that bridges the intercompartmental space between the mother cell and forespore [34 , 41] . Based on these findings , we next tested whether the predicted catalytic activities of SpoIIQ and SpoIIIAA were required for their function . The LytM domain of C . difficile SpoIIQ carries an intact catalytic triad consisting of two conserved motifs: HxxxD and HxH [30] . These motifs coordinate a metal ion ( commonly zinc ) that is essential for endopeptidase activity , which degrades the peptide linkages that crosslink the glycan strands of peptidoglycan [48] . To determine if C . difficile SpoIIQ’s endopeptidase activity is necessary for sporulation , we complemented the spoIIQ mutant with a spoIIQ variant encoding a histidine 120 to alanine mutation ( spoIIQ–/H120A , S3 Fig ) , which should inactive its predicted endopeptidase activity . Analysis of this strain in the heat resistance assay indicated that the H120A mutation caused an ~50% reduction in heat-resistant spore formation relative to wildtype carrying empty vector ( S9 Fig ) . TEM analyses revealed that 52% of spoIIQ–/H120A cells completed engulfment compared to 88% of the wildtype complementation strain ( spoIIQ–/spoIIQ ) and 96% of wildtype carrying empty vector ( S9 Fig ) . These results indicate that the endopeptidase activity of SpoIIQ plays a minor role in regulating C . difficile sporulation under the conditions tested . SpoIIIAA is predicted to function as an ATPase , since strains carrying mutations of conserved residues in the Walker A and B boxes ( S1 Fig , [49] ) in B . subtilis resemble a ΔspoIIIAA mutant [20] . Disruption of the Walker A motif typically prevents ATP binding [50] , while disruption of the Walker B motif typically prevents ATP hydrolysis without affecting ATP binding [51 , 52] . To determine whether the ATPase activity of C . difficile SpoIIIAA is also required for spore formation , we constructed complementation strains encoding SpoIIIAA carrying a Walker A lysine mutation , K167A , a Walker B aspartate mutation , D244A , and a Walker A/Walker B double mutation K167A/D244A ( S1 Fig ) , and tested their ability to complement the heat resistance defect of a spoIIIA mutant . The K167A mutant , expressed from a spoIIIA operon complementation plasmid exhibited ~300-fold defect relative to wildtype carrying empty vector ( Fig 7 ) . Interestingly , only 6% of sporulating K167A cells completed engulfment when analyzed by TEM compared to 20% of the parent spoIIIA−strain carrying empty vector ( Fig 7 ) , and similar results were observed by FM4-64 and Hoechst staining ( S10 Fig ) . The D244A mutant exhibited close to wildtype levels of heat resistance ( ~3-fold decrease ) , consistent with its ability to complete engulfment 35% of the time ( Fig 7 ) . The K167A/D244A double mutant resembled the K167A single mutant in exhibiting a three-log decrease in heat resistance and ~4% engulfment efficiency relative to wild type . In contrast , the wildtype complementation strain ( spoIIIA–/spoIIIA ) exhibited wildtype levels of heat resistance and engulfment completion ( Fig 7 ) . Since we could not test whether the K167A , D244A , or K167A/D244A mutation ( s ) affected SpoIIIAA protein levels due to the absence of a working antibody , we compared spoIIIAA transcript levels in the K167A mutant , whose heat resistance and engulfment defect was more severe than the D244A mutant and equivalent to the double mutant ( Fig 7 ) , relative to wildtype carrying empty vector and the spoIIIA complementation strain . These analyses indicated that spoIIIAA transcript levels in the K167A complementation strain were similar to the spoIIIA complementation strain and wildtype carrying empty vector ( S6B Fig ) . Since the Walker A K167A mutation was considerably more severe than the Walker B D244A mutation , our results suggest that SpoIIIAA function likely depends on its ability to bind , but not necessarily hydrolyze , ATP . Given that B . subtilis SpoIIIAA function completely depends on the presence of intact Walker A and Walker B boxes [20] , C . difficile SpoIIIAA appears to have differential requirements for its function relative to B . subtilis . The engulfment defects of the C . difficile spoIIQ , spoIIIA , and spoIIIAH mutants prompted us to investigate the mechanisms underlying this engulfment defect . In B . subtilis , peptidoglycan hydrolase enzymes that degrade the peptidoglycan layer between the mother cell and forespore drive engulfment [53–55] . Subsequent transformations of this peptidoglycan layer , which involve both the making and breaking of peptide and glycan bonds , essentially “cut” the forespore free of the mother cell until engulfment is complete [56] . Since transpeptidases and/or ligases can incorporate D-alanine into the stem peptide that is conjugated to the glycan strand of peptidoglycan [57] , newly remodeled and/or synthesized peptidoglycan can be metabolically labeled using unnatural D-alanine derivatives conjugated to bioorthogonal functional groups [58 , 59] . To determine if peptidoglycan remodeling and/or synthesis is active during C . difficile forespore engulfment , we incubated sporulating C . difficile cultures with D-alanine bearing an alkyne group and visualized its incorporation into peptidoglycan over time using copper-catalyzed click chemistry [59] . Alkyne D-alanine , referred to as “alkDala , ” was labeled through the azide-alkyne cycloaddition of an azide group conjugated to a fluorescein-derivative [60] . Analysis of wildtype sporulating cells using this metabolic labeling assay revealed that fluorescent peptidoglycan signal ( PG ) was detectable within 10 min of incubating the culture with alkDala ( Figs 8 and S11 ) . After 30 min of incubation with alkDala , the peptidoglycan signal was observed surrounding the forespore and , to a lesser extent , the mother cell ( Fig 8 ) . In contrast , when the spoIIQ mutant was incubated with alkDala for 10 min or longer , peptidoglycan remodeling and/or synthesis was localized primarily at the curved septa at the mother cell-forespore interface ( Fig 8 ) , consistent with the spoIIQ mutant’s engulfment defect ( Fig 1 ) . Incubation of the sigE mutant with alkDala resulted in labeling of the polar septa and mother cell peptidoglycan after 20 min of incubation with alkDala . For comparison , B . subtilis engulfment requires ~45 min to complete [61 , 62] , and sporulation occurs more slowly in C . difficile than in B . subtilis [27 , 28] . To determine the optimal length of time for measuring alkDala incorporation during sporulation , we analyzed the distribution of the alkDala label in sporulating wild type cells 10’ , 20’ , 30’ , and 40’ after alkDala addition . Sporulating wildtype cells with visible peptidoglycan labeling were binned into the following categories: ( i ) no staining of the forespore , ( ii ) labeling of the polar septum , ( iii ) partial labeling of the forespore on the mother cell distal side ( i . e . labeling after engulfment has initiated ) , ( iv ) partial labeling around the middle of the forespore , ( v ) partial labeling of the forespore on the mother cell proximal side , ( vi ) labeling around the entire forespore ( S11 Fig ) . Based on these analyses , we chose to label cells after a 30 minute incubation with alkDala , since this was the earliest time point at which full labeling of the forespore was detected in the majority of wildtype sporulating cells ( S11 Fig ) . As expected , the alkDala probe labeled division septa in all strains ( S12A Fig ) , and fluorescent labeling was not observed in wildtype cells incubated with D-alanine , which cannot undergo cycloaddition ( S12B Fig ) , or at time 0 min , even though the samples were exposed to the azido-fluorophore probe ( Fig 9 ) . To ensure that alkDala specifically labeled newly transformed PG , we incubated WT cells with the cell wall synthesis inhibitors vancomycin and imipenem prior to addition of alkDala and evaluated alkDala incorporation by flow cytometry . Vancomycin inhibits cell wall synthesis by preventing both transpeptidation and transglycosylation [63] , and imipenem covalently inhibits the penicillin binding proteins required for transpeptidation [64] . Incubation of wildtype cells with alkDala resulted in a statistically significant increase in median fluorescent intensity ( MFI ) relative to the MFI of WT cells incubated with Dala ( S13 Fig; p < 0 . 0001 ) . While vancomycin treatment did not reduce alkDala labeling relative to the positive control in a statistically significant manner , imipenem treatment decreased alkDala label incorporation ~5-fold relative to the positive control ( p < 0 . 0001 ) . Taken together , these results suggest that the alkDala probe specifically labels newly synthesized and/or remodeled peptidoglycan , and peptidoglycan continuously surrounds the C . difficile forespore throughout engulfment as previously observed in B . subtilis [65] . Since the metabolic labeling time course demonstrated that the fluorescent signal was maximal after 30 min of alkDala incorporation , we used this labeling period to assess whether C . difficile mutants defective in engulfment could remodel and/or synthesize peptidoglycan around the forespore . Although the metabolic label was evenly distributed around the entire perimeter of wildtype forespores , the label was only partially distributed around the mother cell proximal side of the forespore in engulfment-defective spoIIQ , spoIIIA , spoIIIAH , and sigG mutants ( Fig 9 ) . These results suggest that the engulfment defect of the spoIIQ , spoIIIA , spoIIIAH , and sigG mutants is not due to a failure to activate peptidoglycan remodeling and/or synthesis . Instead , the active peptidoglycan transformations observed in the spoIIQ , spoIIIA , and spoIIIAH mutants appear to be insufficient to drive engulfment to completion .
Since the SpoIIIAA-AH components of the B . subtilis “feeding tube” channel are universally conserved in spore-forming organisms [31 , 32] , and a SpoIIQ-like ortholog is conserved in the Clostridia [32] , we hypothesized that these proteins would play a critical role in regulating C . difficile spore formation . In this study , we have demonstrated that C . difficile SpoIIQ and SpoIIIA proteins control forespore engulfment and integrity and the intimate association of the coat with the forespore ( Figs 3 , 4 and S7 ) . Although SpoIIQ and SpoIIIAH are strongly required for engulfment , the SpoIIIAA-AF proteins appear to be only partially required for engulfment completion in C . difficile , since the spoIIIA mutant completes engulfment ~10–20% of the time ( Figs 1 , 7 , and S7 ) . Given that this mutant produces wildtype levels of SpoIIQ and SpoIIIAH ( Fig 2 ) , and SpoIIQ and SpoIIIAH directly interact at least in vitro ( Fig 6 ) similar to their B . subtilis counterparts [34] , C . difficile SpoIIQ and SpoIIIAH would appear to be sufficient to complete engulfment in some spoIIIA mutant cells . While it remains possible that SpoIIIAG also regulates C . difficile forespore engulfment in the spoIIIA mutant , our observations are nevertheless consistent with the proposal that SpoIIQ-SpoIIIAH complex functions like a “Brownian” ratchet to allow for “zipper-like” engulfment [41] . Indeed , the finding that the SpoIIQ H120A mutant exhibits only a partial defect in engulfment completion and heat-resistant spore formation relative to wild type ( ~50% , S9 Fig ) implies that SpoIIQ-SpoIIIAH complex formation is more important for engulfment completion than the putative endopeptidase activity of C . difficile SpoIIQ . In contrast , B . subtilis SpoIIQ lacks endopeptidase activity , and the SpoIIQ-SpoIIIAH complex is dispensable for engulfment when sporulation is induced by the re-suspension method [20 , 41 , 66] . However , when sporulation is induced by nutrient exhaustion , B . subtilis SpoIIQ is required to complete engulfment [22 , 33 , 41] . This observation suggests that media composition causes changes within sporulating cells such that some sporulation proteins are differentially required for engulfment . Since the 70:30 media used to induce C . difficile sporulation in this study resembles the nutrient exhaustion media used in B . subtilis [67] , it will be interesting to test whether differences in media compositions and sporulation conditions ( e . g . broth vs . plate-based induction ) will lead to differential requirements for C . difficile SpoIIQ and SpoIIIAH during engulfment . Indeed , C . difficile sigG mutants appear to exhibit differences in engulfment completion when sporulation is induced in broth vs . on plates , although slight differences in strain background could be responsible for this difference [27 , 28] . Regardless , the engulfment defects of C . difficile spoIIQ and spoIIIAH mutants suggest that the ancestral function of the SpoIIQ-SpoIIIAH complex is to control engulfment during sporulation [68] . Even though SpoIIQ and SpoIIIAH appear to be sufficient to mediate engulfment in ~15% of C . difficile spoIIIA−cells , the spoIIIA−mutant nevertheless failed to produce heat-resistant spores . Since the spoIIIA−mutant is likely defective in producing the SpoIIIAA-AF proteins ( S4 Fig ) , these proteins would appear to regulate steps beyond engulfment during C . difficile sporulation ( Fig 1 ) . Indeed , our mutational analyses implicate SpoIIIAA’s predicted ability to bind ATP as being critical for engulfment completion , since mutation of the Walker A ATP binding motif ( K167A ) results in an ~300-fold defect in heat-resistant spore formation relative to wild type ( Fig 7 ) . Interestingly , ATP hydrolysis would appear to be less important for SpoIIIAA’s function during spore formation , since the Walker B mutant ( D244A ) exhibits only a 3-fold defect in engulfment and heat-resistant spore formation ( Fig 7 ) . Given that the phenotype of the K167A/D244A double mutant resembles that of the K167A single mutant , nucleotide binding by SpoIIIAA may induce a conformational change within the protein that is necessary for its function . Consistent with this hypothesis , B . subtilis SpoIVA Walker A ATP binding mutants exhibit different phenotypes from Walker B ATPase mutants [52] . Alternatively , a different aspartate residue may substitute for the predicted role of D244 in catalyzing C . difficile SpoIIIAA’s ATPase activity . While this functional redundancy is formally possible , we note that the equivalent Walker B mutation in B . subtilis SpoIIIAA ( D224A ) causes a heat-resistant spore formation defect equivalent to a ΔspoIIIAA mutant [20] . It will be important in future studies to determine whether C . difficile SpoIIIAA binds and hydrolyzes ATP , and whether these activities are necessary to power transport of proteins and/or metabolites from the mother cell to the forespore similar to B . subtilis SpoIIIAA [19–21] . While the C . difficile spoIIIA mutant completed engulfment in ~10–20% of cells yet failed to produce heat-resistant spores , the C . difficile spoIIQ mutant exhibited a severe engulfment defect and produced heat-resistant spores 0 . 5% of the time relative to wild type . The mechanism by which spoIIQ−cells form functional spores remains mysterious given that SpoIIIAH likely binds SpoIIQ during C . difficile sporulation ( Fig 6 ) and is required for heat-resistant spore formation . Interestingly , a differential requirement for SpoIIIAH is observed in B . subtilis , since a spoIIIAH mutant has a 1000-fold less severe phenotype relative to other spoIIIA mutants [20] . While a mechanism underlying these differential phenotypes remains unclear for both C . difficile and B . subtilis , functionally redundant mechanisms appear to exist in both organisms . Testing this hypothesis in C . difficile would be greatly aided by analyses of C . difficile SpoIIQ and SpoIIIA protein complex formation during sporulation . Although the SpoIIQ and SpoIIIA proteins regulate C . difficile forespore engulfment , these proteins appear dispensable for σG activity in the forespore ( Figs 5 and S8 ) as predicted [27 , 29] . Despite these observations , it nevertheless remains possible that these proteins are needed to sustain σG activity after engulfment is complete . Contrary to this model , the number of spoIIIA−cells that activated σG was identical to wild type ( Fig 5 ) . Nevertheless , since we cannot assess the duration of σG activity in the forespore due to the inability to synchronize sporulation in C . difficile [27 , 28 , 45] , it remains possible that the forespore may require resources from the mother cell in a SpoIIQ- and SpoIIIA-dependent manner during late stages of sporulation . Our analyses also uncovered a surprising role for C . difficile SpoIIQ and SpoIIIA proteins in regulating the adherence of the spore coat to the engulfing forespore . TEM analyses revealed that the spore coat appears to localize and anchor to the leading edge of the engulfing membrane but sometimes sloughs away from the mother cell-forespore interface ( Fig 3 ) . Unfortunately , little is known about the mechanisms by which the spore coat localizes around the forespore in C . difficile , since few coat morphogenetic proteins are conserved between C . difficile and B . subtilis [18] . SpoIVA and the clostridial-specific SipL have been shown to function as coat morphogenetic proteins by localizing the coat to the forespore in C . difficile [45] , but how these proteins are recruited to the forespore membrane is unclear . Our results suggest an intriguing link between engulfment completion and adhering coat around the forespore , since the minority of spoIIIA−and spoIIQ−cells that completed engulfment produced coat surrounding the forespore ( S7 Fig ) . Perhaps proteins localized to the leading edge of the engulfing membrane recruit C . difficile coat proteins but are insufficient to adhere the coat to the forespore in the absence of engulfment completion , or mechanical forces that drive engulfment to completion are also required to intimately associate the coat with the forespore . A link between SpoIIQ and coat localization around the forespore has been described in B . subtilis , since SpoIIQ is required for many coat proteins , including the σE-dependent coat protein CotE ( unrelated to C . difficile CotE [47] ) , to surround the forespore in a process known as “encasement” [69] . Since B . subtilis CotE localizes properly in a sigG mutant [69] , the B . subtilis spoIIQ mutant’s ~30-fold encasement defect suggests that components of the coat indirectly interact with the forespore-localized SpoIIQ . It should be noted , however , that CotE in a B . subtilis spoIIQ mutant appears to track along the mother cell-forespore interface [69] , in contrast with C . difficile spoIIQ , spoIIIA , and spoIIIAH mutants in which coat is located some distance from this interface due to an apparent defect in adhering to the forespore ( Figs 3 and 4 ) . Furthermore , the B . subtilis ΔspoIIQ mutant completes engulfment in the conditions used for the coat localization studies , whereas the C . difficile spoIIQ mutant largely fails to complete engulfment ( Figs 3 , S7 and S9 ) . Since B . subtilis CotE localization around the forespore depends upon earlier morphogenetic proteins SpoVM , SpoIVA , and SpoVID [69] , it would be interesting to determine whether B . subtilis “feeding tube” components affect the localization of these earlier morphogenetic proteins , and vice versa . Similarly , C . difficile CotE is a σK-regulated protein that appears to localize to the outermost layers of C . difficile spores [46] , so determining the localization patterns of SpoIVA and/or SipL in the spoIIQ , spoIIIA , and spoIIIAH mutants may provide insight into whether SpoIIQ and/or SpoIIIAA-AH regulate the localization of these coat morphogenetic proteins . Future studies evaluating whether these proteins form a channel in C . difficile , why these proteins are important for forespore integrity , and how these proteins regulate engulfment and coat association with the forespore will provide much-needed insight into how these cellular processes are controlled in C . difficile and potentially other spore-forming organisms .
All C . difficile strains are listed in Table 1 and derive from the parent strain JIR8094 , an erythromycin-sensitive derivative [70] of the sequenced clinical isolate 630 [71] . C . difficile strains were grown on solid brain heart infusion media supplemented with yeast extract ( BHIS: 37 g brain heart infusion , 5 g yeast extract , 0 . 1% ( w/v ) L-cysteine , 15 g agar per liter ) [72] . Taurocholate ( TA; 0 . 1% w/v ) , thiamphenicol ( 5–10 μg/mL ) , kanamycin ( 50 μg/mL ) , cefoxitin ( 16 μg/mL ) , FeSO4 ( 50 μM ) , and/or erythromycin ( 10 μg/mL ) were used to supplement the BHIS media as indicated . Cultures were grown at 37°C , under anaerobic conditions using a gas mixture containing 85% N2 , 5% CO2 , and 10% H2 . Sporulation was induced on media containing BHIS and SMC ( 90 g BactoPeptone , 5 g protease peptone , 1 g NH4SO4 , 1 . 5 g Tris base , 15 g agar per liter ) [73] , at 70% SMC and 30% BHIS ( 70:30 media , 63 g BactoPeptone , 3 . 5 g Protease Peptone , 11 . 1 g BHI , 1 . 5 g yeast extract , 1 . 06 g Tris base , 0 . 7 g NH4SO4 , 15 g agar per liter ) [45] . 70:30 agar ( supplemented as appropriate with thiamphenicol at 10 μg/mL ) was inoculated from a starter culture grown on solid media . 70:30 broth was made as stated above omitting the agar . HB101/pK424 strains were used for conjugations and BL21 ( DE3 ) strains were used for protein expression . E . coli strains were routinely grown at 37°C , shaking at 225 rpm in Luria-Bertani broth ( LB ) . Media was supplemented with chloramphenicol ( 20 μg/mL ) , ampicillin ( 100 μg/mL ) , or kanamycin ( 30 μg/mL ) as indicated . All strains are listed in S1 Table; all plasmids are listed in S2 Table; and all primers used are listed in S3 Table . For disruption of spoIIQ , spoIIIAA , and spoIIIAH , a modified plasmid containing the retargeting group II intron , pCE245 ( a gift from C . Ellermeier , University of Iowa ) , was used as the template . Primers used to amplify the targeting sequence from the template carried flanking regions specific for each gene target and are listed as follows: spoIIQ ( #1052 , 1053 , 1054 and 532 , the EBS Universal primer as specified by the manufacturer ( Sigma Aldrich ) ) , spoIIIAA ( #1049 , 1050 , 1051 and 532 ) , and spoIIIAH ( #1264 , 1265 , 1266 , and 532 ) . The resulting retargeting sequences were digested with BsrGI and HindIII and cloned into pJS107 ( a gift from J . Sorg , University of Texas A&M ) , a derivative of pJIR750ai ( Sigma Aldrich ) . The ligations were transformed into DH5α and confirmed by sequencing . The resulting plasmids were used to transform HB101/pK424 . To construct the spoIIQ complementation construct , primers #1177 and 1178 were used to amplify spoIIQ containing 106 bp of the upstream region using 630 genomic DNA as the template . To construct the spoIIQ H120A complementation construct , SOE primers #1177 and #1851 were used to generate a 5’ fragment ( 590 bp ) containing the H120A mutation; primers #1850 and #1178 were used for the 3’ SOE product using the IIQ complementation construct as a template . To construct the spoIIIA operon complementation construct , primers #1174 and 1175 were used to amplify 211 bp upstream of spoIIIAA and 9 bp downstream of spoIIIAH using 630 genomic DNA as the template . The spoIIIAH complementation construct was made using PCR splicing by overlap extension ( SOE , [74] ) . Primer pair #1174 and 1618 was used to amplify the 5’ SOE product , while primer pair #1617 and 1239 was used to amplify the 3’ SOE product . The resulting fragments were mixed together , and flanking primers #1174 and #1239 were used to generate a fragment corresponding to 211 bp of the spoIIIA upstream region fused to the spoIIIAH gene ( PspoIIIA-spoIIIAH ) . To construct the spoIIIA operon K167A complementation construct , SOE primers #1174 and #1432 were used to generate a 5’ fragment ( 590 bp ) containing the K167A mutation; primers #1431 and #1175 were used for the 3’ SOE product . The flanking primers #1174 and #1175 were used to amplify the K167A IIIA complementation construct . To construct the spoIIIA operon D244A complementation construct , SOE primers #1174 and #1853 were used to generate a 5’ fragment ( 590 bp ) containing the D244A mutation; primers #1852 and #1854 were used for the 3’ SOE product using the IIIA complementation construct as a template . The flanking primers #1174 and #1854 were used to amplify the D244A IIIA mutation insert , digested with NotI and SalI . The plasmid carrying the IIIA complementation construct was also digested with NotI/SalI and then gel purified to separate the plasmid backbone from the wildtype fragment . The D244A IIIA NotI/SalI fragment was ligated to the gel-purified cut vector . To construct the spoIIIA operon K167A/D244A complementation construct , SOE primers #1174 and #1853 were used to generate a 5’ fragment ( 590 bp ) containing the D244A mutation; primers #1852 and #1854 were used for the 3’ SOE product using the K167A complementation construct as a template . The flanking primers #1174 and #1854 were used to amplify the D244A IIIA mutation insert , digested with NotI and SalI , and ligated to the IIIA complementation construct digested with NotI/SalI as described earlier . All complementation constructs except for the D244A and K167A/D244A were digested with NotI and XhoI and ligated into pMTL83151 [44] digested with the same enzymes . To construct strains producing recombinant N-terminally truncated SpoIIQ and N-terminally truncated SpoIIIAH for antibody production , primer pairs #1568 and 1569 and #1566 and 1567 , respectively were used to amplify codon optimized spoIIQ and spoIIIAH genes lacking stop codons off template synthesized by Genscript . The spoIIQ expression construct deletes the sequence encoding the first 30 amino acids of SpoIIQ , while the spoIIIAH expression construct deletes the sequence encoding the first 33 amino acids of SpoIIIAH , which removes the membrane-tethering domains and improves the solubility of the proteins in E . coli . The resulting PCR products were digested with NdeI and XhoI , ligated to pET22b , and transformed into DH5α . The resulting pET22b-spoIIQ and pET22b-spoIIIAH plasmids were used to transform BL21 ( DE3 ) for protein expression . To construct the pET28a-HA-spoIIIAH construct for the affinity co-purification studies , primer pair #1665 and 1614 was used on the codon-optimized spoIIIAH template synthesized by Genscript . To construct the pET28a-HA-spoVT construct for the affinity co-purification studies , primer pair #1691 and 1313 was used to amplify spoVT encoding an N-terminal HA-tag using C . difficile genomic DNA as the template . The resulting PCR products were digested with NcoI and XhoI , ligated to pET28a digested with the same enzymes , and transformed into DH5α . The pET28a-HA-spoIIIAH construct was transformed into BL21 ( DE3 ) to construct strain #1378 . The pET28a-HA-spoIIIAH construct was transformed into BL21 ( DE3 ) to construct strain #1404 . To construct the σG-dependent transcriptional reporter , the σG-regulated promoter of sspA ( PsspA ) was fused to a C . difficile codon optimized SNAP-tag [75] to generate PsspA-SNAP ( Genscript ) with flanking restriction sites . This promoter region has previously been described [28] . The plasmid was transformed into E . coli DH5α , isolated , and digested with NotI and XhoI then cloned into the complementation plasmid pMTL84151 , transformed into E . coli HB101 ( S1 Table ) and subsequently conjugated into C . difficile strains . C . difficile strains were constructed using TargeTron-based gene disruption as described previously ( S4 Fig , [27] ) . TargeTron constructs in pJS107 were conjugated into C . difficile using an E . coli HB101/pK424 donor strain . HB101/pK424 strains containing the appropriate pJS107 construct were grown aerobically to exponential phase in 2 mL of LB supplemented with ampicillin ( 50 μg/mL ) and chloramphenicol ( 10 μg/mL ) . Cultures were pelleted , transferred into the anaerobic chamber , and resuspended in 1 . 5 mL of late-exponential phase C . difficile JIR8094 cultures ( grown anaerobically in BHIS broth ) . The resulting cell mixture was plated as seven 100 μL spots onto pre-dried , pre-reduced BHIS agar plates . After overnight incubation , all growth was harvested from the BHIS plates , resuspended in 2 . 5 mL pre-reduced BHIS , and twenty-one 100 μL spots per strain were plated onto three BHIS agar plates supplemented with thiamphenicol ( 10 μg/mL ) , kanamycin ( 50 μg/mL ) , and cefoxitin ( 16 μg/mL ) to select for C . difficile containing the pJS407 plasmid . After 24–48 hrs of anaerobic growth , single colonies were patched onto BHIS agar supplemented with thiamphenicol ( 10 μg/mL ) , kanamycin ( 50 μg/mL ) , and FeSO5 ( 50 μM ) to induce the ferredoxin promoter of the group II intron system . After overnight growth , patches were transferred to BHIS agar plates supplemented with erythromycin ( 10 μg/mL ) for 24–72 hrs to select for cells with activated group II intron systems . Erythromycin-resistant patches were struck out for isolation onto the same media and individual colonies were screened by colony PCR for a 2 kb increase in the size of spoIIQ ( primer pair #1074 and 1075 ) , spoIIIAA ( primer pair #1302 and 1176 ) , and spoIIIAH ( primer pair #1301 and 1239 ) ( S4 Fig ) . HB101/pK424 donor strains carrying the appropriate complementation construct were grown in LB containing ampicillin ( 50 μg/mL ) and chloramphenicol ( 20 μg/mL ) at 37°C , 225 rpm , under aerobic conditions , for 6 hrs . C . difficile recipient strains spoIIQ– , spoIIIAA– , and spoIIIAH−containing group II intron disruptions , were grown anaerobically in BHIS broth at 37°C with gentle shaking for 6 hrs . HB101/pK424 cultures were pelleted at 2500 rpm for 5 min and the supernatant was removed . Pellets were transferred to the anaerobic chamber and gently resuspended in 1 . 5 mL of the appropriate C . difficile culture . The resulting mixture was inoculated onto pre-dried , pre-reduced BHIS agar plates , as seven 100 μL spots for 12 hrs . All spots were collected anaerobically and resuspended in 1 mL PBS . The resulting suspension was spread onto pre-dried , pre-reduced BHIS agar plates supplemented with thiamphenicol ( 10 μg/mL ) , kanamycin ( 50 μg/mL ) , and cefoxitin ( 10 μg/mL ) at 100 μL per plate , five plates per conjugation . Plates were monitored for colony growth for 24–72 hrs . Individual colonies were struck out for isolation and analyzed for complementation by phase contrast microscopy , Western blot analysis and transmission electron microscopy . A minimum of two independent clones from each complementation strain was phenotypically characterized . For the SNAP-tag expression constructs , a pMTL84151 [44] or pMTL84121 [28] plasmid backbone was used . The complementation protocol was followed as described except that after spots were collected from overnight growth on BHIS plates , 100 μL of the resulting PBS suspension was spotted 7 times onto a BHIS agar plate supplemented with thiamphenicol ( 10 μg/mL ) , kanamycin ( 50 μg/mL ) , and cefoxitin ( 16 μg/mL ) . This procedure was repeated for three plates . C . difficile strains were grown from glycerol stocks on BHIS plates supplemented with TA ( 0 . 1% w/v ) , or with both TA and thiamphenicol ( 5–10 μg/mL ) for strains with pMTL83151-derived or pMTL84151-derived plasmids ( as previously described [27] ) . Cultures grown on BHIS agar plates were then used to inoculate 70:30 agar plates ( with thiamphenicol at 5–10 μg/mL as appropriate ) for 17–24 hrs depending on the assay . Sporulation induced lawns were harvested in PBS , washed once , resuspended in PBS , visualized by phase contrast microscopy , and/or further processed for analysis by transmission electron microscopy , Western blotting , or fluorescence microscopy . C . difficile strains were induced to sporulate as described above , and cells were harvested in 1 . 0 mL PBS , and split into two tubes . One tube was heat shocked at 60–65°C for 25 minutes . Both heat-shocked and non-heat shocked cells were serially diluted , and cells were plated on pre-reduced BHIS-TA plates . After 20 hrs on BHIS-TA , colonies were counted , and cell counts were determined . The percent of heat-resistant spores was determined based on the ratio of heat-resistant cells to total cells , and sporulation efficiencies were determined based on the ratio of heat-resistant cells for a strain compared to wild type . Results are based on a minimum of three biological replicates . One hundred microliters of bacterial cell suspension samples from sporulation assays were prepared as previously described [45] . The anti-SpoIIQ and anti-SpoIIIAH antibodies used in this study were raised in rabbits by Cocalico Biologicals ( Reamstown , PA ) . The antigens SpoIIQ-His6 and SpoIIIAH-His6 were purified on Ni2+-affinity resin from E . coli strains #1301 and 1302 as described above . Cultures were grown and protein expression was analyzed as previously described [27] . Sporulation assay C . difficile cells ( 50 μL of PBS suspension ) were freeze-thawed three times , diluted in 100 μL EBB buffer ( 8 M urea , 2 M thiourea , 4% ( w/v ) SDS , 2% ( v/v ) β-mercaptoethanol ) , and incubated at 95°C for 20 min with vortexing every 5 min . Samples were centrifuged for 5 min at 15 , 000 rpm , and 7 μL of 4X sample buffer ( 40% ( v/v ) glycerol , 1 M Tris pH 6 . 8 , 20% ( v/v ) β-mercaptoethanol , 8% ( w/v ) SDS , and 0 . 04% ( w/v ) bromophenol blue ) , was added . Protein samples were incubated again at 95°C for 15 minutes with vortexing followed by centrifugation for 5 min at 15 , 000 rpm . SDS-PAGE gels ( 12%–15% ) were loaded with 5 μL of the sample . Gels were transferred to Bio-Rad PVDF membrane and blocked in 50% PBS:50% Odyssey Blocking Buffer with 0 . 1% ( v/v ) Tween for 30 min at RT . Polyclonal rabbit anti-SpoVT ( [27] anti-SpoIIQ and anti-SpoIIIAH , antibodies were used at a 1:1 , 000 dilution . Monoclonal mouse anti-Spo0A [27] was used at a 1:10 , 000 dilution . Monoclonal mouse anti-SNAP ( NEB ) was used at a 1:2 , 000 dilution . IRDye 680CW and 800CW infrared dye-conjugated secondary antibodies were used at a 1:20 , 000 dilutions . The Odyssey LiCor CLx was used to detect secondary antibody fluorescent emissions for Western blots . RNA from WT , spo0A– , sigF– , spoIIQ– , sigE– , spoIIIA– , spoIIIAH– , and sigG−strains grown for 17 hrs on 70:30 sporulation media was extracted for qRT-PCR analyses of spoIIIAA transcript . RNA from WT , spo0A– , sigF– , spoIIQ– , sigE– , spoIIIA– , and sigG−strains grown for 25 hr on 70:30 sporulation media was extracted for qRT-PCR analyses of spoVT , CD1430 , and spoVAD transcripts . RNA from WT/EV , spo0A–/EV , spoIIIAA–/EV , spoIIIA–/spoIIIA operon , and spoIIIA–/spoIIIAK167A operon complementation strains grown for 17 hr on 70:30 sporulation media was extracted for qRT-PCR analyses of spoIIIAA transcript . RNA was extracted using a FastRNA Pro Blue Kit ( MP Biomedical ) and a FastPrep-24 automated homogenizer ( MP Biomedical ) . Contaminating genomic DNA was depleted using three successive DNase treatments and mRNA enrichment was done using an Ambion MICROBExpress Bacterial mRNA Enrichment Kit ( Invitrogen ) . Samples were tested for genomic DNA contamination using quantitative PCR for rpoB . Enriched RNA was reverse transcribed using Super Script First Strand cDNA Synthesis Kit ( Invitrogen ) with random hexamer primers . Transcript levels of spoIIIAA and rpoB ( housekeeping gene ) were determined from cDNA templates prepared from 3 biological replicates of WT , spo0A– , sigF– , spoIIQ– , sigE– , spoIIIA– , spoIIIAH– , and sigG−and three biological replicates of WT/EV , spo0A–/EV , spoIIIA–/EV , spoIIIA–/spoIIIA operon , and spoIIIA–/spoIIIAK167A operon . Gene-specific primer pairs for spoIIIAA and rpoB have been previously described [29 , 75] . Transcript levels of spoVT , CD1430 , spoVAD , and rpoB were determined from cDNA templates prepared from three biological replicates of WT , spo0A– , sigE– , spoIIIA– , and sigG– . Transcript levels of CD1430 and spoVAD were analyzed using gene-specific primer pairs #1458 and 1459 , #1708 and 1709 , respectively . Gene-specific primers for measuring spoVT transcript levels have been previously described [27] . Quantitative real-time PCR was performed ( as described by [75] ) . Briefly , using SYBR Green JumpStart Taq Ready Mix ( Sigma ) , 50 nM of gene specific primers , and an ABI PRISM 7900HT Sequence Detection System ( Applied Biosystems ) . Transcript levels were normalized to the housekeeping gene rpoB using the standard curve method and calculated relative to either the spo0A– strain or spo0A– strain carrying empty pMTL83151 vector . The CotE-SNAP previously described by Pereira et al . [28] was transformed into E . coli HB101/ pK424 and conjugated into the indicated C . difficile strains to analyze coat localization in spoIIQ and spoIIIA mutants . C . difficile strains containing SNAP-tag reporters were grown on 70:30 media to induce sporulation . Cells were grown as a lawn for 21 hours on solid 70:30 media and harvested as described by Pereira et al . [28] . Briefly , cells were harvested in PBS and pelleted ( 4 , 000 rpm for 3 min ) , washed once with PBS , reconstituted in 100 μL of PBS . TMR-star SNAP substrate ( NEB ) was added to a final concentration of 3 μM to each tube and cells were incubated for 30 min at 37°C . Cells were pelleted , washed 3 times with PBS , and resuspended in PBS . Hoechst 33342 ( 10 mg/ml ) was added to a final concentration of 15 μg/ml and FM4-64 ( 200 μg/ml ) was added to a final concentration of 1 μg/mL . Strains were harvested from 70:30 plates after 14 hours of growth as a bacterial lawn and re-suspended in 3 mL of 70:30 liquid media . For each strain used , the culture was split into 2 tubes for two conditions , each containing 1 . 5 mL of culture . Alkyne D-alanine or D-alanine ( ACROS Organics ) was added to each tube , respectively , at a final concentration of 2 . 5 mM and incubated at 37°C for 30 min . with mild shaking . After incubation , cells were pelleted ( 8000 rpm for 3 min . ) and washed 3x with PBS . Cells were resuspended in 0 . 7 mL of 2% formaldehyde diluted in PBS and incubated for 10 min on the nutator . Cells were then pelleted and washed 2x with 1 mL PBS . Cells were incubated with 5 mg/mL lysozyme , 37°C , 45 min . , pelleted , washed 2x with 1 mL PBS , and washed once with 3% BSA ( in PBS ) . For the click chemistry reaction , a Click-iT Plus Alexa Fluor 488 Picolyl Azide Toolkit ( Molecular Probes ) was used according to the manufacturer’s instructions . After incubation with Click-iT reagents , samples were pelleted and washed 1x with 3% BSA and 1x with PBS . Samples were resuspended and Hoechst 33342 ( 10 mg/ml ) was added to a final concentration of 15 μg/ml . For the peptidoglycan labeling timecourse , cells were harvested into 5 . 5 ml of 70:30 broth after 14 hours of growth on 70:30 plates , and either alkyne D-alanine or D-alanine ( ACROS Organics ) was added to each tube . One mL samples of each culture was taken at every timepoint ( 0 , 10 , 20 , 30 , and 40 min ) and processed as described above . For evaluation of peptidoglycan labeling after treatment with antibiotics , 1 ml of WT cells in BHIS broth were harvested at late exponential phase for each treatment condition . 2X MIC of antibiotics ( 2 μg/ml vancomycin and 8 μg/ml imipenem ) was added to designated cells and mixed with mild shaking . Dala and alkDala was added immediately after to designated cells and incubated for 30 minutes with mild shaking in the anaerobic chamber . Cells were processed for peptidoglycan labeling as described above . The median fluorescent intensity ( MFI ) of alkDala incorporation was determined using a MACSQuant VYB flow cytometer . MACSQuantify software was used for data collection and FlowJo V . 10 . 0 . 8 was used for data analysis . Cells that incorporated Hoechst dye 33342 ( Molecular Probes ) were evaluated for alkDala staining based on fluorescence in the FITC channel . E . coli BL21 ( DE3 ) strains were grown to mid-log phase in 2YT ( 5 g NaCl , 10 g yeast extract , and 15 g tryptone per liter ) , 225 rpm , at 37°C . 250 μM isopropyl-β-D-1-thiogalactopyranoside ( IPTG ) was added to induce the cells followed by an overnight incubation at 18°C . Cultures were pelleted , resuspended in low-imidazole buffer ( 500 mM NaCl , 50 mM Tris [pH 7 . 5] , 15 mM imidazole , 10% [vol/vol] glycerol ) , and lysed by freeze-thawing and sonication . The insoluble material was pelleted , and the soluble fraction ( Input ) was batch affinity purified using Ni2+ affinity resin and eluted with high-imidazole buffer ( 500 mM NaCl , 50 mM Tris [pH 7 . 5] , 150 mM imidazole , 10% [vol/vol] glycerol ) . The resulting eluates were run on SDS-PAGE gels ( 12% ) and transferred onto a PVDF membrane for Western blot analysis , as described above . For live cell fluorescence microscopy studies , C . difficile strains were harvested in PBS , pelleted , and resuspended in PBS . For initial characterization of mutant phenotypes , cells were resuspended in PBS containing 1 μg/mL FM4-64 ( Molecular Probes ) and 15 μg/mL Hoechst 33342 ( Molecular Probes ) . All live bacterial suspensions ( 4 μL ) were added to a freshly prepared 1% agarose pad on a microscope slide , covered with a 22 x 22 mm #1 coverslip and sealed with VALAB ( 1:1:1 of vaseline , lanolin , and beeswax ) as previously described [27] . DIC and fluorescence microscopy was performed using a Nikon PlanApo Vc 100x oil immersion objective ( 1 . 4 NA ) on a Nikon Eclipse Ti2000 epifluorescence microscope . Multiple fields for each sample were acquired with an EXi Blue Mono camera ( QImaging ) with a hardware gain setting of 1 . 0 and driven by NIS-Elements software ( Nikon ) . Images were subsequently imported into Adobe Photoshop CS6 for minimal adjustments in brightness/contrast levels and pseudocoloring . DIC and fluorescence microscopy for cells that were processed for peptidoglycan labeling experiments were performed and processed with the same equipment as described above with the following differences: a Nikon PlanApo Vc 60x oil immersion objective ( 1 . 4 NA ) was utilized , hardware gain setting 2 . 0 , and fields were imaged with Z-spacing of 0 . 15 μm followed by deconvolution using AutoQuant 3x software ( MediaCybernetics ) . Quantification of total cells undergoing sporulation was determined by analyzing multiple fields for each strain at random . At least 50 cells were enumerated for each strain . Sporulating cells were identified as either having a polar septum with or without DNA staining in the forespore , a DIC-dark forespore with or without DNA staining in the forespore compartment , a DIC-bright forespore without DNA staining , or a free spore ( no mother cell compartment ) .
|
The bacterial spore-forming pathogen Clostridium difficile is a leading cause of nosocomial infections in the United States and represents a significant threat to healthcare systems around the world . As an obligate anaerobe , C . difficile must form spores in order to survive exit from the gastrointestinal tract . Accordingly , spore formation is essential for C . difficile disease transmission . Since the mechanisms controlling this process remain poorly characterized , we analyzed the importance of highly conserved secretion channel components during C . difficile sporulation . In the model organism Bacillus subtilis , this channel had previously been shown to function as a “feeding tube” that allows the mother cell to nurture the developing forespore and sustain transcription in the forespore . We show here that conserved components of this structure in C . difficile are dispensable for forespore transcription , although they are important for completing forespore engulfment and retaining the protective spore coat around the forespore , in contrast with B . subtilis . The results of our study suggest that targeting these conserved proteins could prevent C . difficile spore formation and thus disease transmission .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Regulation of Clostridium difficile Spore Formation by the SpoIIQ and SpoIIIA Proteins
|
Axon-guidance by Slit-Roundabout ( Robo ) signaling at the midline initially guides growth cones to synaptic targets and positions longitudinal axon tracts in discrete bundles on either side of the midline . Following the formation of commissural tracts , Slit is found also in tracts of the commissures and longitudinal connectives , the purpose of which is not clear . The Slit protein is processed into a larger N-terminal peptide and a smaller C-terminal peptide . Here , I show that Slit and Slit-N in tracts interact with Robo to maintain the fasciculation , the inter-tract spacing between tracts and their position relative to the midline . Thus , in the absence of Slit in post-guidance tracts , tracts de-fasciculate , merge with one another and shift their position towards the midline . The Slit protein is proposed to function as a gradient . However , I show that Slit and Slit-N are not freely present in the extracellular milieu but associated with the extracellular matrix ( ECM ) and both interact with Robo1 . Slit-C is tightly associated with the ECM requiring collagenase treatment to release it , and it does not interact with Robo1 . These results define a role for Slit and Slit-N in tracts for the maintenance and fasciculation of tracts , thus the maintenance of the hardwiring of the CNS .
In the Drosophila embryonic ventral nerve cord , about 20 longitudinal axon tracts traverse up and down the nerve cord to connect all hemisegments on either side of the midline . These tracts are inter-connected across the midline by the commissural tracts , which cross the midline only once and never re-cross . The longitudinal tracts , together with those commissural tracts that cross the midline , form the longitudinal connectives on either side of the midline . Pathfinding of longitudinal and commissural tracts have been studied in detail [1–11] . It is well documented that signaling pathways such as Slit-Robo [1–7] or Netrin-Frazzled [8–11] , guide growth cones of these tracts to their synaptic targets . In the absence of these signaling cues , growth cones follow aberrant routes from the very beginning of their journey . Slit-Robo signaling is the main system that mediates pathfinding of growth cones for the longitudinal tracts [3–7] . In Drosophila , the slit gene is transcribed in the midline glia , and the protein is present in the midline glia [1–3] , whereas its receptors , Robo1 , Robo2 , and Robo3 are present in a combinatorial manner in axon growth cones [4–6] . The interaction between Slit and Robo mediates proper projection of growth cones on either side the midline parallel to each other . A loss of function for slit early during neurogenesis , for instance , causes the projection of pCC , a pioneering growth cone for the medial longitudinal tract , to head tangentially towards the midline [3 , 7] . In older slit mutant embryos , all longitudinal tracts are collapsed at the midline . Loss of function for robo genes also causes collapsing of tracts at the midline or on to each other [4–7] . While the role of guidance molecules in axon pathfinding has been well-explored , it is less so whether these molecules are also required for maintaining the position of axon tracts . A previous study had explored the role of Slit on the fasciculation of tracts and their spacing in the mouse diaphragm [12] , but not much is known if the position of tracts and their fasciculation are maintained within the CNS . By examining a mutation in the enzyme involved in the glycosylation of Slit , we have recently shown that Slit-Robo signaling contributes to the maintenance of axon tracts [13] . This issue of maintenance of tracts has major implications since it is essential for maintaining the integrity of the hardwiring of the nervous system , thus , neural function , or dysfunction later in development or in life . While Slit is expressed in the midline glia [1–3] , Slit is also found in tracts [7 , 13 , see also 1–3 , 14] . We ruled out Slit in tracts as cross-reactivity to the antibody or a background staining and showed that tracts-Slit is transported from the midline to the longitudinal connectives along the commissural tracts [7 , 13] . Since both Slit and Robo are present in tracts of the longitudinal connectives , it is most likely that they would interact with each other in tracts to mediate certain function . The Slit protein is processed into a larger N-terminal peptide and a smaller C-terminal peptide . It has been suggested that the C-terminal peptide interacts with PlexinA1 to mediate commissural axon guidance , and the N-terminal peptide interacts with Robo to mediate guidance of longitudinal tracts [15 , 16] . Because of a serendipitous discovery that the expression pattern of Slit is affected in late stage patched ( ptc ) mutant embryos , I sought to explore the post-guidance function of Slit in the Drosophila ventral nerve cord using mutations in ptc . Ptc was originally identified as a major segmentation gene , but it also regulates neurogenesis both in flies and vertebrates [17–21] . In this study , I report a novel effect of loss of function for ptc on Slit in the CNS of older stage embryos . This effect was accompanied by a novel nerve cord defect , which correlated with the changes in the expression of Slit and its localization in tracts . Ptc , together with mutations in another axon guidance molecule , Commissureless ( Comm ) , which is involved in the down-regulation of Robo1 in commissural tracts at the midline [22–24] , allowed me to examine the role of Slit in the maintenance of position and fasciculation of tracts . I also found that Robo1 binds to both full-length Slit and Slit-N in tracts but not to Slit-C . Slit and Slit-N are not freely found in the extracellular milieu , but are associated with the ECM . Compared to Slit or Slit-N , Slit-C appears to be more tightly associated with the ECM of axon tracts requiring collagenase treatment to release it . These results reveal novel insights into this important signaling system in the CNS .
Ptc is involved in multiple developmental and disease processes [17–21] . We have previously reported that misspecification of the identity of neurons that send pioneering axons is responsible for some of the axon guidance defects in ptc mutant embryos [20] . A closer examination of the ventral nerve cord in ptc loss of function embryos showed a novel nerve cord defect in embryos that were older than 13 hours post fertilization ( hpf; at 22 0C ) . As shown in Fig 1 , Fasciclin II ( Fas II ) and BP102 staining of mutant embryos showed that axon tracts progressively moved towards the midline in the anterior-posterior direction resulting in a funneling phenotype ( panels B , D , F , see also Table 1 ) . This funneling defect was seen in 82+/-3 . 6% ( n = 30 embryos , N = 3 independent experiments ) . The difference in the distance between L-L tracts in ptc , but not in wild-type embryos , was statistically significant between the anterior and the posterior regions ( P<0 . 001; S1 Data ) . This defect was not seen in embryos that were younger than 12–13 hpf ( Fig 1A , 1C and 1E; Table 1 ) , but was seen in embryos older than 13 hpf ( Fig 1B , 1D and 1F; Table 1 ) . While the fasciculation of longitudinal tracts was more or less normal in younger stage ptc mutant embryos ( Fig 1E , upper images ) , in older embryos these tracts were frayed and de-fasciculated with no observable discreet bundles ( Fig 1E , lower image , arrowhead ) . The defasciculation defects were much more pronounced in the posterior segments compared to the anterior segments ( Fig 1G ) . These results suggest that in older ptc mutant embryos , tracts not only move towards the midline in the posterior region but also they de-fasciculate . The slit gene is transcribed in the glial cells at the midline [1 , 2; see Fig 2A , 2B and 2E ) . However , the protein is present in the midline glia as well as in axon tracts of the commissures and connectives ( Fig 2A and 2B ) [7 , 13 , see also ref . 1–3 , 14] . The Slit in tracts appears to be transported along the commissural tracts , with the protein profile extending from the midline source cells to the longitudinal connectives along the newly forming commissural tracts ( Fig 2A ) . This can be clearly seen with the ImageJ analysis of the staining . ImageJ analysis across an area that has the Slit-positive midline cells and the commissural tracts crossing these midline cells showed a midline peak that continues with a tiny dip before a smaller peak representing Slit in longitudinal tracts on either side of the midline ( Fig 2A and 2B ) . A similar ImageJ analysis across the midline but between the anterior commissure ( AC ) and the posterior commissure ( PC ) ( between neuromeres ) also showed a midline peak and two smaller peaks corresponding to the Slit in longitudinal tracts , but the dip in between the midline and the longitudinal tracts peaks reached the baseline ( Fig 2A and 2B ) . This dip was due to the absence of axon tracts and consequently the absence of any Slit in that region . ImageJ analysis of the slit-mRNA staining showed a single peak at the midline ( Fig 2A ) . In older ~15 hpf embryos , while the mRNA expression was restricted to the midline cells , a clear ladder-like Slit protein in tracts was seen ( Fig 2B ) . The continuous domain of Slit along the midline now was resolved into two distinct domains , one located at the AC and the other at the PC in each neuromere/segment ( Fig 2B ) . Consistent with Slit being spread to the longitudinal tracts along the AC and PC tracts , Slit was seen in AC and PC but not in between ( Fig 2B ) . ImageJ analysis across the midline and at AC or PC showed a midline Slit-peak , and two peaks on either side of the midline corresponding to Slit in longitudinal tracts ( Fig 2B ) . Whereas analysis across the midline but between neuromeres/segments , showed a dip at the midline because of lack of midline Slit , but only two Slit peaks in longitudinal tracts ( Fig 2B ) . The spreading of Slit from the midline to the tracts along the commissural tracts can be seen with immunofluorescent labeling of Slit in ~14 . 5 hpf old embryos ( Fig 2C ) . The Slit protein was seen strictly along the tracts but not in the areas between AC , PC , and LC , where no tracts are present . A gradient of Slit extending from the midline source to the periphery across the nerve cord was not observed . Staining embryos with an antibody against Slit-N also showed the presence of Slit in tracts and did not reveal any Slit-gradient ( Fig 2D ) . These results show that outside of the midline , Slit is found only in tracts . Consistent with these findings , in embryos mutant for comm , where commissures are mostly absent or greatly reduced [23 , 24 , 7] , Slit was present in the midline but mostly absent or greatly reduced in longitudinal connectives [7] . Thus , commissural tracts appear to play the conduit role in moving Slit from the midline to longitudinal connectives . The expression of Slit in ptc mutant embryos was not affected during the early stages of axon guidance [20 , see Fig 2E ) . However , in older ptc mutant embryos the expression decayed in a highly specific manner ( Fig 2F ) . The mutant embryos that were 14 hpf or older had slit mRNA present in the anterior midline region of the nerve cord , but not in the posterior region ( Fig 2F; 83%+/-11% of the embryos , n = 30 embryos/experiment , N = 3 separate experiments ) . Consistent with this pattern of slit RNA expression , in about the same percentage of embryos , the protein at the midline was restricted to the anterior region in these older ptc mutant embryos ( Fig 2F ) . Moreover , Slit in tracts also followed the midline pattern with the protein present at higher levels in the anterior region and progressively decreased towards the posterior region ( Fig 2F ) . In the remaining ~17% +/-11% of the embryos , a few slit-expressing midline cells could be found in the posterior region or interspersed along the midline ( Fig 2G ) , and about the same percentage of embryos also had Slit protein distributed in a similar fashion in the midline as well as in tracts ( Fig 2G ) . These results indicate that midline is the source of Slit in tracts and the presence of Slit in tracts mirrors Slit expression in the midline . The staining of older stage ptc mutant embryos for Slit also revealed the A-P funneling defect of axon tracts in ptc mutant embryos ( Fig 2F and 2G ) . In these embryos , the gradual disappearance of Slit followed a gradual funneling of tracts ( Fig 2F ) . In those instances where the disappearance of Slit was dispersed or random , the narrowing of tracts followed the loss or reduction of Slit in the midline and in tracts ( Fig 2G ) . These results suggest that Slit is essential for maintaining the position of tracts away from the midline in a parallel trajectory in older stage embryos . Unlike the loss of Slit expression , the abundance of Robo1 in tracts in ptc mutant embryos was uniform along the A-P axis of the nerve cord ( Fig 3A; n = 30 embryos; N = 3 experiments ) , indicating that the limiting factor for the tracts positioning in ptc was Slit and not Robo . Since the Comm protein down-regulates Robo1 only in commissural tracts at the midline , the three Robo1-positive longitudinal tracts ( Medial or M , Intermediate or I , and lateral or L tracts ) that crossed the midline at AC and PC in ptc mutants had Robo1 in tracts at the midline in these regions . Note that the inappropriate midline crossing defects of longitudinal tracts in ptc mutants did not show any specific differences between anterior and posterior regions and were more or less the same in all segments . These results indicate that the funneling defect and the midline-crossing defect are independent of each other . It is also consistent with the fact that the midline-crossing defect is due to the misspecification of NBs and neurons and are ptc-dependent but slit-independent [20] . Next I sought to determine if the loss of Slit expression in older ptc mutant embryos is responsible for the funneling defect and separate it from axon guidance defect that occurs equally in all segments . A UAS-slit transgene and the midline driver single-minded ( sim ) -GAL4 were introduced to a ptc null mutant background and the slit transgene was expressed only in the midline at different time points during development . A transient expression of Slit in the midline was induced in embryos raised at 16 . 5 0C by shifting them to 22 0C ( GAL4 does not or very weakly induces UAS-linked transgenes in 16 . 5 0C but induces at higher levels in 22 0C or above ) . Induction between 11 and 13 hpf ( Fig 3E ) resulted in Slit expression at the midline as well as its presence in axon tracts ( Fig 3B ) . This was sufficient to rescue the funneling phenotype of longitudinal tracts in ptc mutant embryos ( Fig 3B–3E ) . The rescue was seen in 81%+/-9% of embryos ( n = 30 embryos per experiment; N = 3 experiments ) . A shift between 4–13 hpf also rescued the funneling phenotype ( 91%+/-12% , n = 20 , N = 3 ) , the later shift at 15 hpf , however , did not rescue it ( n = 30; N = 3 ) . Measurement of the distance between L-L tracts across the midline in rescued embryos showed that the distance was the same in the anterior and the posterior region . These results indicate that Slit has a direct role in maintaining the position of tracts along the midline , in addition to the initial growth cone guidance [see also ref . 13] . Moreover , the midline expression of the slit gene in ptc mutant embryos improved the fasciculation defects as well ( Fig 3D ) . While the tracts still crossed the midline , which is Slit-independent , individual tracts could be seen in rescued embryos as opposed to in non-rescued ptc mutant embryos where the tracts are mostly de-fasciculated and frayed ( Fig 3D ) . These results suggest that Slit in tracts maintains the fasciculation of axon bundles in longitudinal connectives . A previous paper had reported that very few midline glial cells exist in older ptc mutant embryos [25] . However , we had previously found that ptc mutant embryos had midline cells even at 14 hpf [20] . A re-examination of ptc null embryos by staining with an antibody against Sim , a midline marker [26] , showed that ptc mutant embryos had Sim-positive midline cells except for areas with small gaps and clustering of cells ( Fig 3F ) . One explanation for the discrepancy between these results and that of Hummel et al [25] is that Hummel et al examined an unknown and uncharacterized enhancer-trap line that had failed to complement a ptc allele , unlike the results of Merianda et al [20] or shown here , where well-known and well-characterized ptc alleles were used . Moreover , the midline analysis in Hummel et al [25] used another uncharacterized enhancer-trap line with a midline expression , which itself might be regulated by Ptc . We used Sim , a well-known regulator of fates of all midline cells [26] , in these studies . While the midline expression of the slit transgene in ptc mutant embryos rescued the funneling phenotype , it did not rescue axons aberrantly crossing the midline ( Fig 3C and 3D ) . This is because the midline-crossing phenotype is mainly due to the earlier events of misspecification of neurons that send out pioneering axons for these tracts [20] and are slit-independent . A midline expression of the slit transgene will not rescue this or any segmentation defects . These results argue that the gradual anterior-posterior narrowing of tracts in ptc is due to the corresponding gradual loss of Slit expression from the midline and in tracts . I also examined if the loss of function for ptc progressively reduces the number of neurons in the anterior-posterior direction . Immunostaining of embryos with an antibody against the Elav protein , a pan-neural marker , showed that the number of neurons in ptc embryos is not reduced in the anterior-posterior direction ( Fig 4A ) . However , the entire population of neurons in ptc embryos appeared to progressively move towards the midline with a tighter packing of neurons per hemisegment in the posterior region . This indicates that neurons , which are closely associated with axon tracts , also move towards the midline in ptc embryos . This is likely a secondary effect due to the movement of tracts towards the midline ( see Discussion ) . Embryos mutant for ptc were also examined with Repo , a glial marker . As shown in Fig 4B , there was no significant loss of glial cells , but the location of glial cells had shifted towards the midline in the posterior direction . Since glial cells are attached to the tracts , their movement appears to be the consequence of the movement of tracts . I also examined if a loss of neurons in the nerve cord could cause a shifting of tracts towards the midline . This was done by staining embryos that were mutant for the gene regulator of cyclin A1 ( rca1 ) . Rca1 regulates the expression of zygotic cyclin A and as in cyclin A mutants , in rca1 mutants also Ganglion Mother Cells ( GMCs; these are secondary precursor cells for neurons ) do not divide , but adopt the identity of one of their progeny neurons [27 , 28] . Thus , their nerve cord possesses significantly fewer neurons and glia than wild-type . As shown in Fig 4C , the longitudinal tracts across the midline in rca1 were not any closer , instead , they were further away compared to wild-type . This is likely due to the decrease in the number of neurons in rca1 . An anterior-posterior difference in the position of tracts was also not found in rca1 mutants . Finally , I examined mutants that appear to have the commissural defects similar to ptc to determine if commissural defects could cause the funneling defect . However , the funneling phenotype was not found in such mutants ( Fig 4D ) [see also ref . 23] . A loss of function for comm severely affects the formation of commissures , although they still have a few axons extending to the midline ( see Fig 5A ) . This loss of most of the commissures in comm is due to the upregulation of Robo in commissural growth cones and therefore their inability to overcome the Slit barrier at the midline [24] . These embryos suffer from a significant loss of Slit in tracts ( Fig 5B ) , indicating that proper commissural structures are required for Slit to travel to the longitudinal tracts [see also ref . 7] . Interestingly , we also noticed that in about 40% of comm mutant embryos ( N = 60 embryos ) , Slit was present at higher concentrations in tracts in the anterior region compared to the posterior region ( Fig 5C and 5D ) . This corresponded with a higher expression of Slit in the midline glial cells in the anterior region compared to the posterior region , where Slit-expression was lower or the slit-expressing cells were absent ( Fig 5C and 5D ) . ImageJ analysis of the anterior versus posterior regions of comm embryos for Slit staining quantified and confirmed that there was a significant reduction/loss of Slit in longitudinal connectives from the anterior to the posterior region ( Fig 5D; see also S1A Fig ) . As in ptc mutant embryos , in comm mutant embryos as well , a narrowing of tracts in the anterior-posterior direction was observed ( Fig 5C and 5D ) . This defect corresponded with the gradual reduction in the levels of Slit in tracts . Fas II staining of comm mutant embryos also revealed that while the longitudinal connectives/tracts were farther apart in comm mutant embryos , about twice the distance of wild-type ( see Table 1; S1 Data ) , the tracts were funneling down towards the posterior end ( Fig 5E; 52%+/-4% ) . This funneling defect was seen also with BP102 ( Fig 5A ) and Slit staining ( Fig 5C and 5D ) . On the other hand , in about 50% of the embryos ( N = 60 ) where the tracts had low levels of Slit all along the tracts , no recognizable funneling phenotype was seen ( Fig 5F ) . These results , therefore , provide a second mutational data showing a similar funneling defect as in ptc embryos and its correlation with the decrease in the abundance of Slit in tracts . Each of the longitudinal tracts occupies a specific position along the nerve cord . Slit in tracts appears to regulate the distance between these tracts . While Comm is present and is required in the midline and in commissural tracts spanning the midline , it is not present in longitudinal connectives [23 , 24] . In comm mutants , however , the positioning of longitudinal tracts was aberrant with intermingling of tracts with each other . Additionally , the tracts were defasciculated ( Fig 5E , 5F and 5G , see also S1B Fig ) . ImageJ analysis of such Fas II-stained nerve cord from comm mutant embryos illustrate the intermingling and defasciculation of tracts ( Fig 5G; see also S1B Fig ) . The longitudinal tracts in comm mutants are positioned farther away from the midline compared to wild-type . A loss of cohesion between hemineuromeres due to the absence of commissures could be the reason . However , it appears that the midline is severely affected in comm mutant embryos [29 , 30] . Staining of comm mutant embryos with an antibody against Sim ( Fig 6A ) , Fas II ( Fig 6B ) and sli RNA expression analysis ( Fig 6C ) showed that the midline was severely affected . The midline cells in comm failed to form a single line as in wild-type , but dispersed into two disorganized lines even at an early age ( Fig 6A ) . This early origin of the midline defect in comm affecting the position of tracts was indicated by the finding that in comm the initial position of the medial tract was placed farther away compared to wild-type ( Fig 6B , see also S2 Fig ) . This result argues against the possibility that the commissural positioning defect in comm is due to a loss of cohesion from the absence of commissural structures . Despite their aberrant position , tracts in comm indeed moved closer towards the midline in the posterior region , which strengthens the argument that Slit has a maintenance function , both the position of tracts as well as their fasciculation . In ptc mutant embryos that were ~16 hpf or older did not have slit transcription or the Slit protein in the midline ( Fig 7A ) . But , these embryos still had the Slit protein in axon tracts , with the highest amount in the anterior region of the nerve cord . The position of tracts in the anterior region was more or less normal ( Fig 7A ) . With the Slit in tracts gradually decreasing towards the posterior region , the tracts were also getting progressively narrower ( Fig 7A ) . By taking advantage of the expression pattern of Slit in these >16 hpf ptc mutant embryos , I sought to determine if the Slit in tracts physically interacts with the Robo in tracts , perhaps to regulate inter-tract spacing . If the Slit in tracts physically and locally interacts with the Robo1 in tracts , we should be able to pull down Robo1 with Slit in extracts from older stage ptc mutant embryos . Embryo extracts from >16 hpf old wild-type and ptc mutants were prepared and subjected to immunoprecipitation using an antibody against Slit-C ( which recognizes the full-length Slit as well ) . The immunoprecipitated proteins were resolved on a polyacrylamide gel and probed for the presence of Robo1 . If Robo1 is pulled down by Slit , given that Slit in these older stage ptc embryos is present only in tracts but not in the midline , the interaction between Slit and Robo1 must be occurring in tracts . As shown in Fig 7B , Robo1 was indeed pulled down by immunoprecipitation with anti-Slit in the extract from ~16 hpf ptc mutant embryos . The Slit protein is processed into an N-terminal fragment of~150 kDa Slit-N and a C-terminal fragment of ~40 kDa Slit-C . It has been proposed that Slit-N interacts with Robo to guide longitudinal tracts , whereas Slit-C interacts with PlexinA1 to mediate commissural tracts [15 , 16] . The results presented in Fig 7B show that an antibody raised against Slit-C pulls down Robo1 . This would suggest that Slit-C interacts with Robo1 . However , since this antibody against Slit-C also recognizes full-length Slit , it may be that the antibody pulls down Robo1 through full-length Slit . Therefore , I examined if the same immunoprecipitate complex also has the full-length Slit . As shown in Fig 7C , the immunoprecipitated complex contained full-length Slit , indicating that the full-length Slit interacts with Robo1 ( see also Fig 8 ) . However , no Slit-C was detected in the immunoprecipitate ( Fig 7C; Slit-C migrates around 40 kDa , see also Fig 8 ) . I have recently developed a procedure to detect proteins that are secreted in vivo in embryos ( see Materials and Methods ) . This method involves dissociating cells from embryos in M3 cell culture media with a Dounce homogenizer and analyzing the media and the cellular pellet for the protein in question using Western analysis . When the dissociation of cells was done using 6 strokes with the loose-fitting pestle in the Dounce homogenizer , the full-length Slit was readily recovered in the media ( Fig 8A ) . With 7 strokes , there appeared to be cell-lysis , as indicated by the presence of Tubulin in the media ( Fig 8A ) , although this Tubulin could also be from contaminating or broken axons . Dissociating cells with 4 strokes yielded no Slit in the media but only in the pellet ( Fig 8A ) , indicating that Slit is not freely present in the milieu . The extra mechanical dissociation force caused by 6 strokes appears to release the externalized Slit protein from the ECM into the media . The Slit-N peptide could also be readily detected in embryo extracts with an antibody raised against Slit-N ( Fig 8B ) , indicating that Slit-N is also loosely associated with the ECM . However , Slit-C was only rarely detected in embryo extracts ( Fig 8C ) or in the media with 4–7 strokes ( Fig 8A ) . Thus , Slit-C appears to be more tightly bound to ECM with its epitope perhaps buried within the ECM and not easily accessible/detectable . Only when the ECM somehow gets disrupted Slit-C could be detected in total extracts ( Fig 8C ) . To determine if Slit-C is tightly bound to ECM , I subjected the mechanically dissociated embryonic cells ( 6 strokes ) to collagenase treatment ( see Materials and Methods ) . Collagenase treatment is expected to disrupt the ECM . The supernatant was then subjected to Western analysis . I found that the treatment with collagenase consistently released Slit-C , and was readily detected in Western blots ( Fig 8D ) . This result argues that disruption of the ECM with collagenase releases Slit-C . Since in older stage ptc mutant embryos ( >15 hours old ) , the Slit protein is absent from the midline , but still present in tracts , I sought to determine if the tracts also contain Slit-C . Indeed , cells derived from >15 hours old ptc mutant embryos generated Slit-C but only when treated with collagenase ( Fig 8E ) . Anti-Slit-C pulls down Robo1 as well as Slit ( Fig 7B and 7C ) indicating that Slit-C pulls down the Slit-Robo1 complex . To further confirm this , I immunoprecipitated total cell extract from ~15 hours old wild-type embryos with anti-Robo1 antibody and determined if it pulls down full-length Slit . As shown in Fig 9A , anti-Robo1 pulled down full-length Slit , indicating that Robo1 physically interacts with full-length Slit . However , Slit-C was not detected in the immunoprecipitate ( Fig 9A ) . That Robo1 also complexes with Slit-N is indicated by the result that anti-Robo1 pulls down Slit-N ( Fig 9B ) . Since the absence of Slit-C in the immunoprecipitate could be due to the buried nature of Slit-C in the ECM , I treated total cell extracts with collagenase and then performed IP of this collagenase-treated extract with anti-Robo1 . The IP was then analyzed by Western analysis with the anti-Slit-C antibody . As shown in Fig 9C , analysis of the IP with anti-Slit-C showed that Robo1 does not pull down Slit-C . Thus , Slit-C is unlikely to interact with Robo1 , although it is possible that collagenase treatment somehow altered Slit-C such that it could not interact with Robo1 .
The results shown here argue that Slit in connectives maintains the position of tracts in reference to the midline and axonal fasciculation within each tract-bundle . This appears to occur via Slit interacting with Robo proteins locally in the tracts ( Fig 7B ) . These conclusions are reached based on the following findings . First , the pattern of decay of Slit expression and the funneling phenotype in ptc or comm embryos . A progressive loss of Slit in tracts in an anterior-posterior direction was accompanied by a progressive narrowing of axon tracts ( Figs 1B , 1D , 2F , 5A , 5C , 5D , 5E and 7A ) . In mutant embryos where the loss of Slit was interspersed , the narrowing of tracts corresponded to those regions with loss or reduction of Slit in tracts ( Figs 2G and 7A ) . Second , in ptc mutant embryos of ~16 hours of age or older , the midline expression of Slit was completely lost , but the anterior region of the nerve cord still had a high enough amounts of Slit in tracts ( Fig 7A ) . In such embryos , the tracts were maintained at their more or less correct position in the anterior region , but not in the posterior region where the midline Slit was long gone and the abundance of Slit in tracts was much reduced . The position of tracts was also shifted progressively like a funnel in these embryos ( Fig 7A ) . Third , while the tracts in comm mutant embryos were placed farther away from the midline compared to wild-type or ptc , which appears to be due to midline defects ( Fig 6 ) , whenever there was a loss of Slit in tracts , such embryos showed a narrowing of tracts ( Fig 5 ) . Thus , a continuous presence of Slit at the midline serves as a source for Slit in connectives ( see Fig 10 ) , and the Slit in connectives appears to be essential for maintaining the position of tracts . Since immunoprecipitation of extracts with anti-Slit-C from these older stage embryos pulled down Robo1 , Slit and Robo1 must physically interact with one another in tracts of the connectives . The funneling phenotype in ptc is unlikely due to a secondary , patterning defect or a general impact of defective commissures . Because , a loss of function for ptc affects patterning and other CNS defects ( segmentation and NB specification defects ) equally in all hemisegments , these defects are unlikely to cause the highly specific funneling phenotype . Would the aberrant commissures in ptc pull the longitudinal tracts mechanically closer to the midline ? Again , all of the commissures are equally affected in ptc embryos and this occurs well before any defect in Slit expression occurs . The funneling phenotype is seen only in regions where Slit expression is lost ( mostly towards the caudal end ) . Furthermore , mutations that affect commissures similar to the defects in ptc , do not cause the funneling phenotype seen in ptc ( Fig 4D ) . There does not appear to be any loss of neurons in ptc embryos specific to the posterior region either , only that cells are more tightly packed towards the posterior end of the nerve cord ( Fig 4A and 4B ) . Also , even a significant reduction in the number of neurons in the nerve cord does not cause longitudinal tracts to move towards the midline ( Fig 4C ) . Finally , the rescue experiment shows that the funneling phenotype in ptc could be rescued without rescuing the guidance or commissural defects ( Fig 3C , 3D and 3E ) , effectively separating the two and showing that axon guidance or commissural defects would not necessarily cause the funneling phenotype . This is consistent with our previous finding that axon guidance defects in ptc ( midline-crossing of longitudinal tracts ) are slit-independent but ptc-dependent , and mediated by the misspecification of neurons that pioneer these tracts [20] . Finally , a loss of Slit in tracts , but not in the midline in comm mutants causes longitudinal tracts to move towards the midline , towards each other and also become de-fasciculated ( Fig 5 ) , supporting the possibility that tracts Slit prevents both de-fasciculation and inappropriate movement of tracts towards the midline ( see also S1 Fig ) . In older ptc mutant embryos , both neurons and glia exhibited the funneling phenotype . Neurons and glia are an integral part of axon bundles , like grapes on a wine or beads on a rope , and keeping the tracts/bundles in a specific location within the nerve cord means keeping those neurons and the structure itself in place . The entire nerve cord in that sense is one continuous unit . Whatever that acts on tracts possibly affects the entire structure . Thus , the entire structure in ptc in the caudal region seems to move towards the midline , causing cells in the posterior region to pack more tightly . This phenomenon in ptc or comm may be analogous to the phenomenon of nerve cord retraction—the tracts and the entire nerve cord retracts and progressively gets shorter and shorter beginning with ~13 hours of development . This nerve cord retraction in all likelihood is mediated by shortening of tracts , pulling the entire structure towards the anterior end , thus leading to the condensation of the entire nerve cord . Since Robo1 co-immunoprecipitates only Slit and Slit-N , but not Slit-C ( B , C , Fig 9A and 9B ) , Robo1 must interact with Slit and Slit-N , but not with Slit-C . Also , Slit and Slit-N appear to be loosely associated with the ECM ( Fig 8A and 8B ) , whereas Slit-C is likely buried or tightly bound to ECM , perhaps inaccessible to Robo . These results are consistent with the finding that while Slit-N interacts with Robo to guide longitudinal tracts Slit-C interacts with PlexinA1 to regulate commissural guidance [15 , 16] . The results also show that the full-length Slit also interacts with Robo1 , not just Slit-N , thus potentially mediate axon guidance . Whether Slit versus Slit-N further refines Slit-Robo signaling is not known . The presence of Slit-C in tracts in ptc mutant embryos likely reflects perhaps its role in regulating guidance and maintenance of commissural axons . A following scenario emerges on how Slit-Robo combination could mediate axon guidance to define lateral positioning of longitudinal tracts , and later on regulate tracts position and fasciculation . Soon after its formation , a neuron generates neurites , which repeatedly extend and retract from the cell body . Out of several of these neurites , only one becomes an axon . The axonal growth cone starts to sample its environment while extending from the cell body towards the midline . In the fly embryonic nerve cord , the growth cones of longitudinal axons have Robo proteins , but the three Robo proteins are differentially expressed in growth cones of different tracts . Thus , growth cones of the medial tract have only Robo1 , the intermediate tract have Robo1 and Robo3 , and the lateral tract have Robo1 , Robo2 , and Robo3 ( Fig 10A ) . Thus , the lateral tract growth cones have the highest amounts of Robo proteins , the intermediate tract has the next highest and the medial tract has the lowest Robo . During early neurogenesis , these growth cones will explore their environment and when they reach the midline , the interaction between Slit at the midline and the combined Robo in growth cones specifies the initial position of axons in a Robo-concentration dependent manner . Because the highest combined amounts of Robo is in lateral growth cones , they occupy the lateral-most position . The lowest amounts of Robo in the medial growth cones will specify their position closest to the midline . In this scenario , increasing the levels of Slit at the midline will not alter the eventual positioning of tracts , but the availability of Robo proteins/their saturation would . This was indicated by the previous findings that gain of function for Robo2 and Robo3 proteins in tracts shifted them away from the midline [4–6] but the over-expression of Slit in the midline had no effect [3] . During the maintenance phase ( Fig 10B ) when the growth cones of these tracts have successfully fasciculated with each other , a Slit-Robo interaction can occur only in tracts since growth cones are no longer exploring their environment but are stably fasciculated . But a Slit-Robo1 interaction indeed occurs in tracts ( Fig 7B and 7C ) where both are present . As shown previously [7 , 13] , and in this report , Slit from the midline gets transported to the longitudinal connectives along the commissural tracts ( Fig 10B ) . The longitudinal connectives have both the longitudinal axon tracts as well as the commissural tracts that extend along the longitudinal tracts after crossing the midline ( Fig 10B ) . The Slit in commissural tracts , possibly embedded in the ECM , must then interact with Robo in longitudinal tracts to maintain the fasciculation of individual axons within each tract , as well as their position between and from the midline ( Fig 10B ) . The exact mechanism of Slit transport from the midline along the commissures to the tracts is not known . But , secretion of Slit is essential since loss of function for mummy , a gene that encodes an enzyme in the protein glycosylation pathway , and glycosylates Slit , prevents Slit secretion and causes an absence of Slit in tracts [13] . Consistent with a role for Slit in tracts , mutants in mummy show defasciculation and loss of distance between tracts and from the midline . Does Slit form a morphogen-gradient ? Given our results , it appears unlikely , but perhaps a gradient of a modified Slit ( and thus , unrecognized by the antibody ) or some type of a functional gradient is still a possibility . Our result that Slit-Robo signaling contributes to the maintenance of hardwiring of the CNS has far-reaching implications in deciphering the hardwired neural circuitry networks and their functional aspects in normal and abnormal conditions across organisms .
Standard genetics were used [31] . The following fly stocks were used: ptcIN108 , and ptcH84 which are phenotypically null alleles , ptcdeficiency [Df ( 2R ) ED1742] , slit2 , robo14 , robo1 deficiency [Df ( 2R ) BSC786] , a comm deficiency [Df ( 3L ) BSC845] , a deficiency for kuz [Df ( 2R ) IR52a-d] and a deficiency for Sdc [Df ( 2R ) 48] were from the Bloomington Stock center , comm5 , comm6 were from Mark Seeger and Guy Tear , UAS-slit was from the Goodman lab , rca1 from Nipam Patel , and sim-GAL4 ( on the 3rd chromosome ) was from Steve Crews . For wild-type , Canton-S and Oregon-R flies were used . Mutant chromosomes were balanced using green fluorescent protein ( GFP ) -marked chromosomes ( twi-GFP , CyO , Kr-GFP , CyO ) to enable the selection of homozygous mutant embryos . Mutant embryos were further identified using their mutant phenotypes in other tissues and lineages . To determine if the expression of the slit gene in the midline in ptc mutants restores the axon tract position defects in ptc mutants , UAS-slit was introduced to ptc mutant background and induced in the midline using sim-GAL4 . The UASxGAL4 induction system is sensitive to temperature since the GAL4 protein is not active at lower temperatures , especially at 16 . 5 0C . Therefore , embryos from the above “rescue” cross were allowed to develop at 16 . 5 0C for various developmental time points ( see Fig 3E ) before shifting to 22 0C , where the embryos were allowed to develop until they were fixed for staining or down-shifted to 16 . 5 0C and kept until they were fixed . The embryos were then examined for the distribution of Slit and axon guidance defects . Embryos kept at 16 . 5 0C grow about 40% slower compared to embryos at 22 0C . The timings and stages were then compensated by taking this difference into account while representing the results in Fig 3E . Moreover , when the embryos were shifted to 22 0C , they were ascertained for their correct stages of development by visualizing a subset of them under the microscope after permeabilizing with Halocarbon oil [see ref . 31] . Thus , for example , an 11 hpf embryo shift would mean that embryos were collected for 15 min at 16 . 5 0C , they were then allowed to develop at 16 . 5 0C for an equivalent of 11 hours at 22 0C , which is about 15 . 5 hours of time at 16 . 5 0C . To positively ascertain the developmental stages of embryos before shifting , about 50 embryos were removed from the collection at 15 . 5 hpf , immersed in Halocarbon oil , which allows to monitor the stages of development clearly , and the stages were confirmed . To be certain , only those batches of embryos that showed the correct developmental stages corresponding to development at 22 0C were shifted to 22 0C . Immunochemistry was performed as previously described [7 , 13 , 32] . Monoclonal and or polyclonal antibodies against the following proteins were used at the indicated concentrations: Fas II ( 1:20 , mouse monoclonal 1D4 , Developmental Studies Hybridoma Bank ( DSHB ) , Slit-C ( 1:20 , mouse , C555 . 6D , DHSB ) , Slit-N ( 1:4000 ) , Robo1 ( 1:3 , mouse , 13C9 , DHSB ) , Elav ( 1:4 , mouse , 9F8A9 , DHSB ) , Repo ( 1:4 , 8D12 , DHSB ) . The monoclonal antibody BP102 recognizes an uncharacterized epitope on axons was used at 1:4 ( AB_528099 , DHSB ) . For confocal microscopy , secondary antibodies conjugated to Cy5 ( rabbit , 1:400 , Invitrogen , A10523 ) , fluorescein isothiocyanate ( mouse , 1:50 , Invitrogen , 62–6511 ) , Alexa Fluor 488 ( rabbit or mouse , 1:300 , Invitrogen , A-11008 or A-11001 ) , or Alexa Fluor 647 ( rabbit or mouse , 1:300 , Invitrogen , A-21245 or A-21236 ) were used . For light microscopy , secondary antibodies conjugated to alkaline phosphatase ( AP; rabbit , 1:200 , Pierce , 31341 ) or horseradish peroxidase ( HRP; rabbit , 1:200 , Pierce , 31460 ) were used . Alkaline phosphatase was detected using 5-Bromo-4-chloro-3-indolyl-phosphate and nitro blue tetrazolium ( Promega , S3771 ) . HRP was detected with diamino-benzidine ( Sigma , D4418 ) . Whole-mount RNA in situ hybridization for slit was done as described previously using a Digoxigenin-labeled slit probe synthesized by PCR [20 , 32] and the color reaction was developed by AP reaction . Different genotypes were identified using appropriate markers and phenotypes . Western blot analysis was done as previously described [7 , 13 , 32] . Briefly , about 30 embryos of the specified age were collected and used for protein extraction . Embryos were collected on apple-juice agar plates , transferred to a mesh-wire basket , washed with running water , and the mutant embryos were selected ( absence of green excitation for GFP ) using a Zeiss microscope equipped with an Ultra-Violet ( UV ) light . The embryos were lysed in 40 μL extraction buffer ( 0 . 15 M NaCl , 0 . 02 M Tris pH = 7 . 5 , 0 . 001M EDTA , 0 . 001 M MgCl2 , 1% Triton-X-100 and 1X Protease Inhibitor Cocktail ) using a sonicator for one minute on ice in a 1 . 5 mL Eppendorf vial . A hand-held , sonicator ( Fisher Scientific ) equipped with a disposable pestle ( Fisher Scientific ) was used for sonication . The lysates were centrifuged for 5 minutes at 13 , 000 rpm in a microfuge ( Beckman ) . The supernatant was collected and 10 μL of 4X Laemelli buffer and 1 . 5 μL of the reducing agent ( Invitrogen ) were added , boiled in water for 10 minutes and cooled on ice . About 15 embryos-equivalent amounts were loaded per lane on a 4–12% pre-made SDS-PAGE gel ( Invitrogen ) . Two different primary antibodies recognizing Slit were used: Slit-N , which recognizes the N-terminal portion ( 1: 50000 ) [7]; Slit-C , which recognizes the C-terminal portion ( mouse , C555 . 6D , DHSB , 1:100 ) [3] . For Robo1 Westerns , mouse monoclonal 13C9 ( DHSB ) was used at 1:40 [32] . The chemiluminescent reaction kit ( Millipore ) was used to detect signals . The blots were re-probed with an antibody against Tubulin ( 1:4000 , Abcam ) to determine the loading control . About 200 wild-type and ptc mutant embryos , aged 16 hours of development , were homogenized in 37 . 5 μL of ice-cold lysis buffer [50mM HEPES ( pH 7 . 2 ) , 100mM NaCl , 1mM MgCl2 , 1mM CaCl2 , and 1% NP-40] . The lysates were incubated on ice for 30 minutes , centrifuged at 15 , 000X g for 30 minutes at 4°C . 30 μL of the supernatant was used as starting material for each IP reaction using the Catch and Release v2 . 0 Reverse Immunoprecipitation System ( Millipore #17500 ) . The columns were washed with 1X Wash buffer ( Millipore ) thrice ( 2000X g , 20 seconds ) and the IP reaction was set up by directly adding these ingredients to the column in the following order: 1X 435 μL of the wash buffer , 30 μL of the cell lysate , 25 μL of the antibody against Slit-C and 10 μL of the antibody capture affinity ligand . For immunoprecipitating with anti-Robo1 , the IP reaction was set up as follows: 405 μL of the wash buffer , 30 μL of the cell lysate , 50 μL of the monoclonal antibody against Robo1 and 10 μL of the antibody capture affinity ligand . The columns were then incubated overnight at 4 0C . The flow through was collected and the columns were washed three times with 1X Wash buffer ( 2000Xg , 20 seconds ) , and eluted with 60 μL of PBS-based denaturing elution buffer ( 2000Xg , 20 seconds ) . For equalizing salt concentration between the lysate and the IP samples , 8 μL of the lysate was added to 8 μL of the denaturing elution buffer ( Millipore ) , and 8 μL of the lysis buffer was added to 8 μL of the IP ( which is in the denaturing elution buffer ) . The proteins were then separated on a 4–12% SDS-PAGE and immunoblotted with a monoclonal against Robo1 ( 1:40 , DHSB ) , against Slit-C ( 1:100 ) or against Slit-N ( 1:40000 ) . Signal detection was by a chemiluminescent ECL reaction kit . The blots were re-probed with an antibody against Tubulin ( 1:4000 , Abcam ) to determine the loading control . I have recently developed a method by which one could detect secreted proteins within and outside of cells without having to culture cells . Briefly , about 75 embryos were collected in 500 μL of M3 insect cell culture medium . They were transferred to a 1 . 5 mL Dounce homogenizer . The cells from these embryos were dissociated using the looser fitting pestle and different number of strokes ( without twisting the pestle ) during homogenization . Cells quickly dissociate in M3 media , which is confirmed by visualizing an aliquot of the homogenate under a microscope . The homogenates were transferred to 1 . 5 mL Eppendorf tube and centrifuged at 4000x g for 5 min . The supernatant was collected into a Vivaspin 500 ( Sartorius ) ( molecular weight cutoff 100 kDa ) concentrator and microfuged at 15000x g for 17 min . The resulting ~30 μL of the media were then subjected to Western analysis for Slit . The pellet in the Eppendorf tube was washed once by gently re-suspending the cells in 500 μL of M3 media and microfuging at 4000x g for 5 min . The supernatant was discarded . The pellets ( containing the cells ) were then lysed using the Lysis buffer , and subjected to Western analysis for Slit . About 75 wild-type or ptc mutant embryos were collected directly in 500 μL of M3 insect cell culture medium . The cells were dissociated in a Dounce homogenizer ( using the looser fitting pestle ) by using 6 strokes ( without twisting the pestle ) during homogenization ( see Fig 8 ) . The homogenates were transferred to 1 . 5 mL Eppendorf tube and centrifuged at 4000x g for 5 min . The supernatant was removed and the cell-pellet was re-suspended in 100 μL of PBS , centrifuged at 4000x g for 5 min . The supernatant was discarded , and the pellet was re-suspended in 20 μL of PBS . Collagenase type VII ( 2 μL of 10 mg/mL solution ) was added to these cells and incubated at room temperature for 20 min . The cells were then centrifuged at 4000x g for 7 min , and the supernatant was collected for Western analysis . The cells-pellet was then re-suspended in lysis buffer , homogenized and then subjected to Western analysis for Slit . As a control , another batch of embryos was processed similarly but without adding collagenase . The pellets were then lysed using the Lysis buffer , and subjected to Western analysis for Slit . About 75 wild-type embryos were collected directly in 500 μL of M3 insect cell culture medium . The cells were dissociated in a Dounce homogenizer by using 6 strokes . The homogenates were transferred to 1 . 5 mL Eppendorf tube and centrifuged at 4000x g for 5 min . The media was removed and concentrated using the Vivaspin concentrator ( see above ) . The cells-pellet was re-suspended in 40 μL of the Lysis buffer ( see above under Co-Immunoprecipitation experiment ) . Collagenase ( 4 μL of 10 mg/mL solution ) was added to these cells and incubated at room temperature for 20 min . The cells were then centrifuged at 4000x g for 7 min , and the supernatant was collected and 30 μL of this supernatant was subjected to immunoprecipitation using anti-Robo1 antibody and the IP was processed as described above ( under co-immunoprecipitation experiment ) . Western analysis was done using anti-Slit-C antibody .
|
Early during embryogenesis , the Slit ligand is present only at the midline of the nerve cord . It binds its receptor Robo on growth cones of axons and guides axons such that they form a series of axon tracts called longitudinal tracts on either side of the midline . It has been proposed that Slit regulates axon guidance through a Slit gradient emanating from the midline . Mid-way through embryogenesis , a distinct set of axons begins to cross the midline , forming commissures . Slit from the midline travels along commissural tracts to the longitudinal connectives . The function of Slit in tracts is not known . The Slit protein is processed into a larger N-terminal peptide and a smaller C-terminal peptide . Here , I show that Slit/Slit-N interact with Robo in tracts to maintain the fasciculation and position of axon tracts following axon guidance . In the absence of Slit in mature tracts after their guidance , the tracts de-fasciculate and merge with one another and shift their position towards the midline . This work also shows that Slit and Slit-N are not freely present in the extracellular milieu but associated with the extracellular matrix and both interact with Robo1 . Slit-C is bound tightly in the ECM and does not interact with Robo1 . These results define a role for Slit in tracts for axon maintenance and fasciculation , thus the hardwiring of the CNS itself .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"enzymology",
"neuroscience",
"developmental",
"biology",
"signs",
"and",
"symptoms",
"immunoprecipitation",
"nerve",
"fibers",
"embryos",
"research",
"and",
"analysis",
"methods",
"embryology",
"developmental",
"neuroscience",
"animal",
"cells",
"axons",
"proteins",
"precipitation",
"techniques",
"biochemistry",
"cellular",
"neuroscience",
"hydrolases",
"diagnostic",
"medicine",
"cell",
"biology",
"axon",
"guidance",
"phenotypes",
"fasciculations",
"neurons",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"collagenases"
] |
2017
|
Post-guidance signaling by extracellular matrix-associated Slit/Slit-N maintains fasciculation and position of axon tracts in the nerve cord
|
The ability to tolerate Candida albicans , a human commensal of the gastrointestinal tract and vagina , implicates that host defense mechanisms of resistance and tolerance cooperate to limit fungal burden and inflammation at the different body sites . We evaluated resistance and tolerance to the fungus in experimental and human vulvovaginal candidiasis ( VVC ) as well as in recurrent VVC ( RVVC ) . Resistance and tolerance mechanisms were both activated in murine VVC , involving IL-22 and IL-10-producing regulatory T cells , respectively , with a major contribution by the enzyme indoleamine 2 , 3-dioxygenase 1 ( IDO1 ) . IDO1 was responsible for the production of tolerogenic kynurenines , such that replacement therapy with kynurenines restored immunoprotection to VVC . In humans , two functional genetic variants in IL22 and IDO1 genes were found to be associated with heightened resistance to RVVC , and they correlated with increased local expression of IL-22 , IDO1 and kynurenines . Thus , IL-22 and IDO1 are crucial in balancing resistance with tolerance to Candida , their deficiencies are risk factors for RVVC , and targeting tolerance via therapeutic kynurenines may benefit patients with RVVC .
Candida species are the causative agents of vulvovaginal candidiasis ( VVC ) and recurrent VVC ( RVVC ) , two forms of disease that affect a large number of otherwise healthy women [1] , [2] . Uncomplicated VVC is associated with several predisposing factors , including antibiotic and oral contraceptive usage , hormone replacement therapy , pregnancy and uncontrolled diabetes mellitus , and it usually responds to treatment . In contrast , RVVC , marked by idiopathic recurrent episodes , may be virtually untreatable . Despite a growing list of recognized risk factors , further understanding of anti-Candida host defense mechanisms in the vagina is needed to optimize vaccine development [3] , [4] and immune interventions to integrate with , or even replace , antifungal therapy . Colonization of the vaginal mucosa by the fungus induces both humoral and Th immunity [5]–[7] , with the contribution of epithelial [8] and dendritic cells [6] . Acquired Th1 [9]–[11] and Th17 [12] immunity have been described in murine and human VVC . Thus , multiple effector mechanisms of resistance to the fungus are apparently present in VVC . As IL-22 is known to contribute to antifungal resistance at mucosal surfaces by assuring epithelial integrity [13]–[15] , and low levels of IL-22 are associated with chronic and recurrent mucosal candidiasis [16]–[19] , a role for this cytokine in vaginal immune resistance , beyond the polymorphonuclear neutrophil's ( PMNs ) response and alarmins production [20] , is likely . In addition to resistance mechanisms that reduce pathogen burden during infection , tolerance mechanisms that protect the host from immune- or pathogen-induced damage have recently emerged in the area of animal immunity [21] , [22] . It has been argued that a high rate of infection , but low virulence , should select for host tolerance , whereas the opposite condition should favor resistance [23] . Therefore , it is not surprising that tolerance is a complementary host defense trait that increases fitness in response to low-virulence C . albicans in the host-Candida symbiosis [24] . Considerable evidence for an association of recurrent episodes of symptomatic infection with immune hyper-reactivity to the fungus [25]–[27] point to the contribution of a deregulated immune reactivity to the pathogenesis of VVC and support a role for immunoregulation in this disease . As a matter of fact , protection from VVC is associated with limited or absent inflammatory responses that will not necessarily cause the elimination of the fungus , whereas symptomatic infection is associated with a heavy vaginal cellular infiltrate of PMNs and a variable degree of fungal presence [8] , [28] . A plethora of tolerance mechanisms , despite less clarified than resistance mechanisms , have been described [29] , [30] . In murine VVC , CD4+ CD25+ regulatory T ( Treg ) cells [31] , γ/δ T cells [32] and immunoregulatory cytokines , such as IL-10 and transforming growth factor β , have all been demonstrated . Thus , resistance and tolerance are complementary host antifungal defense mechanisms that likely operate in the vaginal mucosa , where the ability to tolerate the fungus implies immune strategies that favor the induction of non-sterilizing protective immunity in an environment permissive for fungal persistence . Indoleamine 2 , 3-dioxygenase 1 ( IDO1 ) , the rate-limiting enzyme in tryptophan degradation along the kynurenine pathway [33] , is a master regulator of antifungal tolerance at mucosal surfaces [34] . Pathogenic inflammatory responses due to IDO1 deficiency account for the inherent susceptibility of mice to aspergillosis [35] and mucosal candidiasis [36] , owing to the unopposed inflammatory responses that compromise the host's ability to efficiently oppose fungal infectivity . By regulating the balance between Th17 and Treg cells , IDO1 may not only contribute to local immune homeostasis but also limit the pro-survival and virulence-promoting activity of IL-17A on fungal cells [37] . A role for IDO1 in the genitourinary system seems likely , because of the intense IDO1-specific staining in a number of tissues from the genitourinary system [38] . There is also evidence for IDO1 involvement in persistent genitourinary Chlamydia trachomatis infection [39] . However , whether IDO1 contributes to protective tolerance to C . albicans in the vagina is presently unknown . In the current study , we evaluated the role of IL-22 and IDO1 in murine and human VVC . We used mice with selective deficiency of IL-22 or IDO1 to explore innate and acquired Th/Treg mechanisms of antifungal protection and patients with VVC and RVVC in which common genetic variants in the IL22 and IDO1 genes were analyzed and correlated with local cytokine production . We found that genetic deficiencies of IL-22 or IDO1 were associated with VVC in mice , due to impaired resistance and tolerance mechanisms to the fungus . Two functional genetic variants in human IL22 and IDO1 were associated with a decreased risk for RVVC and correlated with increased local expression of IL-22 , IDO1 and kynurenines . This study demonstrates that IL-22 and IDO1 mediate antifungal resistance and tolerance to C . albicans in the vagina and that their deficiencies are risk factors for RVVC .
IL-22 and IDO1 are key mediators of resistance and tolerance to Candida [13] , [36] and other fungi [14] , [35] at mucosal surfaces . To assess the role of IL-22 and IDO1 in murine VVC , we intravaginally infected C57BL/6 , IL-22- or IDO1-deficient mice with Candida blastospores and evaluated resistance to infection in terms of vaginal histopathology , PMN recruitment in vaginal lavages , expression of chemotactic S100A8 and S100A9 proteins , known to mediate PMN migration in murine VVC [8] and local fungal growth . In C57BL/6 mice , histological analysis revealed the presence of fungal and inflammatory cells infiltrating the vaginal parenchyma with signs of epithelial damage at the early stages ( Figure 1A ) . Robust PMN recruitment ( Figure 1B and insets in Figure 1A , dpi 3 ) , significant S100a8 and S100a9 gene expression ( Figure 1C ) and calprotectin production ( Figure 1D ) were also observed . Mice eventually controlled the infection , as indicated by a reduction in fungal burden ( Figure 1E ) , PMN number , S100a8 and S100a9 expression , calprotectin level and amelioration of inflammatory pathology at 21 and 42 dpi ( Figure 1A–D ) . The course of the infection was different in IL-22- vs . IDO1-deficient mice and in those mice vs . their respective wild-type counterparts . IL-22-deficient mice were susceptible to C . albicans in the early but not late stages of infection , as indicated by signs of vaginal epithelial damage and inflammation , robust PMN recruitment , high S100a8 and S100a9 gene expression , calprotectin levels and fungal growth at 3 dpi ( Figure 1A–E ) . An opposite pattern of resistance was observed in IDO1-deficient mice , in which resistance to infection was increased early but not late in infection . Indeed , at days 21 and 42 after the infection , mice were unable to restrict fungal growth and inflammation ( Figure 1A–E ) . Wild-type and IL-22-deficient , but not IDO1-deficient mice , showed resistance to re-infection ( Figure S1A in Text S1 ) , a memory response requiring an intact T cell compartment ( Figure S1B–E in Text S1 ) . These findings suggest that IL-22 mediates early antifungal resistance , whereas IDO1 is required to restrain inflammation during ongoing infection and to provide antifungal memory . To directly prove the role of IL-22 , we evaluated levels of IL-22 , as well as of companion cytokines , such as IL-17A and IL-17F , in infection and the consequences of IL-22 inhibition or administration . We found elevated levels of IL-22 through the infection in IDO1-deficient mice in which high levels of IL-17F , but not IL-17A , were also higher as compared to wild-type mice ( Figure 1F ) . Blocking IL-22 ( Figure 1G ) greatly increased fungal growth in IDO1-deficient mice ( Figure 1H ) , whereas exogenous IL-22 decreased fungal growth in wild- type mice ( Figure 1I ) , a finding confirming that IL-22 mediates antifungal resistance in VVC particularly under conditions of IDO1 deficiency . Experiments in IL-17A- or IL-17F-deficient mice confirmed the superior activity of IL-22 vs . IL-17A in early anticandidal resistance in VVC . Despite both types of mice show a higher fungal burden early in infection , as compared to wild-type mice ( Figure 2A ) , IL-17F-deficient , more than IL-17A-deficient , mice showed signs of inflammation ( Figure 2B ) and PMN recruitment ( Figure 2C and inset in Figure 2B ) associated with high levels of IL-17A ( Figure 2D ) . Thus , confirming recent findings [20] , the neutrophil response in vaginal candidiasis occurs independently of IL-22 and IL-17F . Interestingly , the fact that IL-17A binds fungal cells in the vagina and affects fungal growth and morphology [37] , may account for the numerous hyphae observed in vaginal fluids from IL-17F-deficient mice ( inset in Figure 2B ) . In gastrointestinal candidiasis , IL-22 is mainly produced by NKp46+ NK1 . 1low cells , an innate lymphoid cell ( ILC ) subset expressing the aryl hydrocarbon receptor ( AhR ) [13] . We searched for IL-22-producing NKp46+ cells in the vagina of C56BL/6 mice along with γδ T cells , also known to produce IL-22 at mucosal surfaces [40] . Both types of cells , with the predominance of γδ T cells , were present in the vagina , as revealed by FACS analysis ( Figure 3A ) . However , NKp46+cells mainly produced IL-22 , as shown by intracellular cytokine staining in vitro ( Figure 3B ) and vaginal immunostaining in vivo ( Figure 3C ) . Indeed , IL-22-producing NKp46+ cells expanded in infection in wild-type but not IL-22-deficient mice ( Figure 3C ) . In contrast , γδ T cells produced IL-17A more than IL-22 ( about 15% IL-17A+ vs . 5% IL-22+ cells ) ( Figure 3B ) . IL-22 , but not IL-17A , production was significantly decreased in AhR-deficient mice ( Figure 3D ) , a finding suggesting that IL-22 is produced by vaginal NKp46+ via AhR . As a matter of fact , not only was AhR expression increased during VVC , particularly in IDO1-deficient mice ( Figure 3E ) but AhR-deficient mice were also more susceptible to VVC ( Figure 3F and G ) . Thus , much like in the gastrointestinal tract [41] , IL-22 produced via AhR may serve as a first-line resistance mechanism against yeast infection at the vaginal level . IDO1 is known to contribute to Treg cell function in mucosal candidiasis [13] and Treg cells are essential components of immune memory to the fungus [42] . We looked , therefore , for IDO1 protein and gene expressions , kynurenine-to-tryptophan ratios , a valid indicator of IDO1 activity [43] , and IL-10-producing T cells in mice with VVC . We found that IDO1 was promptly induced in infection at the protein ( Figure 4A ) and gene ( Figure 4B ) expression levels , maintained elevated thereafter and was associated with increased levels of kynurenines , downstream products of IDO1 with immunoregulatory functions [44] , [45] ( Figure 4C ) and increased kynurenine to tryptophan ratio [43] ( Figure 4D ) . Both the kynurenine levels and the kynurenine-to-tryptophan ratio were lower in IDO1-deficient mice ( Figure 4C and D ) . The kynurenines were functionally active as replacement therapy with a mixture of l-kynurenine , 3-hydroxykynurenine and 3-hydroxyanthranilic acid , all molecules downstream of the IDO1 pathway [35] , restored immunoprotection to VVC , as indicated by restriction of fungal growth ( Figure 4E ) , amelioration of tissue inflammation ( Figure 4F ) , decreased IL-17A and increased IL-10 production at 21 dpi ( Figure 4G ) . These data suggested that IDO1 mediates the production of tolerogenic kynurenines in VVC . To define the effector mechanism of tolerance in VVC , we evaluated IL-10 and FoxP3 expression in the vaginal parenchyma of re-infected mice . Immunostaining revealed the presence of cells expressing both IL-10 and FoxP3 in wild-type but not IDO1-deficient ( Figure 4H ) , a finding consistent with the levels of IL-10 production ( Figure 4G ) . Interestingly , the kynurenine levels were also lower in IDO1-deficient than wild-type mice after re-challenge ( 2 . 2±0 . 7 vs . 0 . 5±0 . 3 , wild-type vs . IDO1-deficient mice , 3 days after re-challenge ) . Thus , IDO1 , promptly induced in infection , is apparently required for local immunoregulation via IL-10+ Treg cells . The increased resistance seen later in infection in IL-22-deficient but not IDO1-deficient mice , prompted us to define mechanisms of resistance that are independent from IL-22 but dependent on IDO1 . We evaluated the production and expression of IFN-γ and IL-17A in the vagina and the expression of Th-specific transcription factor genes in purified CD4+ T cells from the draining lymph nodes of re-infected mice . We found that resistance to re-challenge in C57BL/6 or IL-22-deficient mice correlated with high-level production ( Figure 5A ) and expression ( Figure 5B ) of IFN-γ and IL-17A . IFN-γ but not IL-17A , production ( Figure 5A ) and expression ( Figure 5B ) were instead reduced in IDO1-deficient mice in which Tbet expression was also lower and Rorc expression higher as compared to wild-type mice ( Figure 5C ) . Importantly , the production of IFN-γ and IL-17A by CD4+ T cells isolated from mice during the primary infection ( 3 dpi ) was significantly lower than that observed upon re-challenge ( Figure 5A ) . Therefore , these data indicated that an appropriate Th1/Th17 cell balance is required for the expression of acquired antifungal immunity . Studies in IL-10-deficient mice confirmed that IL-10+ Treg cells essentially control this balance . IL-10-deficient mice , although capable of restraining the fungal growth during the primary infection ( Figure 5D ) and after re-challenge ( Figure S1F in Text S1 ) , were unable to control tissue inflammation ( Figure S1G in Text S1 ) and PMN recruitment ( Figure 5E ) during the infection , and this was associated with high-level production ( Figure 5A ) and expression ( Figure 5B ) of IFN-γ and IL-17A , and with high Tbet and Rorc expressions in CD4+ T cells ( Figure 5C ) . Thus IL-10 is required to restrain immunopathology , to which Th17 , more than Th1 , cells contribute as revealed by subsequent studies in IFN-γ- or IL-17RA-deficient mice . While more resistant to the infection in the early stage– likely due to the high levels of IL-22 ( Figure S2A in Text S1 ) –both IFN-γ- and IL-17RA-deficient mice progressively become susceptible to infection , as indicated by the failure to restrain fungal growth during the primary infection ( Figure S2B in Text S1 ) or after re-challenge ( Figure S2C in Text S1 ) and to limit inflammation ( Figure S2D-F in Text S1 ) . In contrast to IL-17A , the levels of IFN-γ were either absent or greatly reduced ( Figure S2G in Text S1 ) , a finding indicating that IFN-γ is a key mediator of acquired resistance to the fungus , to which IL-17RA signaling contributes , as already suggested [13] . Studies in IL-12p40-deficient mice confirmed the combined requirement of Th1 and Th17 cells for optimal antifungal memory resistance , as compared to IL-12p25-deficient or IL-23p19-deficient mice ( data not shown ) . We know that functional yet balanced reactivity to C . albicans at mucosal surfaces requires both the myeloid differentiation primary response gene 88 ( MyD88 ) and the TIR-domain-containing adapter-inducing interferon-β ( TRIF ) as well as different upstream innate receptors [13] . We evaluated resistance to the primary infection and re-challenge in MyD88-or TRIF-deficient mice intravaginally infected with the fungus . Early in infection , fungal burden was higher in MyD88-deficient mice and declined thereafter ( Figure 6A ) . The inflammatory response with PMN recruitment observed early in infection resolved later in infection ( Figure 6B ) , at a time when those mice developed resistance to re-infection ( Figure 6C ) . Growth was instead restrained in TRIF-deficient mice early in infection , but fungal growth was observed in the vagina at a later stage ( Figure 6A ) , when mice effectively controlled fungal growth if re-challenged ( Figure 6C ) , but not the associated inflammatory response ( Figure 6B ) . Consistent with these findings , IL-22 and IL-17A were particularly lower in MyD88-deficient mice as compared to TRIF-deficient mice ( Figure 6D ) , whereas IL-17A more than IFN-γ/IL-10 were present in TRIF-deficient mice ( Figure 6E ) . To identify the upstream innate receptors involved in the response , we infected mice deficient in receptors known to recruit the MyD88 ( TLR2/TLR6/TLR9 ) , TRIF ( TLR3 ) , or both ( TLR4 ) pathways and whose expression was observed in the vagina ( Figure S3 in Text S1 ) . Compared to wild-type mice , resistance to primary infection ( Figure 6F ) and to re-challenge ( Figure 6G ) was similar in TLR9-deficient or TLR6-deficient mice; it was greatly increased in TLR2-deficient mice but impaired in TLR3- or TLR4-deficient mice . Thus , the MyD88 pathway mainly contributes to antifungal mucosal resistance , through the involvement of TLR4 , TLR3 ( this study ) and the beta-glucan receptor Dectin-1 [46] , while the TRIF pathway contributes to tolerance via TLR3 and TLR4 . Because of the above results , we investigated whether genetic variants possibly affecting the functions of IL-22 and IDO1 might influence susceptibility to human VVC and RVVC ( Table S1 in Text S1 ) . Within the set of single nucleotide polymorphisms ( SNPs ) tested ( Table S2 in Text S1 ) , we found that the genotype frequencies for rs2227485 in IL22 and rs3808606 in IDO1 were significantly different between controls and RVVC , but not VVC , patients ( Figure 7A and Table S3 in Text S1 ) . Specifically , the TT genotype at rs2227485 in IL22 was significantly associated with a decreased risk for RVVC [12 . 4% in RVVC vs . 22 . 8% in controls; odds ratio ( OR ) , 0 . 48; 95% confidence interval ( CI ) , 0 . 27–0 . 85; P = 0 . 01] . Likewise , the TT genotype at rs3808606 in IDO1 also correlated with a minor susceptibility to RVVC ( 13 . 8% in RVVC vs . 24 . 0% of controls; OR , 0 . 51; 95% CI , 0 . 29–0 . 88; P = 0 . 02 ) . As dectin-1 mediates IL-22 production in mucosal candidiasis [46] , and the early stop-codon mutation Y238X in DECTIN1 ( rs16910526 ) has been shown to predispose to familial RVVC in homozygous individuals [47] , we also assessed the distribution of this substitution within our study subject group . As expected , we found that the TG genotype conferred an increased risk for RVVC ( 24 . 1% in RVVC vs . 11 . 8% in controls; OR , 2 . 38; 95% CI , 1 . 40–4 . 06 , P = 0 . 001 ) ( Figure 7A and Table S3 in Text S1 ) . Functional analyses confirmed that protection afforded by carriage of the TT genotype at rs2227485 in IL22 correlated with high levels of vaginal IL-22 and decreased levels of pro-inflammatory IL-17A , TNF-α and calprotectin ( Figure 7B and C ) . Likewise , high levels of IL-22 and decreased levels of IL-17A and TNF-α were observed in women carrying the protective TT genotype at rs3808606 in IDO1 ( Figure 7D ) , and they were associated with enhanced IDO1 expression in vaginal cells ( Figure 7E ) and increased kynurenine-to-tryptophan ratio ( Figure 7F ) . Altogether , these data provide evidence that common genetic variants leading to enhanced expression phenotypes of IL22 and IDO1 may contribute to protection in RVVC .
This study disentangles resistance and tolerance components of murine and human C . albicans vaginal infection and introduces the challenging notion of a disease due to a defective tolerance mechanism . While some degree of inflammation is required for protection , particularly at mucosal tissues during the transitional response occurring between the rapid innate and slower adaptive responses , progressive inflammation worsens disease and ultimately prevents pathogen eradication [24] , [35] . Much like in gastrointestinal candidiasis [13] , resistance and tolerance mechanisms in VVC are activated through the contribution of innate and adaptive immune responses , involving distinct modules of immunity , i . e . , IL-22 and Th1/Th17 cells for resistance and IL-10+ Treg cells for tolerance . It has already been shown that IL-22 provides antifungal resistance through disparate mechanisms , including i ) growth control of the early infecting morphotype , the yeasts , via the induction of antimicrobial peptides with anticandidal activity and ii ) epithelial integrity control , via phosphorylation of STAT3 [13] , known to be required for limiting damage and inflammation in mucosal candidiasis [48] . Thus IL-22 variations may explain why some patients are at high risk of vaginal yeast infection . Although many different cell types produce IL-22 [41] , intestinal ILCs expressing AhR , now termed ILC3 [49] , are known to produce , in addition to other cytokines , IL-22 [50] . ILCs reflect many functions of CD4+ T helper cells but expand and act shortly after stimulation . They play fundamental roles early in response to infection and injury , in the maintenance of homeostasis , and possibly in the regulation of adaptive immunity [51] . We found here that NKp46+ cells expanded in the vagina in infection and produced IL-22 , more than IL-17A , likely via AhR . As expression of cytokines by ILCs is regulated by signals provided by epithelial cells in response to microbiota , our finding may provide a mechanistic explanation for the link between microbial dysbiosis and vaginal candidiasis . Indeed , the fact that IL-22 production is driven by commensals may explain how antibiotic therapy and iatrogenic immunosuppression are major predisposing factors in candidiasis and , more generally , how the bacterial-fungal population dynamics impact on vaginal homeostasis and inflammatory diseases . Mutations in IL17F , IL17R and DECTIN1 in patients with chronic mucocutaneous candidiasis , as well as neutralizing autoantibodies against IL-17 and IL-22 in patients with autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy , directly impair IL-17 and IL-22 immunity [47] , [52]–[54] . We found here that IL-22 , more than IL-17A , contributes to resistance to VVC . In mice , resistance was abrogated under conditions of IL-22 or IL-17F , more than IL-17A , deficiency , whereas a common genetic variant in human IL22 leading to enhanced production of IL-22 , but not IL-17A and TNF-α , conferred protection against RVVC . Thus , in addition to functional polymorphisms in genes coding for mannose-binding lectin [55]–[57] , IL-4 [58] , and the inflammasome component NALP3 [59] in predisposing to RVVC , our study identified an IL22 variant that is associated with a decreased risk for RVVC and is consistent with the findings obtained relative to DECTIN1 deficiency [47] . To our knowledge , this is the first genetic variant in IL22 found to be associated with a human disease [41] , and our finding confirms the important functions of the IL-22–IL-22R pathway in regulating immunity , inflammation and tissue homeostasis , and the therapeutic potential of targeting this pathway in human disease [41] . Considered a master regulator of antifungal resistance and tolerance at mucosal surfaces [34] , IDO1 has gained reputation in the field of immune mycology owing to its ability to generate immunomodulatory kynurenines that induce Treg cells suppressing local antifungal T-cell responses , thus favoring fungal persistence [34] . It is known that IDO1 activity exhibits relatively large interindividual variability , in particular under pathological conditions [60] , [61] and that there are naturally occurring polymorphisms that transcriptionally regulate the human IDO1 gene [62] , [63] . Thus , genetic factors are involved in interindividual variability of IDO expression and/or activity . We found that an IDO1 variant leading to enhanced IDO1 expression and concomitant production of kynurenines was associated with decreased risk for RVVC . IDO1 regulates tolerance to the fungus at the vaginal level , as it ameliorates immunopathology that positively correlated with the magnitude of the immune response . It was indeed required for the generation of IL-10+ Treg cells negatively affecting Th1/Th17 cells . Moreover , IDO1 also favored the induction of optimal antifungal memory resistance , an activity likely due to its ability to limit tissue damage , thus allowing for a higher magnitude and duration of the immune response than would have been otherwise possible . Finally , the finding that IL-22 was up-regulated in condition of IDO1 deficiency indicates that the resistance and tolerance mechanisms are to some extent reciprocally regulated . Understanding the mechanisms that are critical for host survival is important for the choice of therapeutic approaches . Antifungal therapy is highly effective for individual symptomatic attacks but does not prevent recurrences . In addition to the associated costs that are very high [64] , there is concern that repeated treatments might induce drug resistance . Thus medical treatments that increase host resistance , such as antibiotics , place selective pressures on pathogens . As tolerance mechanisms are not expected to have the same selective pressure on pathogens , new drugs that target tolerance will provide therapies to which low-virulence fungi , such as C . albicans , will not develop resistance . Kynurenines appear to fulfill this requirement .
Murine experiments were performed according to the Italian Approved Animal Welfare Assurance A-3143-01 and Legislative decree 245/2011-B regarding the animal license obtained by the Italian Ministry of Health lasting for three years ( 2011–2014 ) . Infections were performed under avertin anesthesia and all efforts were made to minimize suffering . The experimental protocol was designed in conformity with the recommendations of the European Economic Community ( 86/609/CEE ) for the care and the use of laboratory animals , was in agreement with the Good Laboratory Practices and was approved by the animal care Committee of the University of Perugia ( Perugia , Italy ) . All patients and control subjects were observed at the Microbiology Department , S . Maria della Misericordia Medical Center ( Perugia , Italy ) and answered a detailed questionnaire reporting social and demographic information , medical history , subjective symptoms for gynecological infections and sexual behavior . The study approval was provided by the University ethics committee ( Prot . 2012-028 ) and informed written consent was obtained from all participants . Enrollment took place between January 2009 and June 2012 . Female C57BL/6 and NOD . SCID ( NOD . CB17-Prkdcscid/NCrCrl ) mice , 8–10 wk old , were purchased from Charles River ( Calco , Italy ) . Homozygous Il22−/− , Ido1−/− , Ahr−/− , Ifng−/− , Il17ra−/− , Il17a−/− , Il17f−/− , Il10−/− , MyD88−/− , Trif−/− , Tlr2−/− , Tlr3−/− , Tlr4−/ , Tlr6−/− and Tlr9−/− mice on the C57BL/6 background were bred under specific pathogen-free conditions at the Animal Facility of the University of Perugia , Perugia , Italy . Mice were treated subcutaneously with 0 . 1 mg of β-estradiol 17–valerate ( Sigma Chemical Co . ) dissolved in 100 µl of sesame oil ( Sigma ) 48 h before vaginal infection . Estrogen administration continued weekly until completion of the study to maintain pseudoestrus . The estrogen-treated mice were inoculated intravaginally with 20 µl of phosphate-buffered saline ( PBS ) suspension of 5×106 viable C . albicans 3153A blastospores from early–stationary-phase cultures ( i . e . , 18 h of culture at 36°C in Sabouraud-dextrose agar with chloramphenicol plates , BD Diagnostics ) . Re-challenge was performed by intravaginal inoculation of 5×106 blastospores , 3 or 5 weeks after the primary infection . In the vaccine-induced resistance experiments , 5×106 heat-killed C . albicans ( HCA ) , obtained by exposing 1×108 cells/ml at 56°C for 30 min , or live low-virulence mutant cells obtained by mutagenesis [65] were intravaginally injected into estradiol-treated mice , 3 weeks before re-challenge . Control mice received PBS . The time course of infection was monitored in individual mice by culturing 100 µl of serially diluted ( 1∶10 ) vaginal lavages on Sabouraud-dextrose agar with chloramphenicol plates . Vaginal lavages were conducted using 100 µl of sterile PBS with repeated aspiration and agitation . CFUs were enumerated after incubation at 36°C for 24 h and expressed as log10 CFU/100 µl of lavage fluid . Quantitative counts of CFU in lavage fluids were evaluated successively in mice anesthetized for each lavage . Cytospin preparations of the lavage fluids were stained with May-Grünwald-Giemsa and observed with a BX51 microscope equipped with a high-resolution DP71 camera ( Olympus ) . IL-22 blockade was achieved as described [13] by intraperitoneally injecting mice with a total of 300 µg of mAb neutralizing IL-22 ( clone AM22 . 3 , mouse IgG2a ) or isotype control mAb ( Sigma-Aldrich ) the day of and 1 day after the infection . Recombinant murine IL-22 ( BioLegend ) was given intravaginally ( 1 µg/mouse ) daily , the day of the infection and 1 and 2 days after . Control mice received PBS . Mice were treated intraperitoneally daily , starting 3 days before and up to 3 days after the infection with PBS or 20 µg/kg of a mixture of l-kynurenine , 3-hydorxykynurenine , and 3-hydroxyanthranilic acid ( Sigma-Aldrich ) . For isolation of CD4+ cells , iliac and inguinal lymph nodes were aseptically removed and cut into small pieces in cold medium . The dissected tissue was then incubated in medium containing collagenase XI ( 0 . 7 mg/ml; Sigma-Aldrich ) and type IV bovine pancreatic DNase ( 30 mg/ml; Sigma-Aldrich ) for 30–45 min at 37°C . A single cell suspension was obtained and incubated with CD4 MicroBeads ( Miltenyi Biotech ) before magnetic cell sorting . Vaginal tissues were chopped into fragments and incubated in 1 . 3 mM EDTA for 30 min at 37°C , as described [66] , followed by digestion for 90 min in collagenase type XI ( 1 mg/ml ) . These digested pieces were minced and filtered through a nylon mesh , and the resulting cells were filtered through a 70-µm filter . NKp46+ and γδ T cells were purified as per the manufacturer's instruction , with the mouse anti-NKp46 MicroBead Kit and the TCR γδ T cell isolation kit ( Miltenyi Biotec ) . Antibodies were as follows: anti-CD3ε ( 145-2C11 ) , anti-γδ ( GL3 ) ( BD PharMingen ) and anti-NKp46 ( CD335 ) ( eBioscience ) . All staining reactions were performed at 4°C on cells first incubated with an Fc receptor mAb ( 2 . 4G2 ) to reduce non-specific binding . For intracellular staining , cells were stimulated with PMA/ionomycin , added of brefeldin and then permeabilized with the CytoFix/CytoPerm kit ( BD Biosciences ) for intra-cytoplasmic detection of IL-17A ( clone: eBio17B7 , eBioscience ) , and IL-22 ( clone 14 . 03 . 01 , R&D System ) . Cells were analyzed with a FACScan flow cytofluorometer ( Becton Dickinson ) equipped with CELLQuest software . For histology , the vaginas were removed and immediately fixed in 10% neutral buffered formalin ( Bio-optica , Milan ) for 24 h . The vaginas were dehydrated , embedded in paraffin , sectioned into 3–4 µm and stained with periodic acid-Schiff reagent . Histology sections were observed using a BX51 microscope equipped with a high-resolution DP71 camera ( Olympus ) . The level of murine and human cytokines in the lavage fluids were determined by Kit ELISA ( R&D Systems ) . The detection limits of the assays were <10 pg/ml for IL-22 , <3 pg/ml for IL-10 , <30 pg/ml for IL-17F , <10 pg/ml for IL-17A and <1 . 6 pg/ml for IFN-γ for murine cytokines and <2 . 7 pg/ml for IL-22 , <15 pg/ml for IL-17A , <15 pg/ml for TNF-α and <4 pg/ml for IL-10 for human cytokines . Data were normalized to total protein levels for each sample as determined using the Bio-Rad Protein assay ( Life Science , Bio-Rad Laboratories S . r . l . Milan , Italy ) and expressed as pg cytokine/mg total protein . Results represent mean cytokine levels ( ± s . e . m . ) from samples pooled from two similar experiments ( n = 3–4 total samples per group ) . Human and murine calprotectin were determined by ELISA ( Hycult biotech , Uden , The Netherlands and Immundiagnostik AG , Bensheim , Germany , for human and murine detection , respectively ) . Real-time PCR was performed using the iCycler iQ detection system ( Bio-Rad ) and SYBR Green chemistry ( Finnzymes Oy ) . Vaginas and purified cells from lymph nodes were lysed and total RNA was extracted using RNeasy Mini Kit ( QIAGEN , Milan , Italy ) and was reverse transcribed with Sensiscript Reverse Transcriptase ( QIAGEN ) according to the manufacturer's directions . The PCR primers were as described [13] . Amplification efficiencies were validated and normalized against Gapdh . The thermal profile for SYBR Green real-time PCR was at 95°C for 3 min , followed by 40 cycles of denaturation for 30 s at 95°C and an annealing/extension step of 30 sec at 60°C . Each data point was examined for integrity by analysis of the amplification plot . The mRNA-normalized data were expressed as relative mRNA in knockout vs . wild-type mice and infected vs . naïve mice . TLR expression was evaluated in vaginal lysates using TLR specific primers and conditions as described [67] . The normalized CT value for the target amplification ( ΔCT , Tlr ) was determined by subtracting the average Gapdh CT value from the average Tlr CT value . The vagina was removed and fixed in 10% phosphate-buffered formalin , embedded in paraffin and sectioned at 5 mm . Sections were then rehydrated and after antigen retrieval in Citrate Buffer ( 10 mM , pH 6 ) , sections were blocked with 5% BSA in PBS and stained with PE anti-IL-17A , -IFN-γ ( XMG1 . 2 ) , -IL-22 , FITC-anti-NKp46 ( eBioscience ) followed by anti-rabbit TRITC ( Sigma ) . Double staining with FITC anti-IL-10 ( JES5-16E3 ) and rabbit polyclonal to FOXP3 ( abcam ) was followed by anti-rabbit TRITC . All mAbs were incubated overnight at 4°C . Images were acquired using a fluorescence microscope ( BX51 Olympus ) with a 20× objective and the analySIS image processing software ( Olympus ) . 4′-6-Diamino-2-phenylindole ( DAPI , Molecular Probes ) was used to counterstain tissues and to detect nuclei . Cells from murine vaginas or from human vaginal washes were lysed in 2× Laemmli buffer ( Sigma-Aldrich ) and the lysates were separated in 10% Tris/glycine SDS gel and transferred to a nitrocellulose membrane . Blots of cell lysates were incubated with mouse anti-human IDO1 antibody clone 10 . 1 ( Millipore Billerica ) or rabbit polyclonal anti-murine IDO1 antibodies [61] . Normalization was performed by probing the membrane with mouse-anti-β-tubulin antibody ( Sigma-Aldrich ) . Images were acquired with LiteAblotPlus chemiluminescence substrate ( Euroclone S . p . A . ) using ChemiDoc XRS and Imaging system ( Bio-Rad Laboratories ) and quantification was obtained by densitometry image analysis using Image Lab 3 . 1 . 1 software ( Bio-Rad ) . The kynurenine to tryptophan ratio was calculated by relating concentrations of kynurenine and tryptophan determined by HPLC with the internal calibrator 3-nitro-L-tyrosine , as described [43] . Chromatography was performed on reversed-phase cartridges LiChroCART RP18 columns , tryptophan was monitored via its fluorescence at 285 nm excitation and 365 nm emission wavelengths , kynurenine was measured by its UV absorption at 360 nm wavelength . The kynurenine to tryptophan ratio ( Kyn/Trp ) was calculated and expressed as µmol kynurenine/mmol tryptophan . The study population included Caucasian women who had ≥4 ( n = 145 ) or 1–3 ( n = 293 ) culture-verified symptomatic episodes of a vulvovaginal Candida infection during a 12-month period ( Table S1 in Text S1 ) . Control subjects consisted of 263 age-matched healthy Caucasian women with no gynecologic complaints , no history of vaginal Candida infection , and who were currently culture-negative for vaginal pathogens . Exclusion criteria were pregnancy , diabetes mellitus , endocrine or immune deficiency disorders , use of immunosuppressive medications , antibiotics or high estrogen content contraceptives , chemotherapy or prior hysterectomy . Cervicovaginal samples were obtained from all participants by instilling 3 ml of sterile saline into the posterior vagina , mixing the saline with secretions and withdrawing the solution with a syringe . All vaginal washes were centrifuged at 12 , 000× g for 10 min to separate the mucus from the PBS wash solution shortly after collection and pellet fractions and immediately frozen at −20°C . Genomic DNA was isolated from the pellet fraction using the QIAamp DNA Mini kit ( Qiagen ) . SNPs were selected either from the literature or based on their ability to tag surrounding variants in the HapMap-CEU population of the International HapMap project , NCBI build B36 assembly HapMap phase III ( http://www . hapmap . org ) . The Haploview 4 . 2 software was used to select haplotype-based tagging SNPs by assessing linkage disequilibrium blocks from the genes of interest with a pairwise correlation coefficient r2 of at least 0 . 80 and a minor allele frequency of ≥5% in the HapMap-CEU population . SNPs evaluated are indicated in Table S2 in Text S1 . Genotyping was performed as previously described [68] , [69] . Primer sequences are available upon request . Each genotyping set comprised randomly selected replicates of sequenced samples and negative controls . Concordant genotyping was obtained for >99% assays . For the functional assays , and unless stated otherwise , measurements were performed in vaginal washes obtained from at least 10 different women without ongoing symptoms and that had negative culture results for each genotype under study , assessed in quadruplicates . Student's T-test or analysis of variance ( ANOVA ) with Bonferroni's adjustment were used to determine statistical significance ( P<0 . 05 ) . The data reported are either from one representative experiment out of three to five independent experiments ( western blotting and RT–PCR ) or pooled from three to five experiments , otherwise . The in vivo groups consisted of 6–8 mice/group . Data were analyzed by GraphPad Prism 4 . 03 program ( GraphPad Software ) . Genotype distributions among controls and VVC and RVVC patients were analyzed by Fisher's exact test and P values less than 0 . 05 were considered as significant . Genotype frequencies were distributed according to the Hardy-Weinberg equilibrium for all SNPs ( P>0 . 05 ) .
|
This study disentangles resistance and tolerance components of murine and human C . albicans vaginal infection and introduces the challenging notion of a disease due to a defective tolerance mechanism . Vulvovaginal candidiasis ( VVC ) and recurrent VVC ( RVVC ) are two forms of disease that affect a large number of otherwise healthy women . Uncomplicated VVC is associated with several predisposing factors , whereas RVVC , marked by idiopathic recurrent episodes , may be virtually untreatable . Despite a growing list of recognized risk factors , further understanding of anti-Candida host defense mechanisms in the vagina is needed to optimize vaccine development and immune interventions to integrate with , or even replace , antifungal therapy . Indeed , medical treatments that increase host resistance , such as antifungals , are highly effective for individual symptomatic attacks but do not prevent recurrences and there is concern that repeated treatments might induce drug resistance . As tolerance mechanisms are not expected to have the same selective pressure on pathogens , new drugs that target tolerance will provide therapies to which low-virulence pathogens , such as C . albicans , will not develop resistance . This study provides a proof-of-concept that targeting tolerance via therapeutic kynurenines may benefit patients with RVVC .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] |
[
"genetics",
"immunology",
"biology",
"microbiology"
] |
2013
|
IL-22 and IDO1 Affect Immunity and Tolerance to Murine and Human Vaginal Candidiasis
|
As with many viruses , rabies virus ( RABV ) infection induces type I interferon ( IFN ) production within the infected host cells . However , RABV has evolved mechanisms by which to inhibit IFN production in order to sustain infection . Here we show that RABV infection of dendritic cells ( DC ) induces potent type I IFN production and DC activation . Although DCs are infected by RABV , the viral replication is highly suppressed in DCs , rendering the infection non-productive . We exploited this finding in bone marrow derived DCs ( BMDC ) in order to differentiate which pattern recognition receptor ( s ) ( PRR ) is responsible for inducing type I IFN following infection with RABV . Our results indicate that BMDC activation and type I IFN production following a RABV infection is independent of TLR signaling . However , IPS-1 is essential for both BMDC activation and IFN production . Interestingly , we see that the BMDC activation is primarily due to signaling through the IFNAR and only marginally induced by the initial infection . To further identify the receptor recognizing RABV infection , we next analyzed BMDC from Mda-5−/− and RIG-I−/− mice . In the absence of either receptor , there is a significant decrease in BMDC activation at 12h post infection . However , only RIG-I−/− cells exhibit a delay in type I IFN production . In order to determine the role that IPS-1 plays in vivo , we infected mice with pathogenic RABV . We see that IPS-1−/− mice are more susceptible to infection than IPS-1+/+ mice and have a significantly increased incident of limb paralysis .
Type I interferon ( IFN ) was first identified as a “factor” that rendered cells resistant to viral infection [1] . It is now known that following viral infection , cells induce type I IFN , which in turn upregulates the expression of numerous antiviral proteins [2] . This class of cytokines is comprised of several genes including multiple IFN-α genes , a single IFN-ß gene , and the less well-defined IFN-ω , -ε , -τ , -δ , and -κ ( for review [3] ) . In addition to having antiviral functions , type I IFNs play a part in activating the adaptive immune response following infection [4] , [5] , [6] . For instance , IFN-α/ß can strengthen the innate immune response by activating antigen presenting cells ( APC ) . Additionally , following maturation in the presence of type I interferon and GM-CSF , monocyte-derived DCs more effectively stimulate an antigen-specific CD8+ T cell response than when differentiated with GM-CSF alone [7] . Viral infection can trigger the type I IFN response via various pattern recognition receptors ( PRR ) , namely Toll-like receptors ( TLR ) and RIG-I-like receptors ( RLR ) . In the case of negative stranded RNA viruses , the members of the TLR family that are generally involved in viral recognition , TLR-3 and TLR-7 , are found on the endosomal membrane . To initiate the signaling cascade , TLR-3 binds double stranded RNA molecules [8] , whereas TLR-7 recognizes immunomodulatory compounds ( ie-imiquimod ) [9] or single-stranded RNA molecules [10] . Although negative stranded RNA viruses do not produce double stranded RNA as part of their normal replication cycle , it is likely that abnormal replication products resulting from errors by viral RNA-dependent RNA polymerases give rise to some level of double stranded RNA in virus-infected cells [11] . TLR-3 and TLR-7 initiate signaling though different adaptor molecules , Trif and MyD88 , respectively; however , the pathways converge on the phosphorylation of IRF-3 . Following phosphorylation , IRF-3 forms protein dimers , which allow for its transport into the nucleus where it can bind to the IFN-ß promoter [12] . Alternatively , RNA viruses can be recognized in the cytoplasm by RLRs , namely RIG-I and Mda-5 [13] . These helicase-like proteins recognize double-stranded RNA and 5′ tri-phosphate groups [14] . In the case of rabies virus ( RABV ) , the negative stranded RNA virus of interest in this study , the leader RNA remains unmodified [15] , [16] and thus provides a potential ligand for these RLRs . Signaling by RIG-I and Mda-5 is mediated through the mitochondria-bound protein IPS-1 , which is also referred to as MAVS , Cardif , or VISA [17] , [18] , [19] , [20] . Similar to what is seen in TLR signaling , RLR signaling culminates with the activation and nuclear translocation of IRF-3 [19] . Rabies virus is a member of the Rhabdoviridae family . RABV has a relatively simple genome , comprised of just 5 proteins: the nucleoprotein , phosphoprotein ( P ) , matrix protein , glycoprotein and the RNA dependent RNA polymerase . Infection with RABV can induce IFN-α/ß production rapidly in vivo . Furthermore , it was seen that a mouse's ability to induce type I IFN , as measured by serum concentrations 4 days post infection , positively correlates to the animal's resistance to RABV [21] . The type I IFN response is also important in driving immunity , as mice injected with anti-mouse IFN-α/ß antibody prior to infection with RABV were more sensitive to the virus than mice injected with a control antibody [22] . However , RABV has the ability to antagonize type I IFN induction [23] . Thus , shortly after infection of fibroblast cells , RABV-P prevents IRF-3 phosphorylation in order to suppress IFN-α/ß production [23] . Although IFN is induced after RABV infection , RABV is able to suppress the IFN response shortly after infection . Therefore , in order to study the receptors responsible for the initial induction of IFN , several groups have used recombinant viruses with lower levels of RABV-P . Using this method , it was determined that IFN-ß promoter activity was seen following recombinant RABV infection of VERO cells transfected with wildtype RIG-I , but not in cells transfected with dominant-negative mutant RIG-I [24] , thus indicating a role for RIG-I in mounting an innate immune response to RABV . Additionally , following infection of human postmitotic neurons with RABV , Prehaud et al . saw an increased production of IFN-ß and TLR-3 mRNAs [25] . Furthermore , the expression of TLR-3 on cerebellar cortex tissues of individuals that had died of rabies , but not on an individual that died of cardiac arrest , verify the viral induced expression of TLR-3 in human brains in vivo [26] . This upregulation of TLR-3 following infection suggests a possible role for TLR-3 signaling in the innate recognition of RABV; however , TLR-3 activation needs to be further studied to conclusively define such a role . Although these results hint at the receptors responsible for interferon expression , there is no evidence that other PRR receptors , such as TLR-7 and Mda-5 , do not also play a role . Furthermore , since the recombinant viruses used in some of these studies exhibit decreased pathogenicity , it is possible that a wildtype virus may act differently following infection . In order to study the IFN-inducing pathways triggered by RABV , we needed to identify a cell type in which RABV-P is unable to antagonize type I IFN signaling . Of note , it has been seen that following infection of dendritic cells ( DCs ) with influenza , another negative stranded RNA virus , the DCs become infected , but this infection is non-productive [27] . Here , we sought to determine whether APCs were productively infected with RABV . Similar to previous reports that human DCs are susceptible to RABV infection [28] , [29] , we saw that mouse DCs became infected; however we also observed that very little viral progeny was released due to limited viral replication . Due to the overall suppression of viral transcription in RABV infected DCs there are presumably low levels of RABV-P that may not be able to inhibit interferon induction . Thus , we decided to utilize infection of DC to study the IFN-inducing capabilities of RABV and found that RLRs are responsible for viral recognition in DCs .
It has been previously shown that RABV-P can inhibit the phophorylation of IRF-3 in fibroblast cells [23] , thus crippling the induction IFN-α/ß . However , RABV is able to infect a variety of cells including neurons [30] and antigen presenting cells ( APC ) [28] , [29] in addition to fibroblasts . Thus , we wanted to determine whether RABV is able to inhibit IFN signaling in other cell types including DCs , which are known to induce the adaptive immune response . In order to check for type I IFN production , we first infected a variety of cell types including fibroblasts ( BSR ) , neuronal cells ( NA ) , macrophages ( Raw264 . 7 ) and DCs ( JAWSII ) with a RABV vaccine strain-based vector , SPBN . Following infection with RABV , cell supernatants were collected and subsequently UV-treated in order to deactivate any infectious virus but retain secreted cellular proteins , such as type I IFN . We then transferred the supernatants to reporter cells , which are sensitive to IFN . Twenty-four hours after supernatant transfer , reporter cells were infected with recombinant vesicular stomatitis virus expressing GFP ( VSV-GFP , [31] ) for 5–8h . VSV replication is highly sensitive to type I IFN [32] , and thus , in the presence of type I IFN , the replication of VSV is suppressed [4] . Following infection with RABV , macrophages as well as DCs , but not fibroblasts or neuronal cells , produce type I IFN that inhibits VSV-GFP replication , as indicated by the lack of GFP expression ( Figure 1A ) . Of note , when BSR , NA , Raw264 . 7 , or JAWSII cells are originally treated with UV-deactivatecd RABV , the supernatants from these cells are unable to block VSV replication ( Figure 1B ) ; therefore , IFN is secreted only after RABV replication . In order to account for the increased amounts of type I IFN produced following RABV infection of macrophages and DCs when compared to the amount produced by fibroblast and neuronal cells , we did a one-step growth curve following infection of the various cell types . Supernatants from infected cells was titered on BSR cells , which are insensitive to type I IFN [4] . We detected that , although all four cell types were infected , only BSR and NA cells produce infectious virus ( Figure 2A ) . There are two possible explanations for the defect in viral production observed here: either a block in viral replication or a defect in viral assembly . In order to compare viral transcription and replication in fibroblast and dendritic cell lines we used quantitative PCR . In fibroblast cells , we saw that the amount of RABV-N messenger RNA ( mRNA ) transcripts increased an average of 1 . 95 logs from 8 hours post infection ( hpi ) to 48 hpi . Similarly , the quantity of RABV-N genomic RNA transcripts ( gRNA ) increased an average of 1 . 2 logs from 8 hpi to 48 hpi . This data indicates that following infection of fibroblast cells both viral transcription ( mRNA ) and replication ( gRNA ) occurs . On the other hand , when looking at the quantity of RABV-N found in dendritic cells following infection we see that there was no increase in the number of mRNA or gRNA viral transcripts when comparing 8hpi to 48 hpi . Thus , it appears that RABV is able to enter APCs , but only limited viral transcription occurs following entry . It is reasonable to assume that decreased levels of transcription might result in low levels of RABV-P . It has been previously shown that recombinant RABV expressing low amounts of RABV-P is unable to inhibit type I IFN induction [23] . Furthermore , we show by Western blotting that cell lysate from RABV infected JAWSII cells contained undetectable levels of RABV P 48hpi ( Figure 2B ) . On the other hand , we were able to detect RABV P in lysates from infected BSR cells as early as 12 hpi after infection . This results support the conclusion that a very low level of RABV P within infected APCs is not able to block the induction of type I IFN and therefore is responsible for the increase in type I IFN production by these cells following infection . In order to better understand the interaction of RABV with host cells following infection , we sought to identify the pathway ( s ) responsible for type I IFN induction in infected cells . Since we detected that DCs make large amounts of IFN following RABV infection we decided to use bone marrow derived DCs ( BMDC ) in our studies . To differentiate BMDCs , we cultured the cells in the presence of 10 ng/ml GM-CSF . After 7 days the majority of cells have matured to DCs as shown by the expression of CD11b+CD11c+ ( Figure 3 ) . In order to identify the PRR that recognizes RABV , we isolated BMDC from mice deficient in various signaling components of PRR pathways . In each experiment , cells were stimulated , and the CD11c+ cell population ( Figure 3 ) was analyzed for production of type I IFN and expression of CD86 , a co-stimulatory molecule that is upregulated on activated DCs . First , we analyzed the role that TLR signaling plays in BMDC activation and type I IFN production following a RABV infection . It has been previously reported that following infection of human postmitotic neurons with RABV , there is an increased production of IFN-ß and TLR-3 mRNAs . In addition , treatment of neurons with poly ( I:C ) , a TLR-3 agonist , generated a similar cytokine profile to that which was seen following RABV infection [25] . Thus , we differentiated BMDCs from TLR-3−/− and congenic wildtype mice and infected the cells with RABV . We then analyzed the infected cells for the presence of CD86 ( Figure 4A ) . As shown in Figure 4C , there is no significant difference in the expression of CD86 on the cell surface of RABV infected BMDCs derived from TLR-3−/− or wildtype mice . As expected , TLR ligands that signal via other TLR receptors , namely TLR-4 ( LPS ) , TLR-9 ( ODN1826 ) , and TLR-7/8 ( R848 ) , equally activate BMDCs derived from wildtype ( wt ) or TLR-3−/− mice . Interestingly , poly ( I:C ) , a known ligand for TLR-3 , was able to activate BMDC isolated from TLR-3−/− mice as well as wt mice . However , it has been previously shown that poly ( I:C ) can also signal through Mda-5 and that Mda-5 is the dominant receptor for mediating type I IFN induction following poly ( I:C ) stimulation in BMDCs [33] , [34] . As the RLR pathway remains intact in TLR-3−/− mice , BMDC activation in TLR-3−/− cells following poly ( I:C ) stimulation is not inexplicable , but rather highlights the need for a better TLR-3 agonist . Taken as a whole and based on the fact that that RABV infection activated BMDC derived from both wt and TLR-3−/− mice equally , we conclude that TLR-3 signaling is not required for the activation of BMDCs following a RABV infection . To our knowledge , TLR-7 has never been investigated in the context of a RABV infection and thus the role that it plays in type I IFN induction and DC activation following RABV infection is unknown . To analyze the function that TLR-7 has in the induction of type I IFN and DC activation , we isolated BMDCs from MyD88−/− and C57BL/6 mice . We detected an equal upregulation of CD86 on BMDCs from MyD88−/− and wildtype mice ( Figure 4B ) . As expected , activation of MyD88−/− BMDCs is significantly reduced following stimulation with ODN1826 and R848 , ligands for TLR-9 and TLR-7/8 respectively , both of which signal via MyD88 [35] ( Figure 4D ) . Thus we conclude that , similar to TLR-3 signaling , the activation of BMDCs following a RABV infection occurs independently of MyD88 signaling . In order to determine if TLR-3 and MyD88 signaling might have an impact on type I IFN production , supernatant from infected BMDCs was collected at various times post infection , and a VSV-sensitivity assay was performed . As seen with BMDC activation , both TLR-3 and MyD88 are dispensable in the induction of type I IFN ( Table 1 ) . We did not detect any VSV-GFP replication on reporter cells following pre-treatment with supernatant from TLR-3−/− , MyD88−/− , or wildtype BMDC , indicating the presence of type I IFN in the supernatant . Having excluded TLRs as the required receptors mediating BMDC activation and type I IFN production , we next looked at the potential role for RLR signaling . Hornung et al . showed that a recombinant RABV expressing low levels of RABV-P signals via RIG-I to induce IFN-ß promoter activity following infection . Furthermore , it was shown that the 5′-triphosphate on the leader sequence of RABV was the ligand for RIG-I [24] . To determine whether the RIG-I pathway is also activated in DCs following RABV infection , we isolated BMDCs from IPS-1+/+ , +/− , or −/− mice . Our results indicate that following infection with RABV , IPS-1+/+ and IPS-1+/− BMDCs express high levels of CD86 on their surface ( Figure 5A–B ) . Of note , IPS-1+/− BMDCs are slightly less activated then IPS-1+/+ cells . On the other hand , IPS-1−/− BMDCs express significantly lower levels of CD86 on their surface at all time points ( Figure 5A–B ) . The TLR ligands LPS , ODN1826 , and R848 equally activated all IPS-1 BMDC samples , indicating that the defect in the IPS-1 −/− BMDCs is specific to the RLR pathways ( Figure 5B ) . As such , when cells are stimulated with RLR agonists , there is a defect in the activation of IPS-1−/− BMDCs when compared to IPS-1+/+ or +/− BMDCs . We see a low CD86 upregulation following both poly ( I:C ) stimulation and infection with a NS1-deficient strain of influenza ( ΔNS1/PR8 ) ( Figure 5B ) . It has been reported previously that poly ( I:C ) can signal via Mda-5 [33] and ΔNS1/PR8 signals exclusively via RIG-I [34] . Taken together this data indicates that BMDC activation is dependent on IPS-1 signaling following a RABV infection . In order to determine whether type I IFN production by BMDC is also dependent on IPS-1 mediated signaling , we assayed for the presence of type I IFN in the supernatants of infected IPS-1 BMDCs by VSV-GFP sensitivity assays and quantified the amount of IFN-ß by ELISA . It was seen that supernatant obtained from IPS-1+/+ and IPS-1+/− BMDCs infected with RABV was able to inhibit VSV-GFP replication , and thus contained type I IFN . On the other hand , the VSV-GFP replication on reporter cells was not inhibited by pre-treatment with supernatants from RABV infected IPS-1−/− BMDCs ( Figure 5C ) . Likewise , IPS-1 +/+ BMDCs produce on average 250 pg/ml IFN-ß while the IPS-1−/− BMDCs produced less than 16 . 7 pg/ml , if any , IFN-ß ( Figure 5C ) . These results indicate that RABV infected IPS-1−/− BMDCs do not secrete type I IFN . Also consistent with the results seen for BMDC activation , IPS-1−/− cells stimulated with RLR agonists produced less type I IFN compared to IPS-1+/+ or +/− BMDCs ( Table 2 ) . It has been shown that IPS-1 mediated pathways are also capable of activating the NF-κB signaling cascade [36] . Thus , we quantified the amount of IL-6 in the supernatant of RABV infected BMDC isolated from IPS-1+/+ , +/− and −/− mice ( Figure 5D ) . We see that there is a significant decrease in IL-6 produced by IPS-1−/− BMDCs compared to IPS-1+/+ BMDCs . However , IPS-1−/− cells do secrete some IL-6 following infection with RABV , and thus , the use of IPS-1 independent pathways to induce NF-κB activation , in contrast to type I IFN activation , seems to be utilized . Mda-5 mediated induction of IFN-ß has been described to occur in response to plus-stranded RNA viruses like picornaviruses , whereas it is reported that RIG-I is responsible for type I IFN induction in response to rhabdovirus infection [34] . However , the function of Mda-5 in the innate immune response to rhabdoviridae has not yet been elucidated . Furthermore , the role of these PRRs following a RABV infection in DCs remains unknown . Therefore we wanted to determine which of the two receptors recognizes RABV . For this approach , BMDCs from Mda-5−/− mice and RIG-I−/− mice were isolated . As shown in Figure 6A , Mda-5−/− BMDCs express high levels of CD86 on their surface at 24 and 48 hpi . Of note , there is a significant reduction of CD86 surface expression on Mda-5−/− BMDCs at 12 hpi when compared to wildtype cells . Likewise , RIG-I −/− BMDCs also have a defect in BMDC activation at 12 hpi , while CD86 expression at 24 and 48 hpi is equal for RIG-I−/− and RIG-I+/+ cells ( Figure 6B ) . In addition , it appears that while Mda-5−/− cells are able to induce type I IFN expression 12 hpi , RIG-I−/− cells have an early defect in type I IFN induction . Importantly , by 48hpi , RIG-I−/− BMDC do produce enough type I IFN to suppress VSV-GFP replication ( Figure 6C ) . This indicates that RABV can induce BMDC activation and type I IFN via both Mda-5 and RIG-I ligation . Furthermore , any perturbation in IPS-1 mediated signaling cascades seems to affect the early response ( 12hpi ) to RABV . Once type I IFN is produced , it will further activate the infected cell via autocrine signaling through IFNAR . Ligation of the IFNAR initiates the Jak/STAT signaling cascade , which culminates in the upregulation of antiviral genes . In addition to antiviral genes , Jak/STAT signaling also upregulates proteins required for type I IFN induction , thus providing a positive feedback for the type I IFN pathway [2] . In order to determine how much IFN induction is directly related to RABV infection and how much is due to positive feedback that is driven by IFN-α/ß production , we infected BMDCs derived from IFNAR−/− mice , which eliminates the contribution of positive feedback . Interestingly , BMDC isolated from IFNAR−/− mice produce enough type I IFN to block VSV-GFP replication on reporter cells after 12 , 24 , and 48 h ( Figure 7A ) . However , we detected a significant decrease in the CD86 cell surface expression of IFNAR−/− BMDC when compared to wt BALB/c mice ( Figure 7B–C ) . Thus , although RABV infection is sufficient to induce type I IFN , the cells need an amplification signal in order to undergo maturation . Additionally , we see a significantly greater infection by RABV in IFNAR−/− cells , presumably due to their inability to induce antiviral gene expression ( Figure 7D ) . Lastly , we wanted to determine the impact that the RIG-I and Mda-5 pathways play in the in vivo response to RABV utilizing IPS-1 −/− mice . Interestingly , we detected that IPS-1−/− BMDC , which do not produce type I IFN , have significantly more RABV-N expression post infection ( Figure 8A ) . This indicates that in the absence of IFN-α/ß induction , viral replication in DCs occurs at a faster rate , which should also increase viral pathogenicity . Therefore , we infected IPS-1 −/− , +/− , and +/+ mice , intramuscularly with SPBN-N2c , a recombinant RABV that is modestly pathogenic after peripheral inoculation [37] . Figure 8B shows that about 60% of the IPS-1+/+ or +/− mice lived , while only 45% of the IPS-1 −/− mice survived infection . More dramatically , nearly 90% of the IPS-1−/− mice had hind limb paralysis 11 days post infection while the IPS-1+/+ and +/− mice exhibited only about 45% paralysis ( Figure 8C ) . This data indicates that RABV infection of IPS-1−/− is more pathogenic than RABV infection in wildtype mice .
It has been previously seen that RABV can infect APCs [28] , [29]; however , the impact of the infection on generating an innate immune response to RABV had not been delineated . We show here that following RABV infection of APCs , unlike fibroblasts or neuronal cells , are able to produce copious amounts of type I IFN . We also determined that infected APCs do not produce novel viral progeny . A similar phenotype has also been seen following influenza infection of DCs . BMDCs become infected by the influenza strain , PR8 , as seen by co-expression of influenza HA and DC marker N418 on 72% of cells . However , infected BMDC do not release viral progeny , as seen by a failure of infected DC supernatants to induce hemagglutination of chicken red blood cells [27] . Non-productive infection of DCs may have significant biological relevance over the course of an infection . Since RABV infection within APCs is easily controlled , the cells become a source of viral antigen , with little risk of spreading infection to neighboring cells . Taken together , APCs seem to be of critical importance during a RABV infection both for the prolonged production of type I IFN as well as a source of viral antigen . In this study , we used APCs as a tool to study the PRRs used to recognize RABV following infection . Interestingly , we see that TLR-3 has no role in inducing a type I IFN response or DC activation despite its previously recognized upregulation following RABV infection [25] . However , recent publications may explain this potential discrepancy . It was reported that TLR-3 is required for the formation of Negri bodies in RABV infected cells and that these bodies are the site of viral replication [38] , [39] . Furthermore , TLR-3 −/− mice are less susceptible to infection with pathogenic RABV , as seen by increased survival and lower viral titers in the brains of TLR-3 −/− animals compared to wt mice [39] . Thus , the requirement for TLR-3 by RABV may explain why it is upregulated following infection despite the fact that it is not required for a type I IFN response . We next sought to identify whether TLR-7 was critical for DC activation and type I IFN production . To our knowledge , no one has directly examined the role of TLR-7 following a RABV infection . Of note , TLR-7 signaling does play a role in the cellular recognition of a closely related Rhabdovirus , VSV . Infection of wild type plasmacytoid DCs ( pDC ) with VSV induced the production of IFN-α . However , infection of pDCs from TLR7−/− or MyD88−/− mice resulted in no cytokine production [40] . Thus , indicating that single-stranded RNA derived from VSV is able to trigger TLR-7 signaling . However , in the case of RABV it appears that MyD88-dependent signaling , and thus TLR-7 , is dispensable for IFN-α/ß production following infection . On the other hand , RLR signaling via IPS-1 is critical for both the activation of DCs and production of type I IFN by infected DCs . It was shown previously that RIG-I signaling is necessary for IFN-ß promoter activity in VERO cells following recombinant RABV infection [24] . However , we show here using RIG-I−/− derived DCs that Mda-5 is also able to induce DC activation and type I IFN production . This is interesting , as Mda-5 is generally recognized as a receptor for positive stranded RNA viruses , not negative stranded RNA viruses like RABV . Of note , another negative stranded RNA virus of the Paramyxoviridae family , Sendai virus , requires MDA-5 signaling for the sustained expression of type I IFN [41] . Our data indicates that RABV can be recognized by either RIG-I or Mda-5 following infection . The use of both RIG-I and Mda-5 receptors has also been observed following infection with West Nile virus ( WNV ) . Following WNV infection , RIG-I−/− cells had a delayed upregulation of host anti-viral genes; however , the ability to respond was conserved . Thus , indicating that another receptor was involved in recognition of WNV , this receptor was identified as Mda-5 [42] , [43] . Following RABV infection in the absence of either RIG-I or Mda-5 , there is a delay in the activation of BMDCs . Furthermore , RIG-I−/− BMDCs have an early defect in type I IFN production . Thus , it appears that in response to a RABV infection , both RIG-I and Mda-5 are utilized in order to rapidly induce high levels of IFN-α/ß production and DC activation . Of note , IPS-1 +/− cells exhibited a phenotype that was intermediate to IPS-1 −/− and IPS-1 +/+ mice . This observation also supports the requirement for rapid induction of IFN-α/ß following a RABV infection . The heterozygous cells are lacking one of the IPS-1 alleles , and this may result in less functional protein in the heterozygous mice compared to homozygous wildtype mice . This again highlights the importance of a rapid response following viral infection in order to control viral replication and spread . The type I IFN response occurs in two phases after infection: the induction of IFN-α/ß following recognition of the pathogen by a PRR and then autocrine or paracrine signaling by IFN-α/ß through the IFNAR to induce upregulation of many other genes . Included among the genes that are upregulated in response to IFNAR signaling are several genes required for PRR signal transduction [2] . In this manner , the infected cell undergoes positive feedback to increase both the host response and PRR signaling . We wanted to identify which arm of the IFN response was responsible for the effects we observed following RABV infection of DCs , viral induction or IFN-α/ß amplification . We saw that both wt and IFNAR −/− mice are able to induce type I IFN production , thus highlighting the host's ability to rapidly induce IFN-α/ß following infection with RABV and indicating that the amplification of IPS-1 signaling by IFNAR signaling is not a critical factor in the induction of type I IFN . Surprisingly , we see that in the absence of IFNAR signaling , there is very little BMDC activation . Thus , it appears that DC activation occurs via IFN-α/ß signaling and is not a direct consequence of viral infection . This fact highlights the importance of a type I IFN response in initiating the adaptive immune response following infection with RABV . Lastly , we sought to determine the biological relevance of IPS-1 mediated PRR signaling following infection with a pathogenic strain of RABV . Although this experiment did not focus specifically on type I IFN production by DCs , it indicates how IPS-1 signaling , and thus IFN-α/ß production and DC activation , impacts the prognosis of infected animals . We saw that 87% of the IPS-1−/− mice in the study became paralyzed , whereas only about 45% of the IPS-1 +/+ or +/− mice exhibited signs of paralysis . There is some data that suggests paralysis following a RABV infection is an early symptom of disease . In humans who present with the less common paralytic rabies , their survival time is slightly longer [44] . Although not significant , this data supports the fact that an early , rapid type I IFN response is an important factor mediating RABV disease outcome . Of note , opposed to vaccine strain of RABV used in the BMDC experiments , the pathogenic RABV strain , SPBN-N2c , infects mostly neurons [37] and we showed here that RABV is able to suppress the type I IFN response in neurons by 12hpi ( Figure 1 ) . Despite this limitation , there is no other model to study RABV pathogenicity . The role that antigen presenting cells play in initiating the immune response to RABV in vivo should also be investigated further . It is known that pathogenic RABV is less immunogenic than vaccine strains of RABV [45] thus it is likely that pathogenic RABV avoids or alters infection of DC in order to elicit a lesser immune response . In summary , we show here that RABV replication is cell type dependent; namely , RABV is able to antagonize the induction of type I IFN in fibroblast and neuronal cells but is unable to inhibit IFN-α/ß induction in APCs . Furthermore , in APCs RABV infection is non-productive due to a defect in viral transcription , and no viral production is observed . Infection of BMDCs allowed us to delineate that RABV is exclusively recognized by either RIG-I or Mda-5 and both receptors are required for a rapid type I IFN response to RABV . This finding has significant implications for the development of a RABV-based vaccine vector . In light of these results , a recombinant RABV expressing a TLR agonist may allow for RABV recognition via TLRs . Such a response may potentiate the type I IFN response and induce better protection in a vaccine setting . We also show here that BMDC activation is secondary to IFN-α/ß induction and requires IFNAR . In addition , IPS-1 mediated signaling does have a role in vivo , as it seems to play a critical role in preventing RABV pathogenesis following RABV challenge .
All animals were handled in strict accordance with good animal practice as defined by the relevant international ( Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) ( Accreditation Status TJU: Full ) ) and national ( TJU Animal Welfare Assurance Number: A3085-01 ) , and all animal work was approved by the Institutional Animal Care and Use Committee ( IACUC ) at Thomas Jefferson University TJU . Animal use protocols are written and approved in accordance with Public Health Service Policy on Humane Care and Use of Laboratory Animals , The Guide for the Care and Use of Laboratory Animals . TJU IACUC protocol number 414A , 414G , and 414I were utilized in this study . The fibroblast cell line used in these studies is a cell clone of BHK-21 ( ATCC: CCL-10 ) , BSR . The neuronal cell line used in these studies is a neuroblastoma cell line referred to as NA [46] . The antigen presenting cell lines used here were JAWSII ( ATCC: CRL-11904 ) and Raw264 . 7 ( ATCC: TIB-71 ) . Mice used in this work are as follows: B6/129S1-Tlr3tm1Flv/J ( TLR-3 −/− , Jackson Laboratory , stock 005217 ) ; B6129SF1/J ( Jackson Laboratory , stock 101043 ) ; MyD88−/− [47]; C57BL/6 ( NIH ) ; IPS-1 [48]; Mda-5 −/− [33]; RIG-I [49]; BALB/c ( NIH ) ; and IFNAR−/− mice [50] . For pathogenicity study , IPS-1 mice were genotyped as described previously [48] . IPS-1 +/+ ( n = 7 ) , IPS-1 +/− ( n = 13 ) , and IPS-1 −/− ( n = 15 ) mice were infected intramuscularly with 106 ffu SBPN-N2c [37] . The weight of the mice was monitored daily , and the animals were euthanized after losing 25% of their body weight , which indicates a severe rabies infection . Cellular supernatants were assessed for the ability to inhibit vesicular stomatitis virus ( VSV ) replication as described previously [4] . Briefly , the cell line of interest was infected with the vaccine strain of RABV , SPBN , at a multiplicity of infection ( MOI ) of 10 , and supernatant was collected at various time points post infection . Alternatively , supernatant from infected BMDCs was used . The supernatants were UV-deactivated with a 254nm UV light source for 15 min . UV-deactivated viral supernatant was then diluted 1∶10 in RPMI-1640 and added to a reporter cell line ( either NA , for cell line experiments or 3T3 cells , for BMDC experiements ) . Following the 24 h pre-treatment , reporter cells were infected with VSV- expressing GFP at a MOI of 5 for 5–8 h . VSV replication was determined by fluorescence under a UV light source . For IFN-ß ELISA ( PML Laboratories ) the manufacturer's protocol was followed with the following modification: 50 µl of sample or standard was loaded into the 96-well plate . For IL-6 ELISA ( eBioscience ) the manufacturer's protocol was followed . Briefly , 5 µg/ml coating antibody was added to MaxiSorb ( Nunc ) plates and kept at 4°C over night . Wells were then washed with 0 . 05% Tween-20/PBS and blocked with Assay Buffer ( eBioscience ) for 2 hours . Plates were again washed with 0 . 05% Tween-20/PBS and then 100 µl standard or sample and 50 µl Biotin-Conjugate was added to the plate . Plates were incubated at room temperature for 2 hours , on a microplate shaker set at 200 rpm , and then washed with 0 . 05% Tween-20/PBS . Subsequently , wells were incubated with Streptavidin-HRP at room temperature for 1 hour , on a microplate shaker set at 200 rpm . The wells were washed and developed with 100 µl of Substrate Solution for 10 min followed by the addition of 100 µl of Stop Solution . Absorbance at 450 nm was recorded for each well . For both ELISAs a fourth-order non-linear regression curve ( Prism software , GraphPad version 4 . 00 ) was fit to the standard curve and used to determine the concentration of the unknown samples . BSR , NA , JAWSII and Raw264 . 7 cells were infected with SPBN at a MOI of 10 . Following 60 min incubation at 37°C , the virus was aspirated , and cells were washed twice with PBS to remove any virus that had not yet infected the cells . Media was then added to the cells , and , at indicated time points , 0 . 3ml of supernatant was removed and stored at 4°C . The aliquots were titered in duplicate on BSR cells . Messenger and genomic RABV-N RNA in SPBN ( MOI-10 ) infected BSR and JAWSII cells was determined by TaqMan probe-based real-time PCR as described previously [4] , [37] . Western blotting was performed as described previously [51] . BMDCs were differentiated as described previously [52] . Briefly , bone marrow ( BM ) was obtained from the mouse's tibia and femur . Following red blood cell lysis using ACK lysis buffer ( Invitrogen ) , the BM cells were cultured in 24-well costar plates at a density of 1 million cells per ml in the presence of 10ng/ml GM-CSF ( Peprotech ) . During the 7 day culture , the cells were washed once by aspirating 600µl of media from the wells and adding 1ml of fresh media supplemented with 10ng/ml GM-CSF . On the seventh day of culture , the non-adherent and semi-adherent cells were collected and used as the BMDCs population . BMDCs were plated in 12-well plates ( Nunc ) at a maximum density of 1 million cells per ml of media . BMDCs were infected with SPBN at an MOI of 10 or ΔNS1/PR8 [53] at an MOI of 1 . SPBN was harvested from BSR cells grown in serum free Opti-Pro media 4 and 7 dpi . Viral supernatant was pooled and spun at 1600 rpm for 10 min to remove cell debris . Alternatively , cells were left uninfected or stimulated with UV-deactivated SPBN , LPS ( 2µg/ml , Sigma ) , ODN1826 ( 2µM , InvivoGen ) , R848 ( 1 µg/ml , InvivoGen ) , or poly ( I:C ) ( 50 µg/ml , InvivoGen ) . Infected BMDC were kept at 37°C with 5% CO2 for 12 , 24 , or 48 h . Following differentiation of BMDC , cells were characterized for expression of DC markers . Briefly , cells were washed in FACS buffer ( 2% BSA/PBS ) and blocked at 4°C for 30–60 m with 2µl rat anti-mouse CD16/CD32 ( Fc block ) ( BD Biosciences Pharmigen ) in 100µl FACS . Cells were then stained with APC-CD11b , PerCP-B220 , and FITC-CD11c ( BD Biosciences Pharmingen ) for 30 min at RT . After staining , cells were washed with FACS buffer and fixed with Cytofix ( BD Biosciences ) for 16–18 hours at 4°C . Samples were washed and resuspended in 300 µl of FACS buffer . Samples were analyzed on BD FACS Calibur and a minimum of 50 , 000 events were counted . Following infection , BMDCs were analyzed for the expression of activation markers . At each given timepoint , BMDCs were removed from wells with cell scrappers and spun at 1600rpm for 5 min . Cells were then blocked at 4°C for 30–60 min with Fc block in 100µl FACS . Cells were then stained with APC-CD11c and PE-CD86 ( BD Biosciences Pharmingen ) for 30 min at RT . After staining , cells were washed with FACS buffer and fixed with Cytofix ( BD Biosciences ) for 16–18 hours at 4°C . Cells were then washed twice in Perm/Wash Buffer ( BD Bioscience ) and then stained with FITC-anti RABV-N ( Centacor , Inc ) for 30 min at RT . After staining , cells were washed with Perm/Wash buffer and then resuspended in 300 µl of FACS buffer . Samples were analyzed on BD FACS Calibur and 20 , 000–30 , 000 APC+ events were counted . All data were analyzed by Prism software ( GraphPad , version 4 . 00 ) . To compare two groups of data we used an un-paired , two-tailed T-test . For all tests , the following notations are used to indicate significance between two groups: *p<0 . 05 , **p<0 . 01 , ***p<0 . 001 .
|
Rabies virus ( RABV ) is a neurotropic RNA virus responsible for the deaths of the at least 40 , 000 to 70 , 000 individuals globally each year . However , the innate immune response induced by both wildtype and vaccine strains of RABV is not well understood . In this study , we assessed the pattern recognition receptors involved in the host immune response to RABV in bone marrow derived dendritic cells ( DC ) . Our studies revealed that Toll like receptor ( TLR ) signaling is not required to induce innate responses to RABV . On the other hand , we see that IPS-1 , the adaptor protein for RIG-I like receptor ( RLR ) signaling , is essential for induction of innate immune responses . Furthermore , we found that RIG-I and Mda-5 , both RLRs , are able to induce DC activation and type I interferon production . This finding is significant as we can target unused pattern recognition receptors with recombinant RABV vaccine strains to elicit a varied , and potentially protective , immune response . Lastly , we show that IPS-1 plays an important role in mediating the pathogenicity of RABV and preventing RABV associated paralysis . Overall , this study illustrates that RLRs are essential for recognition of RABV infection and that the subsequent host cell signaling is required to prevent disease .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology/vaccines",
"virology",
"virology/host",
"antiviral",
"responses",
"virology/immune",
"evasion",
"virology/host",
"invasion",
"and",
"cell",
"entry"
] |
2010
|
Rabies Virus Infection Induces Type I Interferon Production in an IPS-1 Dependent Manner While Dendritic Cell Activation Relies on IFNAR Signaling
|
Postzygotic single-nucleotide mosaicisms ( pSNMs ) have been extensively studied in tumors and are known to play critical roles in tumorigenesis . However , the patterns and origin of pSNMs in normal organs of healthy humans remain largely unknown . Using whole-genome sequencing and ultra-deep amplicon re-sequencing , we identified and validated 164 pSNMs from 27 postmortem organ samples obtained from five healthy donors . The mutant allele fractions ranged from 1 . 0% to 29 . 7% . Inter- and intra-organ comparison revealed two distinctive types of pSNMs , with about half originating during early embryogenesis ( embryonic pSNMs ) and the remaining more likely to result from clonal expansion events that had occurred more recently ( clonal expansion pSNMs ) . Compared to clonal expansion pSNMs , embryonic pSNMs had higher proportion of C>T mutations with elevated mutation rate at CpG sites . We observed differences in replication timing between these two types of pSNMs , with embryonic and clonal expansion pSNMs enriched in early- and late-replicating regions , respectively . An increased number of embryonic pSNMs were located in open chromatin states and topologically associating domains that transcribed embryonically . Our findings provide new insights into the origin and spatial distribution of postzygotic mosaicism during normal human development .
Postzygotic mutations refer to DNA changes arising after the formation of the zygote that lead to genomic mosaicisms in a single individual [1 , 2] . Unlike de novo or inherited germline mutations , postzygotic mutations only affect a fraction of cells in multicellular organisms , and individuals carrying a functional mosaic mutation typically exhibit a milder phenotype [3–5] . The roles of postzygotic single-nucleotide mosaicisms ( pSNMs ) have been demonstrated in numerous cancers [6 , 7] and various types of developmental disorders , including malformations [8 , 9] and autism [10 , 11] . We and another research group have reported the first genome-wide identification and characterization of pSNMs from the peripheral blood samples of healthy individuals [12 , 13] . More recently , the accumulation of postzygotic mutations during aging process has been reported in blood or brain samples [14–17] . Yadav et al . studied pSNMs in apparently benign tissue samples obtained from cancer patients [18] , but the contribution of pre-cancerous mutations could not be completely ruled out and the study was restricted to exonic regions . As such , the occurrence and genomic pattern of pSNMs in normal tissues of healthy individuals remains under-investigated . It has been reported that cancer genomes have distinct mutational signatures resulting predominantly from exposure to mutagenic agents and dysfunction of the DNA repair machinery [19] . Additional genomic factors , such as replication timing and chromatin status , could also impact the distribution of pSNMs in cancer genomes [20–22] . Whether and how these genomic factors might contribute to the genomic distribution of pSNMs in organs of healthy individuals remains largely unexplored [23] . Tumorigenesis has been considered as an evolutionary process in which tumor cells with increased fitness will proliferate faster than normal cells and lead to the clonal expansion of tumor cell population in a specific organ [24 , 25] . Although such events of clonal expansion have been previously reported in apparently normal skin and blood samples [17 , 26] , it remains unclear whether clonal expansion plays a role in other non-cancer tissue types . Understanding the origin and spatial distribution of pSNMs in normal tissues of healthy individuals could provide an important baseline for interpreting their contributions to disease states [27] . The next-generation sequencing technologies ( NGS ) have greatly advanced the study of pSNMs [28] . Sequencing the genomes of single cells after whole-genome amplification or in vivo clonal proliferation have been applied to the study of pSNM profiles of normal human cells , including germ cells [29] , adult stem cells [30] , and neurons [31] . Typically , tens or hundreds of cells from each sample need to be sequenced to identify and quantify pSNMs , which tends to increase the cost [32] . The inaccurate process of whole-genome amplification in single-cell sequencing makes it difficult to distinguish real pSNMs from technical artifacts , and the challenge of rigorously validating the pSNMs in a cell that have been already amplified aggravates the uncertainties [33 , 34] . Bulk sequencing is potentially a reliable and cost-effective alternative that , importantly , allows for rigorous validations of pSNMs [23] . Utilizing bulk whole-genome sequencing ( WGS ) and ultra-deep amplicon re-sequencing , this current study identified and validated pSNMs from 27 different organ samples obtained from five healthy donors and investigated the origin and spatial distribution of pSNMs in the developmental process of these healthy individuals .
Postmortem organ samples derived from five healthy Asian donors ( age 20–45 yr ) were obtained from BioServe , including a total of 27 organ samples from brain , liver , colon , skin , artery , breast , ovary , and prostate ( Table 1 ) . The donors died from motor vehicle accidents and were not known to be affected by any types of cancer or other overgrowth disorders . The samples were sequenced using an Illumina HiSeq X Ten sequencing platform with an average depth of 114-150X ( Table 1 ) . It is expected that many postzygotic mutations occurring at an early stage of embryogenesis may be shared between two or more organs from one individual [13 , 27] . Thus , conventional mutation callers , which require matched negative control samples for comparison , would likely miss these mutations . We had previously developed MosaicHunter [35] , a bioinformatics pipeline that can detect pSNMs without the need of control sample obtained from the same individual . MosaicHunter incorporated a Bayesian genotyper to distinguish pSNMs from germline variants and base-calling errors and a series of stringent filters to remove systematic errors . Using the Bayesian genotyper , we calculated the posterior probability of mosaic genotype versus three germline genotypes across all the genomic sites with at least 5% mutant allele fraction ( the fraction of reads supporting the mutant allele ) and 3 or more reads supporting the mutant allele . As a result , we identified a total of 251 candidate pSNMs in the 27 samples from the five donors; among them 41 pSNMs were found in more than one sample from the same donor ( Table 1 ) . Next , we validated the pSNMs and quantified their minor allele fractions in all of the organ samples ( Methods ) . We used an amplicon-based ultra-deep resequencing method , PASM ( PGM Amplicon Sequencing of Mosaicism ) , which we had previously developed and benchmarked [5] . Of the 251 candidate pSNM sites , 27 were excluded due to failure to design amplicon primers or to get enough sequencing depth in the negative controls . For the remaining 224 sites , the average sequencing depth of the amplicons was greater than 4000X per sample ( S1 Fig ) . The peripheral blood samples of two unrelated healthy Asians ( ACC1 and ACC4 ) served as negative controls . A pSNM was considered validated only if the mutant allele was detected in an organ sample in a mosaic state but undetectable in both the negative controls . Three sites with abnormal copy numbers estimated from the WGS data were further excluded ( S1 Table and Methods ) . In summary , we successfully validated 164 pSNMs in these five donors , with an overall validation rate of 73 . 2% ( Table 1 ) . The full list of the 164 validated pSNMs was described in S2 Table , which was used in the following analyses . The validated pSNMs were located in 21 autosomes and the X chromosome ( Fig 1A ) . We calculated the genomic distance between nearby pSNMs and found no significant difference between the observed and expected distances if pSNMs were uniformly distributed along the human genome ( Kolmogorov-Smirnov test , P-value > 0 . 05 ) , indicating that there was no observable clustering of postzygotic mutations in healthy individuals . The minor allele fractions estimated by PASM ranged from 1 . 0% to 29 . 7% , significantly correlating with the fractions estimated by WGS ( Fig 1B; Pearson’s r = 0 . 89 and P-value < 2 . 2×10−16 ) . The allele fraction of each pSNM varied across the different organs from the same donor ( S2–S6 Figs ) . Based on the presence or absence of the validated pSNMs in the organ samples from an individual donor , we grouped the pSNMs into two categories: 60 were present in two or more organ samples of the same donor ( 27 of which were globally present in all the sequenced organs of the donor ) and 104 were uniquely present in a single organ . Given the low postzygotic mutation rate in healthy individuals [36] , it was unlikely that multiple postzygotic mutation events involving the same nucleotide alteration occurred independently within one individual . It was more likely that the pSNMs shared by more than one organ resulted from mutation events that had occurred at early developmental stages , and the mutant alleles were passed on to cell lineages of more than one organ type . Comparison of the minor allele fractions in the two categories of pSNMs supported this hypothesis . As shown in Fig 2A , the allele fraction of pSNMs shared by more than one organ was significantly higher than that of pSNMs unique to only one organ ( Wilcoxon rank-sum test , P-value = 1 . 2×10−3 ) . In particular , 40% ( 24 out of 60 ) of the pSNMs shared by more than one organ had allele fractions greater than 1/16 , suggesting that they might have originated during the first few cell divisions of embryogenesis [37] . We refer to these pSNMs shared by more than one organ as “embryonic pSNMs” in the following analyses . On average , we identified 4 . 6~14 . 5 embryonic pSNMs from each organ of the five individuals , and the occurrence rate was similar across different organs ( Fig 2B ) . We further compared the allele fractions across multiple organs of the same individual , and found that more than 95% of the embryonic pSNMs showed <5% standard deviation of allele fraction ( S2 Table ) , indicating no dramatic allele fraction change for embryonic pSNMs . Close inspection of the pSNMs unique to only one organ revealed a distinctive type of pSNMs . Two organs had a dramatic excess of organ-unique pSNMs compared to other organs ( Fig 2B ) . Specifically , the liver sample of BBLD1005 and the breast sample of BBL11121 carried 42 and 32 organ-unique pSNMs , respectively , compared to an average of 1 . 1 organ-unique pSNMs for the other organ samples . This suggested that the majority of pSNMs in these two organs might originate organ-specifically after embryogenesis [30] . To further investigate these excessive organ-unique pSNMs in these two organs , we sampled three additional adjacent samples from each organ with varying physical distances to the original samples used for WGS ( S7 Fig ) and applied PASM to profile the allele fractions of validated pSNMs ( S8 and S9 Figs ) . While 8 of the 9 ( 88 . 9% ) embryonic pSNMs could be detected in all three intra-organ samples ( Fig 2C and 2D ) , consistent with our prediction that these mutations occurred early in embryogenesis , the organ-unique pSNMs manifested with a distinct intra-organ pattern . In BBLD1005 , 22 out of 42 ( 52% ) liver-unique pSNMs identified in the original sample ( liver #9 ) were also detected in the physically closest sample ( liver #8 ) , whereas only one liver-unique pSNM was detected in the two samples further away ( liver #2 and #5 ) ( Fig 2C ) . Given that the physical distance between liver #8 and #9 was about 0 . 5 cm and the distance between liver #2/#5 and liver #9 was approximately 3 . 5 and 2 cm , respectively , these results suggested that the majority of liver-unique pSNMs were locally restricted to a small volume of liver cells . A similar observation was made in the breast samples of BBL11121 that breast #7 shared more pSNMs to breast #9 than breast #2 and #5 ( Fig 2D ) . We reconstructed the inter-sample similarity using the minor allele fractions of pSNMs , and indeed the originally-sequenced liver or breast samples shared the largest similarity to their physically nearest samples ( Fig 2E and 2F ) . Analysis of minor allele fractions of the liver- and breast-unique pSNMs revealed a single narrow peak for each organ sample ( S10 Fig ) , with an average of 3 . 1% and 4 . 2% , respectively . Considering that such pSNMs were restricted to a small region within the organ , the narrow peaks likely resulted from clonal expansion events during the process of organ self-renewal that generated a sub-population of cells carrying postzygotic mutations large enough to be detected in bulk sequencing [23] . We refer to these pSNMs as “clonal expansion pSNMs” in the following analyses . Our results demonstrated the presence of clonal expansions in various types of non-cancer organs and highlighted clonal expansion as one of the major sources of pSNMs in clinically unremarkable individuals . If the embryonic pSNMs arose from early mutations during embryogenesis and the clonal expansion pSNMs arose from more recent mutations during organ self-renewal , they may present different mutational characteristics . To explore this possibility , we compared these mutations in terms of mutation spectrum , replication timing , and chromatin status . We first studied the mutation spectrum of the two types of pSNMs identified . For embryonic pSNMs , C>T mutations were the most predominant type ( 65 . 0% ) , with a significant elevated mutation rate at CpG sites vs non-CpG sites ( Proportion Z-test , P-value < 2 . 2×10−16 , Fig 3A ) . The enrichment of C>T mutation at CpG sites could be explained by the spontaneous deamination of 5-methylcytosines ( 5mC ) [20] , which has also been reported as one of the most common signatures in cancers [38] . The predominant C>T mutation at CpG sites for embryonic pSNMs were consistent with previous studies of early pSNMs in human [14 , 39] and mouse [40] . On the contrary , we observed predominant C>A ( 39 . 5% ) and T>C ( 42 . 4% ) mutations for the clonal expansion pSNMs identified in BBLD1005’s liver and BBL11121’s breast samples , respectively ( Fig 3B and 3C ) . Oxidative DNA damage was one of the major cause for C>A mutations [22] , and the higher proportion of C>A mutation in the liver sample could be explained by the accumulated oxidative stress of hepatocytes . Previous studies had reported elevated rates of germline and cancer-related somatic mutations in late-replicating regions [41 , 42] . Using data from the replication timing profile of lymphoblastoid cell-lines [43] , we observed significantly different distributions of replication timing between embryonic and clonal expansion pSNMs ( Wilcoxon rank-sum test , P-value = 9 . 7×10−3; Fig 3D ) . Clonal expansion pSNMs were significantly enriched in late-replication regions ( Permutation test , P-value = 0 . 006 ) , similar to previous reports of germline and cancer-related somatic mutations , while embryonic pSNMs were significantly enriched in genomic regions that replicated earlier ( Permutation test , P-value = 0 . 026 ) . Embryonic pSNMs with a wide range of allele fractions contributed to the early-replication enrichment ( S11 Fig ) , suggesting that the enrichment was not caused by a small number of outliers . This bimodal distribution was confirmed using the replication timing profiles from five other cell-lines: GM12878 , K562 , HeLa-S1 , HepG2 , and HUVEC ( Wilcoxon rank-sum test , P-value < 0 . 05 ) . We further confirmed our finding by using the pSNMs identified from the single-clone sequencing of neuronal progenitor cells [14] , where the mutations which were shared by other brain regions and non-brain tissues were significantly enriched in early-replicating regions than those specifically present in the clone of neuronal progenitor cells ( Wilcoxon rank-sum test , P-value = 0 . 028 ) . The distinct pattern of replication timing between the two types of pSNMs might reflect different mutational effects of replication timing during different stages of human development . Last but not the least , we investigated whether chromatin status contributed to the mutation rate of pSNMs . For embryonic pSNMs , the genomic distance between a pSNM and its closest DNase sensitive zone in embryonic stem cells was significantly smaller than the expectation under uniform distribution ( Permutation test , P-value = 0 . 013 ) . In contrast , clonal expansion pSNMs did not showed the enrichment of DNase sensitive zone ( Permutation test , P-value = 0 . 54 ) . We further found that embryonic pSNMs were significantly enriched in the topologically associating domains ( TADs ) containing embryonically-transcribed genes ( Fisher’s exact test , P-value = 0 . 046 ) , and this pattern was robust with different thresholds for embryonically-transcribed genes ( S12 Fig ) . Moreover , we observed a significantly larger proportion of embryonic pSNMs compared to clonal expansion pSNMs within transcribed chromatin regions using epigenetic data from three cell-lines of different origins ( Fisher’s exact test , P-value < 0 . 05; Fig 3E–3G ) . Analyses of tissue-shared pSNMs versus clone-specific pSNMs previously identified in neuronal progenitor cells [14] further confirmed our finding ( Fisher’s exact test , P-value < 0 . 01; S13 Fig ) . In summary , we reported an elevated rate of postzygotic mutations in open and transcribed chromatin regions during embryogenesis , which might result from the exposure of external or internal mutagens within these regions [44] .
Recent researches have significantly expanded what is known about the functional roles of postzygotic mutations , which now include not only cancers and overgrowth disorders [45] , but also other complex disorders [11 , 46] . With the help of next-sequencing technologies , postzygotic mutations have been identified and validated in healthy individuals [13 , 39] , confirming the theoretical predictions that postzygotic mutations are prevalent and every person is a mosaic [23] . However , the number of rigorously validated postzygotic mutations in healthy individuals has been small , which has hindered statistical analyses of their genomic patterns . In particular , little is known about the genomic patterns of postzygotic mutations in the normal development process of healthy human organs . In this study , we discovered two distinct types of pSNMs , one occurring during early embryogenesis and the other likely to occur during later tissue-specific clonal expansion . Surprisingly , these mutations manifested many distinct features in regard to mutation spectrum , replication timing , and chromatin status , implying dynamic mutational effects across different developmental stages . Unsurprising in hind sight , clonal expansion pSNMs shared many mutational features with previously reported cancer mutations [47] , as tumorigenesis is a specialized process involving clonal expansion of cancer cells [28] . Previous studies reported high proportion of C>A and T>C mutations as well as enrichment of late-replicating regions for clonal expansion pSNMs that were identified from skin fibroblasts [48–50] , which was concordant with our findings of clonal expansion pSNMs in the liver and breast samples . In contrast , our embryonic pSNMs demonstrated a range of unique features , including an elevated C>T mutation rate in CpG sites , an enrichment in early-replicating regions , and a stronger effect of transcribed chromatin status ( Fig 3 ) . Similar patterns in mutation spectrum ( S14 Fig ) , replication timing ( S15 Fig ) , and chromatin status ( S3 Table ) could be observed between the embryonic pSNMs that were globally present in all the sequenced organs of the donor and those only present in some but not all the sequenced organs . To further cross-validate our findings , we further analyzed an independent pSNM list that had been identified from human neuronal progenitor cells [14] , and confirmed the varied genomic patterns between pSNMs which originated at different developmental stages ( Results ) . In addition to WGS , the elevated C>T mutation rate in CpG sites was also reported in high-fraction pSNMs identified from whole-exome sequencing data [11 , 46] . Two of the reasons why the study of postzygotic mutations in healthy organs lags behind that of tumors include the lack of matched control samples in healthy individuals and the significantly lower abundance of postzygotic mutations . Our results showed that 38% of the validated pSNMs were shared by more than one organ , proving the importance of using a control-free pSNM-caller such as MosaicHunter . Furthermore , the high specificity of MosaicHunter compared to other callers enabled us to generate a candidate list that was specific enough to be validated . Single-cell sequencing has been demonstrated to be an alternative approach to study postzygotic mutations [14 , 15] . However , compared to single-cell sequencing , bulk sequencing is able to not only provide the genomic location of the postzygotic mutations but also their allele fractions ( Fig 1B ) , which are informative for assessing the proportion of cells that carry the mutation as well as reconstructing the lineage similarity across multiple samples within an individual ( Fig 2E and 2F ) . The list of pSNMs that had arisen locally during clonal expansion events in the liver and breast samples identified in our study deserve further discussion here . A cell clone with fitness advantage can predominantly proliferate faster and drive all private mutations that were originally carried by that clone to higher allele fractions , allowing them to be detected by bulk sequencing [26] . Because early embryonic pSNMs might affect only a fraction of cells in a certain organ , clonal expansion events could , in theory , make some pSNMs become undetectable from bulk sequencing if the carrier clone was out-competed . Indeed , we observed the breast #7 and breast #9 samples of BBL11121 had lost nine pSNMs that were detected in other breast samples and other organs of the same individual ( Fig 2D ) . These results demonstrated the dynamics of allele fraction for pSNMs driven by clonal expansion events in healthy individuals . We further screened 1407 cancer-related genes from BBLD1005’s liver samples using panel sequencing , and identified four more pSNMs with allele fraction around 1% ( S4 Table ) . However , none of the clonal expansion pSNMs had been previously reported in cancer studies and more functional experiments might be required to examine their relationship with clonal expansion . The current ~100X WGS bulk sequencing data in our study might not provide enough sensitivity to detect the whole spectrum of pSNMs , especially for those with allele fractions less than 1% . The genomic pattern we reported here were based on the analysis of eight organ types from five individuals . With reduced cost of NGS technology , we can expect a better-characterized spectrum of pSNMs in more and more organ samples and individuals in the future . A combination of deeper bulk sequencing and single-cell sequencing on the same organ sample could provide additional insights for pSNMs with lower allele fractions or even those present in only one or a few cells . This will enable a better characterization of postzygotic mutations in the human population and shed new light on distinguishing clinically-relevant postzygotic mutations from the genomic background .
Twenty-seven postmortem organ samples from five donors were obtained from BioServe Biotechnologies ( Beltsville , MD , USA ) , with an approved protocol from the Institutional Review Board of Medical Informatics Multi-Media Systems , Inc . ( 5-2-07 ) and written informed consent obtained from all participants or their legal guardians ( Table 1 ) . The clinical histories of all five donors showed no diagnosis of cancer or other known overgrowth disorders . Each organ sample was dissected into nine pieces ( roughly 0 . 5×0 . 5×0 . 5 cm each ) perpendicular to its long axis and labeled from #1 to #9 ( S7 Fig ) . The peripheral blood samples of two unrelated clinically unremarkable individuals of Asian descent ( ACC1 and ACC4 ) were collected with written informed consent and approval by the Institutional Review Board at Peking University ( IRB00001052-13025 ) . Genomic DNA was extracted using an AllPrep DNA/RNA Mini Kit ( Qiagen , Hilden , Germany ) after homogenization . Genomic DNA extracted from one piece ( labeled as #9 ) of each of the 27 organ samples was used for WGS and subsequent validation . To reduce potential bias introduced by library preparation , three sequencing libraries were constructed independently for each sample using a KAPA LTP Library Preparation Kit for Illumina platforms ( Kapa Biosystems , Wilmington , MA , USA ) . Size selection was performed for each library with a target insert size of 350–450 bp using a Pippin Prep system ( Sage Science , Beverly , MA , USA ) . Libraries were purified using Agencourt AMPure XP beads ( 1 . 0× volume; Beckman Coulter , Brea , CA , USA ) and underwent subsequent quality control using a 2100 Bioanalyzer ( Santa Clara , CA , USA ) . Each library was sequenced in one lane on an Illumina Hiseq X Ten platform ( Illumina , San Diego , CA , USA ) using 150-bp paired-end reads . Raw sequencing reads were aligned to the GRCh37 human reference genome using the paired-end mode of BWA [51] . The aligned reads were processed using Picard and GATK [52] for the removal of duplicated and ambiguous reads ( mismatch >4 ) , indel realignment , and base-quality recalibration . The average depths of processed reads in each organ sample ranged from 83X to 113X ( Table 1 ) . CNVs and indels were called using CNVnator [53] and GATK [52] , respectively , and all the involved regions as well as annotated repetitive regions were masked , because these regions were more vulnerable to false positives due to mis-alignment or abnormal copy numbers . To maximize the detection sensitivity of pSNMs , the single-sample and paired-sample modes of MosaicHunter [35] were applied to each organ sample with default parameters of genotyper and filters . For the paired-sample mode , the WGS data of the other organs obtained from the same donor served as paired controls . Candidates with at least 5% mutant allele fractions and 3 reads supporting the mutant allele were considered in our subsequent analyses . We randomly chose 16 candidates that are present in the latest versions of dbSNP [54] or the 1000 Genomes Project [55] to validate . As a result , all of them were genotyped as heterozygous rather than pSNMs , suggesting that they were more likely to be caused due to the large variation of allele fraction in WGS . Therefore , we only considered the candidates absent in both databases for thorough validation below . All the candidate pSNMs were analyzed using the standard workflow of PASM , which had been previously benchmarked by pyrosequencing and micro-droplet digital PCR [5 , 56] . PCR primers for PASM were successfully designed for 241 out of the 251 candidate pSNMs , with their amplicon lengths ranging between 380 and 420 bps; for a few candidates located in highly homologous genomic regions , two rounds of nested PCR were carried out to achieve higher amplification specificity ( S5 Table ) . For one-step amplicons , 35 cycles of PCR amplification was carried out using 2X Ex-Taq Premix ( Takara Bio , Dalian , China ) . For nested amplicons , additional 15 cycles of first-round amplification and 25 cycles of second-round amplification was carried out to capture the target region specifically . Amplicons from different organ samples and individuals were barcoded during library preparation and then pooled and sequenced using either the Ion Torrent PGM or Ion S5 XL sequencer ( ThermoFisher , Guilford , CT , USA ) , following the manufacturer’s protocols . Sequencing of the same PASM library on the PGM and S5 XL platforms showed a similar distribution of depth-of-coverage ( S16 Fig ) . In each sample , we calculated the 95% credible intervals of minor allele fractions for each candidate site with at least 30X depth [5] . Candidates with 95% credible intervals between 0 . 5% and 40% were considered a mosaic genotype , whereas those sites with a 95% credible intervals’ lower bound below 0 . 5% or an upper bound above 40% were considered a homozygous or heterozygous genotype , respectively . Five candidate sites without sufficient depth in the negative control samples were removed . As shown in S17 Fig , the inter-organ and intra-organ variation of the minor allele fractions for pSNMs could not be explained by the technical variance induced by DNA extraction or PASM . The differences in pSNM profiles across multiple samples from one individual were assessed using the Euclidean distance ( D ) of the square root of the allele fraction for all the validated sites . The relative similarity ( RS ) between sample i and j was defined as RSij=1−Dij−min ( D ) max ( D ) −min ( D ) CNVkit [57] was applied to estimate copy number from the WGS data , and the 100 bp windows centered by each pSNM were considered . The individual-specific copy number was then normalized using the mean across all the five individuals . Three of the 167 PASM-validated pSNMs were found to demonstrate abnormal copy number ( < 1 . 7 or > 2 . 3 for autosomes and females’ X chromosome and < 0 . 9 or > 1 . 1 for males’ X chromosome ) in the corresponding carriers ( S1 Table ) . The estimated copy numbers were 2 . 79 , 2 . 80 , and 2 . 59 for these three pSNMs ( Q7 , P35 , and N26 ) , respectively . Their mutant alleles were globally present in all the sequenced organs of the carrier , and the average allele fractions were close to 1/3 ( 26 . 2% to 30 . 9% ) . Rather than postzygotic mutations , these three sites were very likely to be explained by the germ-line events of copy number gain which made the allele fraction of involved heterozygous mutations deviate from 50% . Therefore , we excluded these three sites from all the following analyses . The genome-wide annotation of DNA replication timing was extracted from two independent studies [43 , 58] . For Hansen et al . , we downloaded the wavelet-smoothed signal datasets of five different cell-lines , including GM12878 , K562 , HeLa-S1 , HepG2 , and HUVEC [58] . For Koren et al . , the genome-wide profile was averaged from six lymphoblastoid cell-lines [43] . In both studies , a higher value represented an earlier DNA replication timing . For the permutation test , because our MosaicHunter pipeline only considered candidate pSNMs in non-repetitive regions , we compared the observed median of replication timing for each type of pSNMs against the null distribution estimated by using 1000 times of genome-wide random shuffling among non-repetitive regions . The DNase-seq data of the embryonic stem cell-line H1-hESC was downloaded from the ENCODE project [59] , and the DNase I sensitive zones were identified using the HotSpot algorithm [60] . For the permutation test , we randomly permutated the genomic positions of pSNMs among non-repetitive regions and assessed the median distance between each permutated pSNM and its closest DNase I sensitive zone . Each permutation was replicated 1000 times to estimate the distribution under the null hypothesis . The annotation of chromatin states in HepG2 , HMEC , and K562 cell-lines was downloaded from the UCSC Genome Browser [61] to represent cell types derived from the three primary germ layers in early embryo . Chromatin states were inferred from ChIP-seq data of ten epigenetic factors using a Hidden Markov Model [62] . The inferred chromatin states of active promoter , enhancer , and transcription were defined as “transcribed” chromatin status , whereas the other states involving repressed elements and heterochromatin were defined as “repressed” chromatin status . The annotation of TADs for the H1-hESC cell-line was downloaded from the ENCODE project [59] , which was generated based on the Hi-C data of Dixon et al . [63] . The expression profile of the same cell-line was also downloaded from the ENCODE project under GEO accession number GSM958733 . A gene was defined as embryonically transcribed if its FPKM was larger than 20 in H1-hESC . We also used other FPKM thresholds to confirm the robustness of our finding . To screen for potential driver mutations related to the clonal expansion events in BBLD1005’s liver samples , we captured 1407 cancer-related genes from the liver #2 and #8 samples of BBLD1005 using a Roche SeqCap panel ( Pleasanton , CA , USA ) designed by Genecast Biotechnology ( Beijing , China ) . The captured libraries were sequenced by Illumina Novaseq6000 ( Illumina , San Diego , CA , USA ) using 150-bp paired-end reads , with an average depth of ~2000X . Sequencing reads were aligned to the GRCh37 human reference genome by BWA . The pSNMs were identified using MosaicHunter , and the 95% credible intervals of mutant allele fraction were estimated by the Bayesian model implemented in PASM .
|
Genomic mosaicism led by postzygotic mutation is the major cause of cancers and many non-cancer developmental disorders . Theoretically , postzygotic mutations should be accumulated during the developmental process of healthy individuals , but the genome-wide characterization of postzygotic mosaicisms across many organ types of the same individual remained limited . In this study , we identified and validated two types of postzygotic mosaicism from the whole-genomes of 27 organs obtained from five healthy donors . We further found that the postzygotic mosaicisms arising during early embryogenesis and later clonal expansion events show distinct genomic patterns in mutation spectrum , replication timing , and chromatin status .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"liver",
"variant",
"genotypes",
"genetic",
"mapping",
"oncology",
"mutation",
"developmental",
"biology",
"epigenetics",
"dna",
"chromatin",
"dna",
"methylation",
"chromosome",
"biology",
"gene",
"expression",
"chromatin",
"modification",
"dna",
"modification",
"mutant",
"genotypes",
"carcinogenesis",
"biochemistry",
"cell",
"biology",
"anatomy",
"nucleic",
"acids",
"heredity",
"embryogenesis",
"genetics",
"biology",
"and",
"life",
"sciences"
] |
2018
|
Distinctive types of postzygotic single-nucleotide mosaicisms in healthy individuals revealed by genome-wide profiling of multiple organs
|
Exonization of Alu elements is a major mechanism for birth of new exons in primate genomes . Prior analyses of expressed sequence tags show that almost all Alu-derived exons are alternatively spliced , and the vast majority of these exons have low transcript inclusion levels . In this work , we provide genomic and experimental evidence for diverse splicing patterns of exonized Alu elements in human tissues . Using Exon array data of 330 Alu-derived exons in 11 human tissues and detailed RT-PCR analyses of 38 exons , we show that some Alu-derived exons are constitutively spliced in a broad range of human tissues , and some display strong tissue-specific switch in their transcript inclusion levels . Most of such exons are derived from ancient Alu elements in the genome . In SEPN1 , mutations of which are linked to a form of congenital muscular dystrophy , the muscle-specific inclusion of an Alu-derived exon may be important for regulating SEPN1 activity in muscle . Realtime qPCR analysis of this SEPN1 exon in macaque and chimpanzee tissues indicates human-specific increase in its transcript inclusion level and muscle specificity after the divergence of humans and chimpanzees . Our results imply that some Alu exonization events may have acquired adaptive benefits during the evolution of primate transcriptomes .
Alu is a class of primate-specific transposable elements that belongs to the short interspersed nuclear elements ( SINE ) family [1] . The rapid expansion of Alu during primate evolution has produced over one million copies of Alu elements in the human genome [2] . Until recently , Alu elements were considered as “junk DNA” , with no important functional or regulatory roles [1] . However , recent studies suggest a substantial influence by Alu elements on evolution of the human genome and regulation of gene expression [3] . Alu is a major source of new exons in primate genomes [4]–[6] . Alu elements have several sites resembling consensus splice sites in both sense and antisense orientations [7] . Therefore , the insertion of Alu elements into intronic regions may introduce new exons into existing , functioning genes . The evolutionary history of several such “exonization” events has been characterized in detail [8] , [9] . For example , in p75TNFR , the insertion of an Alu element and a series of subsequent nucleotide substitutions created a new alternative first exon [8] . Sorek and colleagues investigated the splicing pattern of 61 Alu-containing exons using human mRNA and EST sequences [4] . All Alu-containing exons were alternatively spliced . The vast majority of these exons were included in the minor transcript isoforms , based on ESTs pooled from all tissues [4] . This is consistent with the hypothesis that the creation of a new minor-form alternative exon reduces the initial deleterious effects of exonization events [10] . However , due to the high noise in EST sequencing [11] and the low EST coverage for these Alu-derived exons [4] , it was difficult to assess the splicing patterns of individual exons tissue by tissue . Regardless , there have been anecdotal reports for Alu-containing exons to have splicing patterns other than minor-form alternative splicing . Based on the tissue origins of human EST sequences , Mersch et al . predicted a few Alu-containing exons to be tissue-specific [12] . In another study , an Alu-containing exon of FAM55C was shown to be constitutively spliced in a neuroblastoma cell line [13] . These data suggest that the splicing profiles of exonized Alu elements may be more diverse than previously expected . In this study , we combined a genome-scale Exon array analysis with RT-PCR experiments to investigate the splicing profiles of exonized Alu elements in human tissues .
We collected a list of 330 Alu-derived exons , using annotations from the UCSC Genome Browser database [14] and Affymetrix human Exon 1 . 0 arrays ( see details in Materials and Methods ) . We first analyzed the splicing signals of these exons as well as their evolutionary rates during primate evolution . For the purpose of comparison , we also analyzed 13103 constitutively spliced exons and 5389 exon-skipping cassette exons in the human genome , which were collected after applying a set of stringent filtering criteria to exons in the Alternative Splicing Annotation Project 2 ( ASAP2 ) database ( see Materials and Methods ) . Our analysis showed that Alu-derived exons had significantly weaker splicing signals compared to constitutively spliced exons and typical cassette exons . For each exon , we scored its 5′ and 3′ splice site using models of consensus splice sites in MAXENT [15] . The median 5′ splice site score of Alu-derived exons was 7 . 35 , compared to 8 . 27 for cassette exons and 8 . 88 for constitutive exons , a statistically significant difference ( P = 3 . 0e-6 for Alu-derived exon vs cassette exons; P<2 . 2e-16 for Alu-derived exons vs constitutive exons; Wilcoxon rank sum test ) . We observed the same trend for the 3′ splice site . The median 3′ splice site score of Alu-derived exons was 6 . 79 , significantly lower than the scores of cassette exons ( 7 . 86 ) and constitutive exons ( 8 . 87 ) . In addition , Alu-derived exons had a lower density of exonic splicing regulatory elements ( ESRs ) . We used two sets of ESRs from the studies of Goren et al [16] and Fairbrother et al [17] . For each exon , we calculated the density of ESRs as the number of nucleotides covered by ESRs divided by the total length of the exon . The average ESR density of Goren et al was 0 . 484 on Alu-derived exons , compared to 0 . 500 on cassette exons and 0 . 532 on constitutive exons ( P = 0 . 04 for Alu-derived exon vs cassette exons; P = 6 . 5e-14 for Alu-derived exons vs constitutive exons ) . The same trend was observed for ESRs of Fairbrother et al: the average density was 0 . 144 on Alu-derived exons , which was significantly lower than the density on cassette exons ( 0 . 268 ) and constitutive exons ( 0 . 328 ) . We also found that Alu-derived exons had much higher evolutionary rates during primate evolution , compared to constitutive exons and cassette exons . Recently , the genome sequences of several non-human primates have become available . Therefore , we can study the sequence evolution of Alu-derived exons in primates after the initial Alu insertion events . To determine the evolutionary rate of different classes of exons , we analyzed the pairwise alignments of the human genome to the genomes of chimpanzee , orangutan , macaque and marmoset , which were increasingly distant from humans [18] . For exons present in both human and chimpanzee genomes , the overall nucleotide substitution rate of Alu-derived exons was 1 . 34% , compared to 0 . 73% for cassette exons and 0 . 52% for constitutive exons ( P≤2 . 2e-16 in Alu vs cassette exons and Alu vs constitutive exon comparisons , Wilcoxon rank sum test ) . Similarly , between human and orangutan genomes , the overall nucleotide substitution rates of Alu-derived exons , cassette exons and constitutive exons were 3 . 69% , 1 . 81% , and 1 . 31% respectively . The same trend was also observed in pairwise comparisons of human-macaque and human-marmoset genomes ( see Table 1 ) . We also obtained similar results when we restricted our analysis to exons smaller than 250 nt ( data not shown ) . These comparative analyses span the last ∼50 million years of primate evolution [18] . Taken together , these data are consistent with the hypothesis that the majority of primate-specific human exons derived from Alu elements are evolutionary intermediates without established functions [4] , [6] . The high evolutionary rate of Alu-derived exons observed in primate genome alignments probably reflects the combined effect of reduced negative selection pressure on non-functional Alu exons as well as positive selection pressure on Alu exons with adaptive benefits . However , distinguishing the effect of positive selection from that of the reduced negative selection is a difficult task in general [19] , [20] . Identifying the subset of Alu exonization events that have undergone positive selection using sequence-based approaches is particularly difficult for some practical reasons . Most Alu-derived exons are short ( median length of the 330 exons is 121 nucleotides ) . They are too new to have homologous sequences from distantly related species – homologous sequences of these exons may only exist in non-human primates . Thus , for most exons the number of nucleotide differences between homologous sequences is small , which significantly decreases the power of statistical tests . Although SNP-based approaches have been applied to genome-wide scans of positive selection on the human genome [21]–[26] , the regions identified by these studies are typically very large , making it a major challenge to locate the causal allele for positive selection [27] . In addition , SNP-based methods are sensitive to the temporal phases of positive selection [28] , influenced by the ascertainment bias [29] , and confounded by demographic factors [19] , [30]–[32] . For example , the Alu-derived exon of ADAR2 ( ADARB1 ) is a well-known case of functional exonization . This exon inserts an in-frame peptide segment into the catalytic domain of ADAR2 , altering its catalytic activity [33] . Using HapMap ( I+II ) SNP data [21] , [34] , we tested for the reduction of SNP heterozygosity , the skewed allele frequency spectrum with Tajima's D [35] and Fay and Wu's H [36] , and the increased population differentiation ( Fst ) [26] , [37] ( see details of the analysis in Text S1 ) . We did not observe evidence of positive selection on this ADAR2 exon using these metrics ( see Figure S1A ) . Similarly , SNP-based tests did not indicate evidence of positive selection for the alternative first exon of p75TNFR ( see Figure S1B ) , the result of another well-known functional exonization event [8] . These data show the limitation of using sequence-based approaches to identify functional Alu exonization events . A direct approach to assess the impact of individual Alu-derived exons on mRNA and protein products is to examine the splicing patterns of these exons in human tissues . Therefore , we proceeded with a large-scale splicing analysis of Alu-derived exons , using Affymetrix Exon array data of 330 exons in 11 human tissues and RT-PCR experiments of 38 exons , described in detail below . To examine the splicing patterns of Alu-derived exons , we used a public Affymetrix Exon 1 . 0 array data set on 11 human tissues ( breast , cerebellum , heart , kidney , liver , muscle , pancreas , prostate , spleen , testes , thyroid ) [38] , with three replicates per tissue . The Affymetrix human Exon 1 . 0 array is a high-density exon-tiling microarray platform designed for genome-wide analysis of pre-mRNA splicing , with over six million probes for well-annotated and predicted exons in the human genome [39] , [40] . Most exons are targeted by a probeset of four perfect-match probes . We compiled a list of 330 Exon array probesets targeting the 330 Alu-derived exons ( see details in Materials and Methods ) . In each of the 330 probesets , we had at least three probes to infer the splicing profile of the exon , after we filtered probes showing abnormal intensities ( Materials and Methods ) . Using a series of statistical methods that we developed for Exon array analysis [41] , [42] , for each probeset targeting an Alu-derived exon , we calculated the background-corrected intensities of its multiple probes and the overall expression levels of the gene in 11 tissues . These data were used to infer the splicing patterns of the exon . A large fraction of the 330 Alu-derived exons had low probe intensities in all surveyed tissues . Using a presence/absence call algorithm we developed for Exon array analysis , which compares the observed intensity of a probe to its predicted background intensity , we summarized a probeset-level Z-score for each exon in individual tissues as in [41] . A high Z-score suggests that the target exon is expressed . 174 ( 53% ) Alu exons had a Z-score of greater than 6 in at least one tissue , including 119 ( 36% ) exons whose Z-score was greater than 10 in at least one tissue . We also applied the same Z-score calculation to 37687 “background” probes on Exon array . These probes do not match any known genomic and transcript sequence in mammalian genomes [43] , so we can use their Z-score to estimate the false positive rate of the analysis . 5% of the background probes had Z-score greater than 6 in at least one tissue , including 3% whose Z-score was greater than 10 in at least one tissue . Based on these false positive rate estimates , at the Z-score cutoff of either 10 or 6 , we estimated that 33%–48% of the 330 Alu-derived exons in our study were expressed in some of the tissues . The remaining exons were not expressed at all or were expressed at very low levels in these 11 adult tissues . Of course , this is only a rough estimate , because the Z-score of individual probesets could be affected by a variety of microarray artifacts such as low probe-affinity or cross-hybridization [44] , [45] . Overall , these data are consistent with the observation that most Alu-derived exons had low transcript inclusion levels in EST databases [4] . Such Alu-derived exons may represent non-functional evolutionary intermediates that are rarely incorporated in the transcripts [9] . It is also possible that some of these exons are indeed expressed in other tissues or developmental states . Despite the low transcript abundance of many Alu-derived exons , a small fraction of exons showed highly correlated probe intensities with the overall expression levels of their corresponding genes across the surveyed tissues , suggesting stable exon inclusion . We found 19 Alu-derived exons where three probes or more correlated with gene expression levels , including the well-characterized Alu-derived exon in ADAR2 ( ADARB1 ) that inserts an in-frame peptide segment to ADAR2's catalytic domain [5] . Detailed descriptions of these 19 exons are provided in Table 2 . Among the 19 “correlated” exons , 12 were in the 5′-UTR . One exon was in 3′-UTR and one exon was part of a non-coding transcript . The remaining five exons were in coding regions , including two that introduced premature termination codons . This distribution is consistent with the hypothesis that most functional Alu exonization events do not contribute to the proteome but may play a role in regulating gene expression [46] , [47] . Similar to the finding by a recent study of species-specific exons [48] , we observed an excess of Alu-derived internal exons in 5′-UTR as compared to 3′-UTR . This may reflect stronger negative selection pressure against exon creation in 3′-UTR because such exons could trigger mRNA nonsense-mediated decay . The 5′-UTR Alu exons may influence the transcriptional or translational regulation of their host genes , as suggested by Goodyer and colleagues [49] . Several types of splicing patterns could explain the observed correlation between probe intensities and estimated gene expression levels . These “correlated” exons could be constitutively spliced , or alternatively spliced at similar levels across tissues , or alternatively spliced but with certain variations in exon inclusion levels from tissue to tissue . However , we could not distinguish these situations based on Exon array data alone , since uncertainties in microarray probe affinity [44] prevent estimations of the absolute transcript abundance of individual exons . To uncover the exact splicing patterns of the “correlated” exons we analyzed all 19 exons by RT-PCR , using RNAs from all available tissues surveyed by Exon array ( purchased from Clontech , Mountain View , CA ) except breast tissue . For each exon , we designed RT-PCR primers targeting its flanking constitutive exons . The identities of all PCR products close to the expected sizes of exon inclusion or skipping forms were further confirmed by sequencing ( Materials and Methods ) . We discovered three major categories of splicing patterns in these 19 exons ( Table 2 ) . Six exons ( in FAM55C , NLRP1 , ZNF611 , ADAL , RPP38 , RSPH10B ) were constitutively spliced . For example , the four probes of an Alu-derived exon in NLRP1 had a minimal correlation of 0 . 86 with the expression levels of NLRP1 in the Exon array data ( Figure 1A ) . Our RT-PCR analysis showed a single isoform corresponding to the exon inclusion form in all surveyed tissues ( Figure 1B ) . In FAM55C , an Alu-derived exon was shown previously to be included in the only isoform product in a human neuroblastoma cell line [13] . We found all four probes of this FAM55C exon had a minimal correlation of 0 . 78 with the overall gene expression levels ( Figure 1C ) . Our RT-PCR experiments showed that this exon was constitutively spliced ( Figure 1D ) . In another three tested genes ( SLFN11 , NOX5 , B3GALNT1 ) , the Alu-derived exons were alternatively spliced , but the transcript inclusion levels varied in individual tissues . For example , the SLFN11 exon was included in the major transcript product in most tissues but appeared as the minor form in pancreas . We observed Alu exon inclusion isoforms of varying lengths that resulted from alternative splice site usages of the Alu-derived exon and its upstream alternative exon ( Figure 1E ) . In NOX5 , a single exon-inclusion isoform was detected in most tissues , but an additional exon-skipping isoform was detected in liver , pancreas and testes ( Figure 1F ) . In the remaining 10 tested genes , the exons were alternatively spliced with varying levels of transcript inclusion , but no exon showed evidence of tissue-specificity in our semi-quantitative RT-PCR analyses ( see Table 2 and Figure S2 ) . We also conducted RT-PCR analyses of 11 “uncorrelated” exons ( Table S1 ) . The lack of correlation between probe intensities of an exon and overall gene expression levels can be due to a number of reasons . If the target Alu-derived exon has very low transcript inclusion levels , or if the probes have poor binding affinity to the target exon , the intensities of the microarray probes could be largely saturated by microarray noise , resulting in poor correlation with the overall gene expression levels . It is also possible that the correlation pattern of a highly expressed Alu-derived exon is obscured due to microarrray artifact ( such as cross-hybridization ) in a subset of samples . Thus , by analyzing “uncorrelated” exons , especially those with high probeset-level Z-scores in individual tissues , we may discover additional Alu-derived exons with high transcript inclusion levels . Indeed , among six RT-PCR tested “uncorrelated” exons whose probeset-level Z-score was greater than 7 in at least three tissues , we found two constitutive exons , three exons with medium to high transcript inclusion levels , as well as one exon in the minor transcript isoform ( see Table S1 and Figure S3 ) . By contrast , among five exons whose probeset-level Z-score was smaller than 3 in all 11 tissues ( suggesting weak exon inclusion ) , four exons had very weak exon-inclusion transcripts in all surveyed tissues . The exon in FAM124B had medium transcript inclusion levels ( see Figure S3 ) . Taken together , our RT-PCR analysis of 19 “correlated” exons and 11 “uncorrelated” exons indicates that a subset of Alu-derived exons have acquired strong splicing signals , so that they are included in the transcript products at high levels . Moreover , while prior EST-based analyses suggested all Alu-derived exons to be alternatively spliced [4] , we provide experimental evidence that some Alu-derived exons are constitutively spliced in a broad range of normal human tissues . Our analysis of the “correlated” Alu exons revealed that some exons had varying transcript inclusion levels in different tissues . It is possible that exons with strong tissue-specific splicing patterns do not have highly correlated intensities with the overall gene expression levels , and were missed by the above analysis . Therefore , we combined computational analysis and manual inspection of Exon array data to specifically search for tissue-specific exons ( see Materials and Methods ) . We selected three exons ( in ICA1 , ZNF254 , FAM79B/TPRG1 ) that appeared to exhibit strong tissue-specific splicing patterns for RT-PCR . We also selected five other Alu-derived exons with prior experimental evidence for exon inclusion in at least one tissue or cell line [9] , [12] , [50] , [51] , regardless of whether reliable Exon array probes existed for these exons ( Table 3 ) . Our RT-PCR experiments detected four exons with tissue-specific splicing patterns ( also see Figure S4 for the other four exons with no tissue-specificity ) . In ICA1 , the Exon array data suggested testes-specific exon inclusion ( Figure 2A ) . The RT-PCR analysis detected a strong band corresponding to the exon inclusion form specifically in the testes ( Figure 2B ) . In ZNF254 , the RT-PCR analysis indicated strong exon inclusion in cerebellum , which was consistent with the Exon array profile ( Figure 2C–D ) . We also found that this exon was almost completely skipped in pancreas , although this pattern was not observed in the Exon array data . In PKP2 , the exon inclusion form was shown to be the minor isoform in HT29 , a colon cancer cell line [9] . Our RT-PCR result showed that this exon was skipped in all other surveyed tissues but was included in the minor transcript product in the pancreas ( Figure 2E ) . Some tissue-specific Alu-derived exons have interesting functional implications . For example , SEPN1 encodes selenoprotein N , 1 , which is expressed in skeletal muscle and has been suggested to play a role in protection against oxidant damage [50] . Mutations in SEPN1 were linked to a form of congenital muscular dystrophy [50] . SEPN1 is expressed as two alternatively spliced isoforms . The full-length isoform contains an Alu-derived exon , which is predicted to be the minor isoform based on EST data . The Alu-derived exon contains a second in-frame TGA selenocysteine residue . However , the protein product corresponding to the exon inclusion isoform was not detected by Western blot in the HeLa cell [52] . Our RT-PCR result indicated a strong muscle-specific increase in the inclusion level of this Alu-derived exon ( Figure 2F ) . It will be interesting to investigate whether this splicing pattern represents a mechanism for modulating SEPN1 activity in muscle . To further elucidate the evolution of this muscle-specific Alu exon in SEPN1 , we obtained matching macaque and chimpanzee tissues and analyzed the splicing pattern of this exon in primate tissues using semi-quantitative RT-PCR as well as realtime quantitative PCR ( see Materials and Methods ) . RT-PCR analysis of this exon in macaque tissues showed no exon inclusion ( see Figure 3B ) , consistent with the fact that this Alu exon was absent from the corresponding SEPN1 region in the rhesus macaque genome . In chimpanzees , both exon inclusion and skipping forms were produced , but the exon inclusion levels were significantly lower compared to human tissues based on the RT-PCR gel pictures ( Figure 3B ) . The splicing difference of this SEPN1 exon between humans and chimpanzees was further confirmed by realtime qPCR using isoform-specific primers ( Figure 3C–D ) . These data depict the evolutionary history during the creation of an Alu-derived primate-specific exon and the establishment of its tissue-specific splicing pattern . Our results suggest that the strong transcript inclusion and muscle-specificity of the human SEPN1 exon was acquired after the divergence of humans and chimpanzees . In this study , we conducted RT-PCR analysis of 38 Alu-derived exons in 10 human tissues . 26 of the 38 exons had at least medium inclusion levels in certain tissues . These exons are in genes from a wide range of functional categories ( see the complete list in Table S2 ) . Analyses of these 26 exons revealed several interesting characteristics . 23 of the 26 exons were derived from the antisense strand of Alu elements , among which 14 were from the right arm of the antisense Alu ( see Figure S5 ) , consistent with a recent report that the right arm of Alu antisense strand is a hotspot for exonization [53] . Moreover , of these 26 exons , 23 were from AluJ class and 3 were from the AluS class . By contrast , in the total set of Alu-derived exons in our study , 211 were from AluJ and 111 were from AluS , a 4-fold shift in the ratio of AluJ to AluS ( 7 . 7 in the “substantially included” set versus 1 . 9 in the total set; P = 0 . 01 , one-tailed Fisher exact test ) . In the human genome , AluJ is outnumbered by AluS at a ratio of 1 to 2 . 3 [14] ( Figure 4 ) . The similar trend was also found in the 19 “correlated” exons; 16 were from the AluJ class and 3 were from the AluS class . Taken together , these data are consistent with the fact that AluJ is the oldest Alu subclass in the human genome [54] , so that exons derived from AluJ elements had more evolutionary time to accumulate nucleotide changes that strengthened exon inclusion in the transcript products . We did not observe a significant difference in the splice site score and ESR density of the 26 substantially included Alu exons compared to other Alu-derived exons ( data not shown ) . This could be due to the lack of statistical power . Alternatively , it may reflect the current lack of knowledge of the complete set of cis-elements that regulate splicing [55] , [56] . Future experimental studies ( such as mini-gene experiments ) are needed to dissect the exact regulatory elements important for strong transcript inclusion and/or tissue-specific splicing of individual Alu-derived exons .
Our study reveals diverse splicing patterns of exonized Alu elements in the human transcriptome . Most new exons originated from Alu elements probably represent non-functional splice forms that are included in the transcripts at low frequencies [4] , [6] . However , a small subset of exonization events , in particular those associated with more ancient Alu elements , could evolve strong splicing regulatory signals to become constitutive or tissue-specific , possibly driven by positive selection . The analysis of high-density exon tiling array data across a broad range of tissues provides an efficient approach to identify such exons . Considering the incomplete coverage of Exon 1 . 0 arrays on human transcribed regions , and the high noise in the observed intensities of probes targeting individual exons [57] , [58] , we expect that many constitutive or tissue-specific Alu-derived exons are missed by this study . Also , while we focus on primate-specific exons derived from Alu repeats , a recent study by Alekseyenko and colleagues identified nearly 3000 human-specific exons created by de novo substitution in intronic regions during primate evolution [59] . With improved exon microarray platforms and analysis algorithms in the future , more species-specific exons with regulatory roles are likely to be discovered . Our data provide novel insight into the evolutionary impact of newly created exons in eukaryotic genomes . During evolution , new exons are frequently added to existing functioning genes via a variety of mechanisms , such as exonization of transposable elements , exon duplication , and de novo exonization from intronic regions [6] . Modrek and Lee found that the birth of new exons was strongly coupled with widespread occurrence of alternative splicing in eukaryotic genes [60] . Through pairwise comparisons of human and rodent genomes , they showed that nearly 75% of human alternatively spliced exons with low transcript inclusion levels were absent from the corresponding genomic sequence of the rodent orthologs . By contrast , the number was less than 5% for constitutive exons [60] . This pattern was corroborated by subsequent analyses of exon creation events in vertebrates using multiple genome alignments [48] , [59] . Based on these observations , Modrek and Lee proposed an evolutionary model that alternative splicing can facilitate the evolution of new exons – the creation of a new exon in the minor transcript isoform keeps the original gene product intact , which reduces the negative selection pressure against the new exon , allowing it to evolve towards an adaptive function [10] , [60] . On the other hand , this evolutionary model also predicts that the vast majority of new exons found by comparative genomics analyses are non-functional evolutionary intermediates . In fact , most previous genomic studies have focused on the low transcript inclusion levels of new exons [4] , [6] , [48] , [59] , [60] . It is unclear to what extent new exons could have produced functional and regulatory novelties . In this study , based on a large-scale splicing analysis of human tissues , we show that a number of primate-specific exons derived from Alu retrotransposons have a major impact on their genes' mRNA/protein products in a ubiquitous or tissue-specific manner . In SEPN1 , the strong transcript inclusion and muscle-specificity of the Alu derived exon represents a human-specific splicing change after the divergence of humans and chimpanzees . These data suggest that some new exons may contribute to species-specific differences between humans and non-human primates . Our study has discovered a large list of Alu-derived exons with substantial transcript inclusion levels . This exon list can be valuable for a variety of further investigations . These exons provide candidates for detailed mechanistic analyses and can be used to characterize the splicing regulatory mechanisms of Alu-derived exons . If suitable tissue samples from closely or distantly related primate species are available , it will be possible to precisely reconstruct the evolutionary events preceding the emergence of constitutive or tissue-specific Alu-derived exons . Further experimental studies will be needed to elucidate the functional significance of individual exonization events ( e . g . the muscle-specific inclusion of the Alu-derived exon in SEPN1 ) .
We downloaded a public Affymetrix Exon 1 . 0 array data set on 11 human tissues ( breast , cerebellum , heart , kidney , liver , muscle , pancreas , prostate , spleen , testes , thyroid ) [38] , with three replicates per tissue ( http://www . affymetrix . com/support/technical/sample_data/exon_array_data . affx ) . We compiled a list of Exon array probesets targeting exonized Alu elements . The locations of Alu elements in the human genome were downloaded from RepeatMasker annotation of the UCSC Genome Browser database [14] . The locations of internal exons ( i . e . exons flanked by both 5′ and 3′ exons ) in human genes were taken from the UCSC KnownGenes database [14] . This database combines transcript annotations from multiple sequence databases [14] . To eliminate long exonic regions likely resulting from intron retention events , we removed probesets whose probe selection regions were over 250 bp as in [61] . We then defined an exon as Alu-derived if the Alu element covered at least 25 bp of the exon and over 50% of the total length of the Exon array probe selection region . We collected 526 Exon array probesets targeting such Alu-derived exons . Since microarray probes targeting Alu repeats may cross-hybridize to off-target transcripts , we used a conservative approach to identify and remove individual probes showing abnormal intensities ( see “Analysis of Exon array data” below ) . After probe filtering , we collected a final list of 330 Exon array probesets , with at least three reliable probes in each probeset to infer the splicing profiles of Alu-derived exons . We collected 13103 constitutively spliced exons and 5389 exon-skipping cassette exons in the human genome , after applying stringent filtering criteria to exons in the Alternative Splicing Annotation Project 2 ( ASAP2 ) database [62] . ASAP2 determined the splicing patterns of human exons based on the analysis of mRNA/EST sequences [62] . Constitutive exons were defined as those without any evidence of exon skipping in mRNA/EST data . To ensure that no skipping form was missed due to incomplete transcript sampling in EST databases , each constitutive exon included in our study was required to have at least 50 exon inclusion ESTs . We obtained 13103 high-confidence constitutive exons using this criterion . For exon skipping cassette exons , we collected 5389 ASAP2 exons with at least 3 inclusion ESTs and at least 3 skipping ESTs . For each exon , we scored its 5′ and 3′ splice sites using consensus splice site models in MAXENT [15] . For 5′ splice site , we analyzed 3 nucleotides in exons and 6 nucleotides in introns . For 3′ splice sites , we analyzed 3 nucleotides in exons and 20 nucleotides in introns . We also calculated the density of exonic splicing regulatory elements ( ESRs ) . Two sets of elements were used separately: ( i ) 285 exonic splicing regulatory elements from Goren et al [16]; ( ii ) 238 exonic splicing enhancers from Fairbrother et al [63] . For each exon , the ESR density was calculated as the number of nucleotides covered by ESRs , divided by the total length of the exon . To determine the nucleotide substitution rate of exons in primates , we downloaded and analyzed the UCSC pairwise genome alignments of the human genome ( hg18 ) to the genomes of chimpanzee ( panTro2 ) , orangutan ( ponAbe2 ) , rhesus macaque ( rheMac2 ) and marmoset ( calJac1 ) [64] . In each pairwise alignment , we defined an exon to be conserved in a non-human primate if there was at least one homologous region that covered at least 80% of the human exon with at least 80% sequence identity . We included a conserved exon in the nucleotide substitution rate analysis if there was a single ( unambiguous ) homologous region in the genome alignment . For such exons , we calculated the nucleotide substitution rate between the human genome and the genome of a non-human primate as the number of conserved nucleotide within the aligned region , divided by the total length of the aligned region . The alignment analysis was performed using Pygr [65] , a python bioinformatics library that provided efficient access to alignment intervals in the UCSC genome alignments . Briefly , we first predicted the background intensities of individual Exon array probes , using a sequence-specific linear model [41] , [66] trained from “genomic” and “anti-genomic” background probes on the Exon 1 . 0 array [43] . For every probe , the predicted background intensity was an estimate for the amount of non-specific hybridization to the probe . This background intensity was subtracted from the observed probe intensity before downstream analyses [41] . Second , for each gene we used a correlation-based iterative probe selection algorithm to construct robust estimates of overall gene expression levels , independent of splicing patterns of individual exons [42] . Third , since oligonucleotide probes for Alu-derived exons may be more likely to cross-hybridize than typical Exon array probes , we used two independent methods to identify and remove individual probes with abnormal probe intensities . We searched all 25mer oligonucleotide probes against all RefSeq-supported exon regions , allowing up to 3 bp mismatches . Once a potential off-target gene was found for a probe , we calculated the Pearson correlation coefficient between the probe's intensities and the off-target gene's estimated expression levels across the 11 tissues [45] . We defined a probe to be cross-hybridizing if there was an off-target gene within 3 bp mismatches , and if the computed Pearson correlation coefficient was above 0 . 55 . Such probes were removed from further analyses . We also detected probes whose intensities were higher than 95% of all other probes for RefSeq-supported exons of the same gene in at least 3 of the 11 tissues . Such probes were regarded as outlier probes and were also removed . After probe filtering , we collected a final list of 330 Exon array probesets , with at least three reliable probes in each probeset to infer the splicing profiles of Alu-derived exons . For each Alu-derived exon , using a presence/absence call algorithm that compares the observed intensity of a probe to its predicted background intensity , we summarized a probeset-level Z-score for exon expression in individual tissues as in [41] . We also calculated the Pearson correlation co-efficient of individual probes' intensities with the overall gene expression levels in 11 tissues ( estimated from all exons of a gene , see [41] , [42] ) . We defined a probe to be “correlated” with gene expression levels if the Pearson correlation co-efficient was above 0 . 6 . We defined an exon to be “correlated” if it had at least three probes correlated with gene expression levels . We used a two-step approach to identify strong tissue-specific exons , by combining computational analysis and manual inspection of Exon array data . For each probe of an exon in a tissue , we calculated a “splicing index” , defined as the background-corrected probe intensity divided by the estimated gene expression level [40] . We used a Z-score method used by Graveley and colleagues [67] to test whether the splicing index of a particular tissue was an outlier compared to other tissues . A highly positive Z-score suggests tissue-specific exon inclusion . After this initial computational screening , we manually inspected the Exon array data of potential tissue-specific exons . Total RNA samples from 10 human tissues were purchased from Clontech ( Mountain View , CA ) . Single-pass cDNA was synthesized using High-Capacity cDNA Reverse Transcription Kit ( Applied Biosystems , Foster City , CA ) according to manufacturer's instructions . For each tested Alu-derived exon , we designed a pair of forward and reverse PCR primers at flanking constitutive exons using PRIMER3 [68] . Primer sequences and positions are described in Table S3 . Two µg of total RNA were used for each 20 ul cDNA synthesis reaction . For each candidate Alu exonization event , 1 µl of cDNA were used for the amplification in a 25 µl PCR reaction . PCR reactions were run for 40 cycles in a Bio-Rad thermocycler with an annealing temperature of 62°C . The reaction products were resolved on 2% TAE/agarose gels . All of the candidate DNA fragments corresponding to exon inclusion and exon skipping forms were cloned for sequencing using Zero Blunt TOPO PCR Cloning Kit ( Invitrogen , Carlsbad , CA ) . Total RNA samples from rhesus macaque tissues ( brain , skeletal muscle , pancreas ) were purchased from Biochain Inc ( Hayward , CA ) . Frozen tissue samples ( cerebellum , skeletal muscle , liver , kidney ) of two chimpanzees were generously provided by Southwest National Primate Research Center ( San Antonio , TX ) . RNA was prepared using TRIzol ( Invitrogen ) according to the manufacturer's instructions . Single-pass cDNA was synthesized using the High Capacity cDNA Reverse Transcription Kit ( Applied Biosystems , Foster City , CA ) . The quantitative real-time polymerase chain reaction ( qRT-PCR ) was performed using Power SYBR Green PCR Master Mix ( Applied Biosystems , Foster City , CA ) . The following primers were used in qRT-PCR: SEPN1 Exon 3 skipping form: forward: 5′-GGGACAGATGGCCTTTTTCT-3′; reverse: 5′-AGTTGACCCTGTTAGCTTCTCAG-3′ ; SEPN1 Exon 3 inclusion form: forward 5′- GGAGTTCAAACCCATTGCTG -3′; reverse: 5′- AATTGAGCCAGGGAAGTTGA -3′ . These qPCR primers match perfectly to their transcript targets in human and chimpanzee . Using a mathematical method described by Pfaffl [69] , we calculated and presented the SEPN1 exon 3 inclusion level as a ratio to the exon 3 skipping level in each sample .
|
New exons have been created and added to existing functional genes during eukaryotic genome evolution . Alu elements , a class of primate-specific retrotransposons , are a major source of new exons in primates . However , recent analyses of expressed sequence tags suggest that the vast majority of Alu-derived exons are low-abundance splice forms and represent non-functional evolutionary intermediates . In order to elucidate the evolutionary impact of Alu-derived exons , we investigated the splicing of 330 Alu-derived exons in 11 human tissues using data from high-density exon arrays with multiple oligonucleotide probes for every exon in the human genome . Our exon array analysis and further RT-PCR experiments reveal surprisingly diverse splicing patterns of these exons . Some Alu-derived exons are constitutively spliced , and some are strongly tissue-specific . In SEPN1 , a gene implicated in a form of congenital muscular dystrophy , our data suggest that the muscle-specific inclusion of an Alu-derived exon results from a human-specific splicing change after the divergence of humans and chimpanzees . Our study provides novel insight into the evolutionary significance of Alu exonization events . A subset of Alu-derived exons , especially those derived from more ancient Alu elements in the genome , may have contributed to functional novelties during primate evolution .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"computational",
"biology/alternative",
"splicing",
"genetics",
"and",
"genomics/gene",
"expression",
"evolutionary",
"biology/genomics"
] |
2008
|
Diverse Splicing Patterns of Exonized Alu Elements in Human Tissues
|
Prophylactic interventions such as vaccine allocation are some of the most effective public health policy planning tools . The supply of vaccines , however , is limited and an important challenge is to optimally allocate the vaccines to minimize epidemic impact . This resource allocation question ( which we refer to as VaccIntDesign ) has multiple dimensions: when , where , to whom , etc . Most of the existing literature in this topic deals with the latter ( to whom ) , proposing policies that prioritize individuals by age and disease risk . However , since seasonal influenza spread has a typical spatial trend , and due to the temporal constraints enforced by the availability schedule , the when and where problems become equally , if not more , relevant . In this paper , we study the VaccIntDesign problem in the context of seasonal influenza spread in the United States . We develop a national scale metapopulation model for influenza that integrates both short and long distance human mobility , along with realistic data on vaccine uptake . We also design GreedyAlloc , a greedy algorithm for allocating the vaccine supply at the state level under temporal constraints and show that such a strategy improves over the current baseline of pro-rata allocation , and the improvement is more pronounced for higher vaccine efficacy and moderate flu season intensity . Further , the resulting strategy resembles a ring vaccination applied spatiallyacross the US .
There is an abundance of literature on the modeling , analysis , and control of epidemics . We briefly mention three areas that are closely related to our paper , namely , mobility modeling , disease modeling , and designing interventions to control the spread of epidemics . We refer to [11] [12] for surveys on these topics . We develop a framework for national seasonal/pandemic influenza planning using realistic datasets , a mechanistic model of disease spread , and a greedy optimization algorithm for vaccine allocation . Our specific contributions are discussed below .
Our approach in building the national scale model involves two broad steps: Model calibration is the process of estimating parameters of the computational model that can reproduce observed characteristics in the ground truth . In the context of epidemiology , beyond forecasting , calibrated models allow us to perform counterfactual ( i . e . , what-if ) studies , and address resource allocation questions like VaccIntDesign . In this section , we will briefly describe our approach and the ground truth used for the two-stage calibration of the national-scale influenza model . We begin with the assumption that the ground truth of interest y is a noisy version of the simulation model η ( ⋅ ) at some unknown input parameter configuration θ ^ . We use a gaussian error model , which are simple and adopted widely for many applications , including epidemics [30 , 31] . We adopt importance sampling [32] scheme to produce posterior realizations of the calibration parameters . We begin with sampling from an easy-to-sample importance distribution Im ( θ ) ( say , uniform ) , and run the simulation model η at each of those samples . The importance weights are computed as the ratio of the posterior distribution ( proportional to the product of likelihood and prior distribuion ) and importance distribution evaluated for each of the samples . The samples along with the normalized weights then constitute an estimate of the posterior distribution . Further , it is often useful to factorize the likelihood function , if possible , when the simulation model is required to be calibrated to several different criteria [31] . One possible way is to sequentially calibrate the model to different criteria . In addition to simplifying the computation of importance weights , the approach allows user to introduce more samples as needed using the intermediate calibrated parameter space . More details on the statistical framework and the two-stage posterior exploration is provided in S1 Appendix . Additional methods . Finally , in our case , since we are interested in using a single calibrated model for the optimization study ( as against a weighted ensemble provided by the posterior distribution ) , we consider the Maximum a Posteriori ( MAP ) estimate i . e . , model configuration with highest frequency to be the calibrated model . We now consider the problem of determining the spatial allocation of vaccines across the US to minimize a chosen objective function . In addition to the complexity introduced by non-linear dynamics of the disease model , we also need to account for the temporal constraints imposed by vaccine production and delivery logistics . Formally , the VaccIntDesign problem involves determining the vaccine allocation vector X that minimizes the total attack size given by f ( X ) . This can be expressed as: minimize X f ( X ) subject to ∑ i X i , t ≤ B t , for all t , where Bt is the total number of vaccines available at time t . Our goal in the VaccIntDesign problem is then determine the amount of vaccine allocated to each patch i at time t , denoted by Xi , t . The VaccIntDesign problem is very challenging , and its exact complexity remains open . A strategy that has been useful in many kinds of intervention design problems is to design a greedy allocation , which selects each decision variable based on the marginal improvement to the objective function . If the problem involves submodular maximization , such a strategy is guaranteed to give a constant factor approximation; see , e . g . , [22] [23] . In contrast , VaccIntDesign involves a minimization , and the objective functions are neither submodular or supermodular , in general . Nevertheless , the greedy strategy is a reasonable approach for designing vaccine allocation strategies , and we study it here with an allocation step size of L . The algorithm begins with an initial zero allocation . For each week w , the algorithm allocates the next set of L vaccines to the state s which leads to the maximum reduction in the objective value f ( X ) . The algorithm is repeated for week w , until we exhaust Bw , and then proceed to the next week’s supply of vaccines . Note that the computation of marginal benefit of allocating L additional vaccines to state s subject to population constraints , can be computed in parallel . As a generalization , we have also included the lookahead duration d ( in weeks ) as an additional parameter . This means that the potential allocations at a greedy stage of week w are evaluated by their reduction of attack size at week min ( w + d , T ) where T is the total duration of the epidemic . While this includes the total attack size ( full lookahead , when d ≥ T ) as a special case , it also allows us to explore the resulting trade-off due to varying forecast horizons . The detailed algorithm is provided in S1 Appendix . Additional methods .
For the current study , we begin with the disease model calibrated to the 2014-15 influenza season . Given the best fit model Mθ⋆ , we define the optimization study scenarios as follows: A scenario is defined by a ( v , E ) tuple and is derived by setting the vaccination efficacy to v in model Mθ⋆ and calibrating the transmissibility β to achieve national attack size of E under pro rata vaccine allocation . We do this to simulate multiple seasons that spread spatially like the 2014-15 season , but vary in their severity ( captured by the national attack size E ) and the efficacy of seasonal vaccine ( captured by v ) . In our study setting , we construct 12 scenarios , where v takes values in {0 . 2 , 0 . 35 , 0 . 5} and E takes values in {40 , 61 , 73 , 86} where the values are in millions of cases , corresponding to different severity levels based on past seasons of seasonal influenza . Thus for each target attack size E , we have three scenarios , in which E is achieved by assigning vaccines at v efficacy . In our study ( restricted to contiguous US , including DC ) , the number of states S = 49 . Also , we set the number of weeks W = 40 , roughly the period from September to May corresponding to the influenza season . Therefore , the allocation profile X has 1960 spatio-temporal dimensions . The temporal constraint B is based on historical vaccine uptake schedue available from CDC FluVaxView [36] . CDC FluVaxView provides monthly coverage estimates nationally for the past influenza seasons . We scaled it by the national population to get a vaccine uptake schedule and converted it to the temporal constraint B . Note that CDC also provides the vaccine supply and distribution schedule [37] , however , we noticed a considerable delay between the supply and uptake schedules , so we chose to use the uptake schedule to reflect ground reality .
Current policies for vaccine interventions are designed based on a host of social and political issues , and tend to be fairly simplistic . For instance , Department of Health and Human Services ( HHS ) directives for targeting pandemic vaccines are based on age group [5] , and the allocation of the national vaccine supply and other resources is typically done proportional to the state population . There has been a lot of interest in developing more effective interventions , e . g . , [2] [4] [25] [26] . For instance , Medlock et al . [2] developed an optimal vaccination strategy for the H1N1 outbreak; their model showed a prioritization for a different age group than the ones recommended by CDC directives . All prior methods are restricted to simple models , and only focus on non-temporal interventions in which the allocation is done ahead of time . In reality , vaccine supply varies over time , and the real problem involves finding an allocation that respects the supply constraints and optimizes the epidemic outcomes . Our current model can be extended in several ways . Firstly , the model calibration process can be refined to match more detailed trajectories of influenza spread , like the ILI % time series , or the in-season burden estimates being produced by CDC starting 2018-19 season [33] . Such approaches can then be used to do real-time forecasting and provide vaccine allocation recommendations for an ongoing influenza season . Further , instead of selecting the MAP model for vaccination study , one could use an ensemble of calibrated models based on the posterior distribution , thus being able to quantify uncertainty in the vaccine allocation policy’s effectiveness . Another aspect of the real-world dynamics currently not being captured in our model is that of residual immunity . The national influenza model can be improved by taking into account the co-circulating and dominant influenza strains , as well as the strains present in the recommended vaccine for the season . Note that while improving over pro-rata allocation , greedy algorithm , even with the lookahead duration , may lead to sub-optimal policies . One can develop algorithms that earmark resources for regions with high spreading capacity , thus potentially improving the effectiveness of vaccine allocation . Finally , the logistics of the supply of medical resources , such as medicines , medical equipment ( e . g . , ventilators ) , and medical staff is also very complex . The health infrastructure is generally optimized for typical demand for such resources , and any surge , as would happen during a pandemic outbreak , would place a severe strain on hospitals . Ajao et al . [27] show that over 50 , 000 ventilators might be needed in the event of a national influenza pandemic outbreak . Since local and state health systems are usually unprepared for such a surge in demand , the Office of the Assistance Secretary for Preparedness and Response ( ASPR ) maintains a stockpile of mechanical ventilators in strategic locations [38] , which can be deployed during an emergency . While existing efforts partially address the question of optimizing stockpile redistribution [28] , a mechanistic model like the one developed in this paper will help design better national-scale studies for pandemic preparedness exercises , and develop strategies for allocation of vaccines and other resources during such emergencies . In conclusion , we have presented a national level seasonal influenza model , based on short-range and long-range mobility datasets , and used it to optimize the spatio-temporal allocation of vaccines . For the scenario under consideration , we find that the national attack size can be reduced by up to 17% by allocating the early vaccines to regions around the origin of the epidemic . Most states still end up with close to their overall pro-rata quota of vaccines , however , these findings demonstrate that shifting when and where these vaccines are administered has a sizable impact on the national attack size . Achieving these optimal outcomes would require better surveillance and the ability to accelerate vaccine uptake at will , which presents multiple challenges . However , the study shows there is ample room for improvement and this framework provides means for developing a play-book for epidemic containment .
|
Annual vaccination campaigns continue to be one of the prime measures which help alleviate the burden of seasonal influenza . Due to production and logistic constraints , there is a need for prioritization policies associated with vaccine deployment . While there is general consensus on age-based or risk-based prioritization , spatial optimization of vaccine allocation has not yet been explored in sufficient detail . In order to do this , we develop a mechanistic model of influenza spread across the United States , and propose a greedy mechanism for spatial optimization . We test the methodology on different realistic scenarios with temporal constraints on vaccine production .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"influenza",
"applied",
"mathematics",
"immunology",
"social",
"sciences",
"human",
"mobility",
"simulation",
"and",
"modeling",
"vaccines",
"preventive",
"medicine",
"algorithms",
"mathematics",
"infectious",
"disease",
"control",
"vaccination",
"and",
"immunization",
"human",
"geography",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"geography",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"vaccine",
"development",
"viral",
"diseases",
"physical",
"sciences"
] |
2019
|
Optimizing spatial allocation of seasonal influenza vaccine under temporal constraints
|
Despite recent advances in our ability to detect adaptive evolution involving the cis-regulation of gene expression , our knowledge of the molecular mechanisms underlying these adaptations has lagged far behind . Across all model organisms , the causal mutations have been discovered for only a handful of gene expression adaptations , and even for these , mechanistic details ( e . g . the trans-regulatory factors involved ) have not been determined . We previously reported a polygenic gene expression adaptation involving down-regulation of the ergosterol biosynthesis pathway in the budding yeast Saccharomyces cerevisiae . Here we investigate the molecular mechanism of a cis-acting mutation affecting a member of this pathway , ERG28 . We show that the causal mutation is a two-base deletion in the promoter of ERG28 that strongly reduces the binding of two transcription factors , Sok2 and Mot3 , thus abolishing their regulation of ERG28 . This down-regulation increases resistance to a widely used antifungal drug targeting ergosterol , similar to mutations disrupting this pathway in clinical yeast isolates . The identification of the causal genetic variant revealed that the selection likely occurred after the deletion was already present at high frequency in the population , rather than when it was a new mutation . These results provide a detailed view of the molecular mechanism of a cis-regulatory adaptation , and underscore the importance of this view to our understanding of evolution at the molecular level .
Evolutionary adaptation is the process that has given rise to the ubiquitous , yet remarkable , fit between all living organisms and their environments [1] . The origins of these adaptations at the molecular level have been a subject of great interest , with active debate surrounding the relative roles of two major classes of molecular mechanism: changes in protein sequences vs . changes in the expression levels/patterns of those proteins [2]–[5] . Until recently , the evidence cited in favor of both mechanisms was either anecdotal ( involving studies of single genes ) or theoretical in nature [2]–[4] . However , the advent of methods for characterizing gene expression adaptation genome-wide [6]–[9] ( as well as methods for measuring cis-regulatory changes that may or may not be adaptive [10]–[11] ) has paved the way for this question to be addressed in an unbiased , systematic fashion [5] . Although the distinction between protein sequence vs . gene expression regulation is important , it is only one of many levels at which molecular mechanisms can be distinguished . For example among cis-regulatory adaptations , mutations might act via alterations in transcription factor ( TF ) binding , nucleosome positioning , mRNA processing , binding of RNA-binding proteins , etc . As the field matures , it is likely that the distinctions between these more detailed mechanistic levels will be of increasingly greater interest , since only by investigating these mechanisms will we fully understand the nature of adaptation at the molecular level . In order to investigate the molecular mechanism of an adaptation , it is generally necessary to first identify the causal mutation ( s ) ( though see [12] ) . This prerequisite has been a significant bottleneck in studies of cis-regulatory adaptation . Because we cannot computationally predict the effects of most non-coding mutations , and such mutations can act at long distances from their target genes in many species ( resulting in a large search space ) , only a handful of causal mutations underlying cis-regulatory adaptations have been reported . For example , large deletions of an enhancer driving the pelvic expression of the Pitx1 gene in sticklebacks have been found to result in adaptive pelvic reduction in freshwater populations [13] . In another case , five non-coding mutations at the ebony locus contributed to dark abdominal pigmentation found in high-altitude populations of Drosophila melanogaster [14] ( although other examples exist where causal cis-regulatory mutations have been identified [15]–[17] , these have not been shown to be adaptive ) . However even for these intensively studied cis-regulatory adaptations , and others where important factors such as fitness effects have been estimated [18] , the molecular mechanisms by which the causal mutations act—e . g . which TFs and/or epigenetic states are affected by the mutations—remain unknown . We previously reported a genome-wide scan for gene expression adaptation between two strains of the budding yeast Saccharomyces cerevisiae: a laboratory strain ( BY4716 , hereafter “BY” ) and a vineyard strain ( RM11-1a , hereafter “RM” ) [8] . We found that over 200 genes had likely been subject to recent positive selection in these strains via reinforcing cis and trans-acting regulatory adaptations . Among these genes , there was a particularly strong enrichment of down-regulating mutations in one metabolic pathway: ergosterol biosynthesis . Ergosterol is an abundant lipid component of the fungal plasma membrane , and is of major biomedical importance , being targeted by numerous antifungal drugs [19] . Indeed , a common mechanism of resistance to ergosterol-targeting drugs ( such as amphotericin B ) is reducing ergosterol levels via disruption of this pathway [19]–[21] . We previously found that six genes within the pathway ( underlined and red in Figure 1A ) showed the strongest signs of selection , based on patterns of reinforcing cis/trans-regulatory mutations , as well as a population-genetic signature of selective sweeps in the genomes of multiple strains [8] . This represents the first known example of a polygenic gene expression adaptation , from any species . Here , we sought to gain a deeper understanding of this adaptation .
Because our initial identification of the polygenic gene expression adaptation within the ergosterol ( ERG ) biosynthesis pathway was based on expression data from genome-wide microarrays [22] , we first sought to more precisely measure the cis-regulatory divergence at these loci . This divergence can be measured for any gene as the ratio of mRNA abundances of the two alleles present in a hybrid diploid: in the absence of cis-acting differences , the mRNA from the two alleles will be present in equal amounts ( as they are in the genomic DNA ) , whereas they will be unequal in the presence of cis-regulatory divergence . To measure this ratio we employed pyrosequencing , a method that accurately quantifies allelic ratios at individual heterozygous sites [23] . Of the six genes we previously implicated , five were amenable to this approach ( the sixth , ERG26 , lacked any BY/RM sequence differences in its mRNA , so the alleles could not be distinguished ) . All five showed reproducible allelic imbalance in the expected direction ( lower expression from the BY allele ) , with magnitudes ranging from 1 . 13-fold to 1 . 94-fold ( Figure 1B ) . This result confirms that the “local eQTL” ( genetic markers showing a statistical association with a nearby gene's expression level ) previously mapped for these genes [22] likely represent cis-acting genetic variants . To investigate if the polygenic adaptation extends beyond the six genes we originally identified , we also performed pyrosequencing on three additional ERG genes adjacent in the pathway to those already implicated: ERG25 , ERG27 , and ERG2 ( allelic bias of ERG1 , the other adjacent pathway member , could not be measured because it has no sequence differences between BY and RM ) . We found reproducible allelic bias in favor of RM for both ERG25 and ERG27 , but not for ERG2 ( Figure 1B ) . This suggests that the adaptive down-regulation extends to a total of at least eight genes , forming a contiguous block within the ERG pathway ( Figure 1A , in red ) that has been specifically targeted by natural selection . Interestingly , in addition to the clear clustering of the down-regulated genes within the pathway , the genes with the strongest cis-regulatory differences correspond precisely to the core proteins in a stable complex organized by Erg28 . Erg28 is the only known member of the ERG pathway lacking enzymatic activity; it is an endoplasmic reticulum transmembrane protein , highly conserved across eukaryotes ( including humans ) , that acts as a scaffold promoting co-localization of ERG enzymes [24]–[26] . Erg28 physically interacts most strongly with Erg27 ( and is thus shown next to Erg27 in Figure 1 ) , but has also been found to interact strongly with itself and three other proteins: Erg25 , Erg6 , and Erg11; its other interactions are significantly weaker [24] . These five interacting proteins are not only all components of the polygenic adaptation ( Figure 1 ) , but are specifically those components with the strongest cis-acting down-regulation: all five have at least 1 . 25-fold differences between RM and BY alleles , while no other genes quite reach this threshold ( Figure 1B ) . This pattern suggests that the precise magnitude of down-regulation may be influenced both by pathway position and by membership in the protein complex organized by Erg28 . We decided to focus on ERG28 for further investigation . Not only is Erg28 the central member of the protein complex apparently targeted by natural selection , but sequence divergence in its promoter region was also minimal: there are only two sequence differences between BY and RM in the 590 bp upstream of the ERG28 transcription start site ( TSS ) . These are one two-bp deletion ( located in an 11 bp poly-A tract 112 bp upstream of the TSS , termed the AA112Δ allele ) , and one T/C SNP ( 229 bp upstream of the TSS , the T229C allele ) ( Figure 2A ) . Because promoters in S . cerevisiae are compact ( generally <400 bp [27] ) , we decided to focus on these two candidate variants . To definitively identify the mutation ( s ) underlying a cis-regulatory adaptation , the mutations must be individually tested for their effects on expression of the associated gene . Therefore we constructed allelic replacement strains in which individual BY variants were introduced into the RM genome . Using a method of in vivo site-directed mutagenesis known as delitto perfetto [28] , we engineered strains that differed only by the desired mutation . We refer to the two resulting strains as RM AA112Δ and RM T229C ( Figure 2b ) . If a mutation can fully account for the 1 . 30-fold cis-acting difference between the RM/BY alleles of ERG28 ( Figure 1B ) , and no additional mutations have any effect , then this mutation can be deemed causal . To test if this was the case for either of our candidate mutations , we measured the expression level of ERG28 in each strain , as well as in wild-type RM , by quantitative PCR ( qPCR ) . While we found no effect of the T229C mutation ( 1 . 05-fold difference ) , we observed that the AA112Δ mutation led to a 1 . 26-fold decrease in mRNA level ( Figure 2c ) , indistinguishable from the 1 . 30-fold change expected for the causal mutation ( s ) . To further test if the AA112Δ mutation could fully account for the RM/BY difference , we mated the RM AA112Δ strain with BY , and measured the allelic ratio of ERG28 mRNA in the resulting diploid strain . The causal mutation would be expected to reduce the 1 . 30-fold allelic difference to ∼1 , while any non-causal mutation would have the same the 1 . 30-fold allelic imbalance found in the BY/RM hybrid . Consistent with the qPCR results , the RM AA112Δ/BY hybrid strain showed a 1 . 03-fold difference between alleles , while the RM T229C/BY hybrid showed a 1 . 27-fold difference ( Figure 2d ) . Together , these results suggested that the AA112Δ mutation likely accounted for all , or nearly all , of the cis-acting divergence at ERG28 between RM and BY . We considered two potential mechanisms for how the AA112Δ mutation may be down-regulating transcription: nucleosome positioning and TF binding . Both processes are known to play important roles in determining rates of transcription initiation , and could potentially be affected by a 2-bp deletion . Nucleosome positioning was an especially plausible mechanism because the 11-bp poly-A sequence in which the 2-bp deletion occurred is a strong nucleosome-disfavoring sequence [29] . Therefore we took advantage of published data on genome-wide nucleosome positions from BY and RM [30] to determine whether the nucleosome overlapping the deletion was affected . There was no significant difference between BY and RM in the nucleosomal occupancy or positioning at this location ( nor was it differentially acetylated on histone H3 lysine 14 [30] ) , suggesting that nucleosome occupancy was not greatly affected by this deletion . We therefore turned to TF binding as a second possible mechanism . Utilizing a published map of putative TF binding sites [31] we identified two highly conserved ( across Saccharomyces sensu stricto ) binding sites for the TFs Mot3 and Sok2 , flanking the deletion ( Figure 3a ) . Mot3 is a well-known repressor of ERG pathway genes , exerting its greatest effect in hypoxic or hyper-osmotic conditions [32]–[33] , whereas Sok2 has not been previously linked to the ERG pathway to our knowledge . Neither binding site motif is directly affected by the 2-bp deletion; rather the only effect is on their spacing , reducing the distance between motif centers from 16 bp to 14 . To test if the AA112Δ deletion may affect the regulation of ERG28 by either of these two TFs , we created knockout strains for each TF in both the wild-type RM and RM AA112Δ backgrounds . Several outcomes are possible ( Figure 3b ) . First , if the TF does not regulate ERG28 , then deleting it should have no effect in either genetic background . Second , if the TF does regulate ERG28 but is not affected by the AA112Δ deletion , then the effect of TF deletion should be equal in the two backgrounds . Finally , if the AA112Δ deletion is affecting the TF's regulation of ERG28 , then the effect of TF deletion will depend on the background—for example , having an effect on ERG28 expression in wild-type RM but not in RM AA112Δ . Consistent with Mot3's known role as a repressor of ERG pathway genes , we found that ERG28 was induced 1 . 85-fold in an RM mot3Δ strain compared with wild-type RM ( Figure 3c ) . Likewise , Sok2 was found to be an activator of ERG28 , with 1 . 21-fold lower expression in RM sok2Δ compared to wild-type RM . However neither TF had any measurable effect on ERG28 expression when deleted from the RM AA112Δ strain ( Figure 3c ) . This suggests that although both TFs regulate ERG28 in RM , this regulation was abolished by the 2-bp deletion . The effect of AA112Δ on regulation of ERG28 by Mot3 and Sok2 suggested that their binding to the promoter may be affected by the deletion . To investigate this , we performed chromatin immunoprecipitation ( ChIP ) . Specifically , we HA-tagged both TFs in both wild-type RM and RM AA112Δ backgrounds , and quantified their binding to specific regions by quantitative PCR ( qPCR ) . We found that for both factors , binding at the ERG28 promoter was reduced in RM AA112Δ , compared to wild-type RM: Sok2 showed ∼19-fold lower binding , while Mot3 had ∼31-fold lower binding ( Figure 3d ) . This suggests that the loss of ERG28 regulation by these TFs in the AA112Δ background ( Figure 3c ) is likely due to their severely reduced binding . In order to investigate the phenotypic effects of the AA112Δ allele , we measured the growth rates of our engineered strains and RM in several environments ( see Materials and Methods ) . While we did not observe any fitness advantage of the RM AA112Δ strain in most conditions ( e . g . rich synthetic defined [SD] media; paired t-test p = 0 . 46 for RM AA112Δ vs . RM and p = 0 . 83 for RM T229C vs . RM; Figure 4a ) , we did find a growth advantage of this strain in the presence of the antifungal drug amphotericin B ( Figure 4b ) . Specifically , RM AA112Δ had a 1 . 3% higher growth rate than RM when grown in the presence of the drug ( p = 0 . 014 ) , whereas RM T229C had no measurable difference from RM ( p = 0 . 86 ) . This suggests that the fitness benefit conferred by the AA112Δ allele is condition-specific . Our identification of the AA112Δ allele as causal allows us to examine the distribution of this adaptive mutation across other yeast strains , in order to study its history . In particular , we wished to address the question of whether the selection occurred when the deletion was a new mutation that just recently arose ( e . g . in the laboratory ) , or whether it was present as “standing variation” in S . cerevisiae for some time before the selection occurred . Population geneticists have theorized about the consequences of selection acting on pre-existing variation , as opposed to waiting for rare advantageous mutations to occur , but few clear examples exist [34]–[36] . To distinguish between these alternatives , we first examined the distribution of the AA112Δ allele across a set of 36 sequenced strains of S . cerevisiae [37] . The deletion is present in 12/36 sequenced strains ( in addition to BY; Figure S1 ) . These 12 strains are diverse in terms of both geography ( from the Americas , Asia , Africa , and Europe ) and lifestyle ( lab strains , wild strains , sake strains , palm wine strains , and other fermentation strains ) . Furthermore they are genetically diverse , as evidenced by their lack of clustering within the S . cerevisiae phylogeny ( Figure S1 ) . This broad distribution across the species suggests that the AA112Δ allele is present at appreciable frequency in many populations of S . cerevisiae . To further investigate this , we sequenced the ERG28 promoter in EM93 , the wild strain that accounts for ∼88% of the BY genome [38]–[39] . Since EM93 is a diploid , we sequenced the promoter in the four spores from a single EM93 tetrad , in order to capture both alleles with no ambiguity . We found that the AA112Δ mutation was heterozygous within EM93 , supporting our inference that it is commonly found in the wild . Together , these results suggest that the selection on ERG28 in the BY lineage [8] was likely acting on standing variation , as opposed to a new mutation . Because EM93 is heterozygous , we can infer the selective sweep most likely occurred in the descendants of EM93 , after its introduction to the laboratory . To attempt a similar analysis for the seven other ERG genes involved in this adaptation ( Figure 1A ) , we sequenced their promoters in the same four EM93 spores . Because we do not know the causal variants , we performed this analysis at the level of promoter haplotypes ( sets of co-occurring alleles ) . We found that for all seven genes , the complete BY promoter haplotype was either homozygous ( for two genes , ERG25 and ERG26 ) or heterozygous ( for five genes ) in EM93 , indicating that their cis-acting down-regulations were likely not due to new mutations occurring in the lab . Each of these BY haplotypes was also observed in between zero and six additional sequenced strains , indicating that some of the haplotypes are segregating at an appreciable frequency in S . cerevisiae . However the absence of a complete BY haplotype does not imply the absence of the causal BY variant , since most ERG promoter variants are not in perfect linkage disequilibrium with their neighboring variants . For example , although the AA112Δ variant was found in 12 strains ( Figure S1 ) , only five of these also had the T229C variant ( and thus the complete BY promoter haplotype ) . This highlights the importance of identifying causal variants in order to study the evolutionary histories of specific adaptations .
We have identified the causal mutation underlying a cis-regulatory adaptation that affects the ergosterol biosynthesis pathway in yeast , and characterized its molecular mechanism of action . The mutation , a 2 bp promoter deletion , reduces the expression of ERG28 by ∼1 . 3-fold . This effect is mediated by two TFs , Mot3 and Sok2 , which bind immediately adjacent to the deletion; these TFs bind and regulate the wild-type RM ERG28 promoter , but not the ERG28 AA112Δ promoter . Although it may seem surprising that a 2 bp deletion outside of TF binding sites can have such a strong effect on TF binding , it is consistent with previous work . First , most between-strain variation in the binding of the Ste12 TF in yeast cannot be linked to variation in any known TF motif , even when only considering those binding sites where occupancy was associated with nearby genetic markers [40] . Second , it was recently shown that changes in the positions of TF binding sites as small as 1–2 bp can result in substantial ( >1 . 5-fold ) effects on transcription [41] . Finally , minor changes in the copy number of very short tandem repeats in yeast promoters can also impact transcription [42] . It is also at first counterintuitive that decreased binding of a repressor ( Mot3 ) could contribute to the down-regulation of ERG28 by AA112Δ , in particular since the repressive effects of Mot3 appear to be stronger than the activation by Sok2 ( Figure 3C ) . We hypothesize that the AA112Δ mutation may have altered the TF binding landscape upstream of ERG28 , not only for Mot3 and Sok2 , but possibly for other TFs or their cofactors as well . The deletion's effect on transcription would then be determined by this altered landscape . In addition to the focus on ERG28 , our results also further characterize the polygenic ERG pathway adaptation as a whole . We found that two genes not implicated in our previous analysis of microarray data [8] , ERG25 and ERG27 , also show reduced expression from the BY allele ( compared to RM ) . Moreover , our precise measurements of the cis-acting effect size for each ERG gene led us to an intriguing discovery: the five proteins that form the core of a complex at the ER membrane are also the five with the strongest cis-regulatory change . This pattern suggests an exquisite specificity of selection , in which the precise level of down-regulation is determined not only by position within the pathway , but also by membership in a particular protein complex . While a handful of causal mutations underlying cis-regulatory adaptations in other model organisms have been previously reported [13]–[14] , their molecular mechanisms are unknown . Compared to these , our knowledge of the ERG28 AA112Δ mutation is now relatively detailed , though still incomplete; for example , how the deletion disrupts binding has not been established . A plausible explanation is that Sok2 and Mot3 may bind cooperatively to the ERG28 promoter in wildtype RM; if this cooperativity is disrupted by the 2-bp deletion ( which brings the binding sites ∼6 . 8 Å closer together and changes their relative angles by ∼70° ) , then neither factor would bind well to the AA112Δ promoter . At the phenotypic level , we found that AA112Δ confers a condition-specific growth advantage in the presence of the antifungal drug amphotericin B . Because the AA112Δ mutation may also lead to a fitness advantage in other environments that were not tested , we cannot conclude whether amphotericin B is related to the specific selection pressure that gave rise to the ERG pathway adaptation in BY . However our results are quite consistent with previous observations that the down-regulation or inactivation of ERG pathway genes confers resistance to amphotericin B in diverse clinical yeast isolates [19]–[21] . Thus in addition to aiding our understanding of the molecular mechanisms of cis-regulatory adaptation , our results may shed light on potential mechanisms by which antifungal drug resistance can evolve .
We carried out all strain engineering in RM , as opposed to BY , because BY contains a very recent loss-of-function transposon insertion in the transcription factor HAP1 , which alters the regulation of many ERG genes , including ERG28 . Because this mutation was so recent ( not even present in the very closely related lab strain W303 [8] ) , it must have happened after the ERG28 cis-regulatory adaptation , so the functional HAP1 in RM should more accurately reflect the original effects of any cis-regulatory mutations . In vivo site-directed mutagenesis , known as delitto perfetto , was performed as described [27] . Briefly , the pCORE-UH cassette , containing K . lactis URA3 and hyg , was amplified using primers containing ∼70 bp of homology to the RM ERG28 promoter ( Table S1 ) . This PCR product was transformed into RM , and correct incorporation into the ERG28 promoter was verified by PCR . The site of incorporation was chosen in between the two candidate genetic variants , so that the same CORE cassette transformant could be used for engineering both mutations . The CORE cassette was then removed by separately transforming two PCR products from the BY ERG28 promoter , containing the desired mutation ( either AA112Δ or T229C ) as well as enough flanking DNA sequence ( identical between RM and BY ) to allow specific targeting of the PCR product . Because the efficiency of delitto perfetto is maximized when transforming longer DNA molecules , as well as double-stranded DNA [20] , transforming long PCR products from BY ( as opposed to shorter , single-stranded synthetic oligonucleotides ) is a useful modification . Counter-selection of the resulting transformants on 5-FOA allowed isolation of successfully engineered strains that had replaced the CORE cassette with the desired mutation , which were then sequence-verified . The complete coding regions of MOT3 and SOK2 were replaced with the hphMX6 antibiotic resistance gene via PCR-mediated gene disruption [43] in both RM and RM AA112Δ . Transformants were grown on hygromycin B , and verified by PCR . These two TFs were also HA-tagged at their C-termini via transformation of a PCR product including the HA tag , hphMX6 , and flanking regions with 40 bp of homology to the targeted regions [43] . Transformants were grown on hygromycin B , and then verified by PCR and sequencing . Table S2 lists all strains used in this work . With the exception of growth rate experiments ( Figure 4 ) , all strains were grown in standard YPD media at 30°C , and harvested in log-phase ( OD600 ∼1 ) for either RNA extraction or chromatin immunoprecipitation . We extracted total RNA with the Epicentre Biotechnologies RNA Purification kit , which includes a DNase treatment to remove contaminating genomic DNA . RNA concentration was quantified with a NanoDrop2000 spectrophotometer . For cDNA synthesis , total RNA samples were diluted to a concentration of 500 ng/µL . RNA was reverse transcribed into cDNA with SuperScript III RT ( Invitrogen ) , following manufacturer protocols . Pyrosequencing was performed on a PyroMark Q24 ( Qiagen ) , following manufacturer's protocols . Primers ( Table S1 ) were designed to target individual SNPs in transcribed regions using the PyroMark Assay Design Software ( Qiagen ) . Negative controls using no primers , or no cDNA template , were performed for each assay . cDNA was diluted 1∶100 prior to qPCR . qPCR was performed on an Eco Real-Time PCR machine ( Illumina ) following manufacturer's protocols . To quantify changes in ERG28 mRNA abundance , six control genes previously noted for their stability across conditions [44] were measured in each experiment: ACT1 , TDH3 , ALG9 , TAF10 , TFC1 , and UBC6 . All experiments were done in at least biological triplicate and technical duplicate . Experiments in Figure 3c were done in biological sextuplicate and technical quadruplicate . Data were analyzed using qBase Plus software ( Biogazelle ) [45] . Chromatin immunoprecipitation ( ChIP ) was performed essentially as described [46] . Briefly , wildtype cells and cells expressing either Mot3-HA or Sok2-HA were grown to mid-log phase in 100 mL YPD . Cross-linking was performed by treating yeast with 1% formaldehyde for 15 minutes at 25°C . Chromatin was isolated from whole-cell extracts generated by spheroplasting and sheared by sonication . Immunoprecipitation was performed from 5 µg chromatin using mouse monoclonal anti-HA ( Invitrogen , clone 5B1D10 ) and immune complexes were captured with Ultralink Immobilized Protein A/G resin ( Pierce ) . Protein-DNA complexes were eluted with 1% SDS/0 . 1 M NaHCO3 . Eluates were incubated at 65°C overnight to reverse cross-links and treated with proteinase K ( Invitrogen ) and RNAse A . DNA was phenol-chloroform extracted , ethanol-precipitated , and resuspended in water prior to qPCR . ChIP DNA was amplified on an Eco Real-Time PCR machine ( Illumina ) following manufacturer's protocols . We quantified the abundance of the ERG28 promoter region containing the Mot3 and Sok2 binding sites , as well as part of the ACT1 coding region as a control to quantify the amount of DNA in each reaction . The concentration of ERG28 promoter DNA was normalized against this control before comparing across genetic backgrounds ( RM vs . RM AA112Δ ) for each TF . To perform quantitative growth rate measurements ( Figure 4 ) , we grew strains in 96-well plates and measured OD600 at 15-minute intervals using an automated plate reader ( Tecan ) until cultures reached saturation . Data shown in Figure 4 are the mean log2 ratios of the maximum log-phase growth rates ( estimated by Magellan software , Tecan ) for 48 replicate growth curves of each strain . Growth conditions were SD media alone or 0 . 8 ug/ml amphotericin B in SD media , both at room temperature ( 22°C ) . P-values were calculated using a paired t-test , pairing wells in the same row on each plate . Other conditions tested in an initial screening phase were hyperosmotic stress ( NaCl or menadione ) and temperature stress ( heat/freezing ) .
|
Evolutionary adaptation is the process that has given rise to the ubiquitous , yet remarkable , fit between all living organisms and their environments . The molecular mechanisms of these adaptations have been a subject of great interest , but we still know very little about their mechanisms , particularly in the case of regulatory adaptations . In this work , we investigate the molecular mechanism of a regulatory adaptation that we previously identified in ERG28 , a component of the ergosterol biosynthesis pathway in budding yeast . Ergosterol is an abundant lipid component of the fungal plasma membrane , and is of major biomedical importance , being targeted by numerous antifungal drugs . We identified the causal mutation underlying the ERG28 adaptation , a two-base deletion in its promoter which leads to lower abundance of its mRNA . This deletion acts via disrupting the binding of at least two transcription factors , Mot3 and Sok2 , to the promoter . The deletion increases resistance to a widely used antifungal drug , Amphotericin B , which targets ergosterol . This effect is reminiscent of misregulation of the ergosterol pathway observed in clinical yeast isolates that have evolved resistance to Amphotericin B . Our results may therefore have medical implications , while also advancing our basic understanding of evolutionary mechanisms .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
The Molecular Mechanism of a Cis-Regulatory Adaptation in Yeast
|
The interferon ( IFN ) system represents the first line of defense against a wide range of viruses . Virus infection rapidly triggers the transcriptional induction of IFN-β and IFN Stimulated Genes ( ISGs ) , whose protein products act as viral restriction factors by interfering with specific stages of virus life cycle , such as entry , transcription , translation , genome replication , assembly and egress . Here , we report a new mode of action of an ISG , IFN-induced TDRD7 ( tudor domain containing 7 ) inhibited paramyxovirus replication by inhibiting autophagy . TDRD7 was identified as an antiviral gene by a high throughput screen of an ISG shRNA library for blocking IFN’s protective effect against Sendai virus ( SeV ) replication . The antiviral activity of TDRD7 against SeV , human parainfluenza virus 3 and respiratory syncytial virus was confirmed by its genetic ablation or ectopic expression in several types of mouse and human cells . TDRD7’s antiviral action was mediated by its ability to inhibit autophagy , a cellular catabolic process which was robustly induced by SeV infection and required for its replication . Mechanistic investigation revealed that TDRD7 interfered with the activation of AMP-dependent kinase ( AMPK ) , an enzyme required for initiating autophagy . AMPK activity was required for efficient replication of several paramyxoviruses , as demonstrated by its genetic ablation or inhibition of its activity by TDRD7 or chemical inhibitors . Therefore , our study has identified a new antiviral ISG with a new mode of action .
Interferon ( IFN ) system provides the first line of immune defense against viral infections in vertebrates [1–3] . It is designed to inhibit viral infection by blocking virus replication and eliminating the virus-infected cells . The Pattern Recognition Receptors ( PRRs ) , e . g . Toll Like Receptors ( TLRs ) , RIG-I Like Receptors ( RLRs ) and cyclic AMP-GMP synthase ( cGAS ) /stimulator of IFN genes ( STING ) , are located in distinct cellular compartments , to sense specific viral components , such as the viral nucleic acids [4–9] . Upon ligand stimulation , the PRRs trigger rapid downstream signaling pathways via respective adaptor proteins to activate the transcription factors , e . g . Interferon Regulatory Factors ( IRFs ) and Nuclear Factor-κB ( NF-κB ) . The co-operative action of these transcription factors triggers the synthesis of Type-I interferons e . g . IFN-β , an extensively studied antiviral cytokine . After synthesis in the infected cells , IFN-β is secreted and acts on the infected as well as yet uninfected cells via Janus Kinase ( JAK ) /Signal Transducer of Transcription ( STAT ) signaling pathways to trigger the synthesis of a number of antiviral genes . All biological effects of IFN are executed by the induced proteins , encoded by Interferon Stimulated Genes ( ISGs ) , which are either not present or expressed at a low level in untreated cells , but can be transcriptionally upregulated by IFN-action [3 , 10 , 11] . Most ISGs can also be induced directly in the virus-infected cells without IFN-action [12] . The ISGs perform all physiological and pathological , including viral and non-viral , functions of IFNs . The ISGs function singly or in combination with other ISGs to inhibit virus replication . The antiviral activities of only a handful of these ISGs have so far been identified . Among them , Protein Kinase R ( PKR ) , 2’5’ Oligoadenylate Synthetase ( OAS ) , Mx1 , IFN-induced protein with tetratricopeptide repeats ( IFIT ) , tripartite motif ( TRIM ) family are most well-known for their antiviral activities against a wide spectrum of viruses in vitro and in vivo [13–20] . PKR , upon binding to viral double-stranded RNA ( dsRNA ) , is activated and phosphorylates eukaryotic initiation factor ( eIF2α ) , leading to the translational inhibition of cellular and viral mRNAs [21] . Mx1 is a broad antiviral ISG that acts at an early stage of virus replication , by sequestering the viral components from the desired destination within the cells [18] . OAS recognizes dsRNA and produces 2’ , 5’-oligoadenylates , which activate the latent ribonuclease , RNase L that degrades both cellular and viral RNAs [14] . The IFIT family of ISGs recognizes viral mRNAs and thereby inhibiting their translation [17 , 19] . IFIT proteins also directly modulate cellular translation machinery by inhibiting eIF3 activities [22] . The TRIM family of proteins , which possesses E3 ubiquitin ligase activity , has diverse cellular functions [20] . In addition to directly interfering with virus life cycle , the ISGs often exert their antiviral actions by amplifying the cellular IFN responses [10] . Many of the ISGs also serve as PRRs or signaling intermediates , which are expressed at low levels and are transcriptionally induced by IFN signaling . IFN can inhibit many stages of virus replication: viral entry , transcription , replication , translation , assembly or egress . IFN induced transmembrane proteins ( IFITM ) mediate antiviral resistance to a wide range of viruses [23] . IFITM proteins target the attachment and uncoating , two very early stages of viral entry , of several enveloped viruses [24] . Viperin inhibits Hepatitis C Virus ( HCV ) replication by localizing to the cellular lipid droplets , the site of viral replication [25] . Viperin also inhibits the budding and release of Influenza A virus by disrupting lipid rafts [26] . Tetherin ( BST2 ) prevents the release of human immunodeficiency virus ( HIV-1 ) by tethering HIV-1 virion particles to the cell surface [27] . In addition to targeting individual steps , multiple ISGs may target different steps of virus replication and elicit a large cumulative antiviral effect [3] . A single ISG can also function in cell type-specific manner to exhibit antiviral defense against multiple viruses . Ifit2 provides protection against a wide range of viruses in specific cells and tissues [17] . Viruses take advantage of cellular machineries or their components to achieve productive replication in the infected cells . Autophagy is an evolutionary conserved cellular degradation pathway , which has numerous physiological functions , including maintenance of cellular homeostasis and host defense [28–30] . Autophagy is induced by various cellular stresses , such as nutrient deprivation or microbial infection . Autophagy generates double-membranous cytoplasmic structure , known as autophagosome , which fuses with the lysosome , leading to the degradation of undesired cellular contents [31] . Adenosine monophosphate ( AMP ) -dependent Kinase ( AMPK ) directly phosphorylates the critical Ser/Thr residues of Unc-51 like autophagy activating kinase 1 ( ULK1 ) to initiate the autophagy pathway [32] . Autophagy is regulated by class III PI3 Kinase ( PI3K III ) and mammalian target of rapamycin ( mTOR ) [33 , 34] . How autophagy regulates virus replication is not clear; however , many viruses utilize autophagy or its components to promote their replication . Autophagy induced by virus infection can have both pro- or anti-viral effects [35–37] . Hepatitis C Virus ( HCV ) , Dengue virus and Poliovirus use autophagosome membranes for their replication [38–40] . Human parainfluenza virus 3 ( HPIV3 ) triggers autophagy in the infected cells and the viral P protein blocks the degradation of autophagosome , to enhance the intracellular virus yield [41] . Measles virus sequesters RIG-I within autophagosome , to evade the antiviral action of IFN [42] . In dendritic cells , Respiratory Syncytial Virus ( RSV ) induces autophagy , which regulates the adaptive immune responses [43] . How the IFN system regulates ‘virus-induced autophagy’ is unclear . Paramyxoviruses are strong inducers of IFN and ISGs in the infected cells . To identify the ISGs that can inhibit their replication , we performed a high throughput genetic screen of individual ISGs for their ability to inhibit the replication of Sendai virus ( SeV ) ( family: Paramyxoviridae , sub-family: Paramyxovirinae , genus: Respirovirus ) . Our screen identified a small subset of antiviral ISGs , including Tudor domain containing 7 ( TDRD7 ) , which strongly inhibited the replication of SeV in human and mouse cells . The TDRD family of proteins contains multiple Tudor domains and have roles in cellular RNA metabolism . Tdrd7-/- mice show defects in lens development and spermatogenesis , which are related to the deficiency in Tdrd7-associated mRNAs [44 , 45] . Our results demonstrate that TDRD7 inhibits the replication of not only SeV , but also other paramyxoviruses , in multiple cell types . In-depth mechanistic studies revealed that the antiviral effect of TDRD7 is mediated by its ability to inhibit ‘virus-induced autophagy’ , which is required for paramyxovirus replication .
To identify the ISGs that block SeV replication , we set up an unbiased genetic screen using a shRNA library against human ISGs ( GIPZ lentiviral shRNAmir , Open Biosystems ) in HeLa cells . The library is a 96-well formatted commercial system , which packages lentiviruses encoding individual ISG shRNA , a GFP reporter , and a puromycin selection cassette [46] . The GFP expression allowed a flow cytometry-based screen , in which lentivirus-transduced cells were tracked by GFP . The SeV-infected cells were stained with an antibody against the whole virion . In each well of cells , all ISGs were induced by IFN-β-pretreatment but the expression of only one of them was prevented by transduction of the cognate shRNA; the cells were then infected with SeV and the degree of virus replication was measured . Reversal of IFN-mediated inhibition of virus replication in a specific well indicated that the corresponding ISG was responsible for inhibiting SeV replication . We optimized the assay by knocking down the expression of genes required for IFN signaling and , therefore , would exhibit a phenotype . In HeLa cells , the knockdown of IRF9 enhanced the expression of viral protein ( SeV C ) in comparison to the NT control ( Fig 1A , lanes 1 , 3 , 5 ) . Moreover , IRF9 knockdown cells reversed the IFN-β-mediated inhibition of SeV C protein expression ( Fig 1A , lanes 2 , 4 , 6 ) . We used these cells to develop a flow cytometry-based quantitative assay to measure SeV replication as percent of SeV-positive cells within the GFP-expressing cell population ( Fig 1B ) . The results indicate that knockdown of IRF9 reversed the IFN-β-mediated inhibition of SeV replication ( Fig 1B ) . We quantified the GFP-expressing SeV-antigen positive cells ( % infectivity ) by flow cytometry ( Fig 1C ) . IRF9 knockdown led to enhanced SeV replication ( % infectivity ) in both untreated and IFN-β-treated cells ( Fig 1D ) . We screened the shRNA library , which consists of 814 lentiviral constructs ( multiple targets for each ISG ) expressing shRNA against more than 300 human ISGs . The shRNA library was lentivirally expressed in HeLa cells , using the strategy outlined in Fig 1E . A shRNA against IRF9 was used as an internal control to validate the effect of IFN-β ( as in Fig 1D ) for each screening experiment . Percent infectivity was quantified for each shRNA ( as described in Fig 1C ) , and was used to calculate z-score for the individual shRNAs and normalized to that of IRF9 ( Fig 2A and S1 Table ) [47] . A set of 25 primary shRNA hits were shortlisted on the basis of high z-scores ( >1 . 9 , Fig 2B ) , and used for secondary validation . The primary hits consisted of both known , as well as some novel , antiviral ISGs . In order to further validate these primary hits , stable HeLa cell lines expressing the individual ISG shRNAs were generated . In the ISG shRNA-expressing cells , the reversal of IFN-β-mediated inhibition of SeV C protein expression was measured . As a control , we used IRF9-specific shRNA expressing cells , which alleviated the IFN-β-mediated suppression of SeV C protein level ( Fig 2C , left ) . From the secondary validation assay , we narrowed down to a small subset of five ISGs , the knockdown of which reversed IFN-β-mediated inhibition of SeV C protein expression ( Fig 2C and 2D ) . Importantly , individual knockdown of any of these five ISGs , elevated the level of SeV replication in IFN-β-treated cells ( S1 Fig ) . Among the five anti-SeV ISGs identified by our screen , we focused on TDRD7 ( in human and Tdrd7 in mouse ) because: ( a ) it is a cytosolic protein and , therefore , is a potential candidate to inhibit paramyxoviruses , which replicate in the cytosol , ( b ) it has defined functional domains , which may be required for its antiviral action , ( c ) it has no known functions as an ISG or a viral restriction factor and ( d ) our previous microarray results indicated its robust transcriptional induction by virus infection [48] . In HeLa cells , our screening system , TDRD7 was transcriptionally induced by IFN treatment ( S2A Fig ) and the stable knockdown of TDRD7 ( S2A Fig ) had no impact on the cell viability ( S2B Fig ) . We further examined the inducibility of Tdrd7 in various mouse cells . In RAW264 . 7 cells , Tdrd7 mRNA was induced , as expected , by IFN-β; it was also induced by adding poly ( I:C ) to the medium to activate the TLR3 signaling pathway or transfecting poly ( I:C ) to activate the RLR pathway ( Fig 3A ) . SeV infection , which activates the RLR pathway , induced Tdrd7 mRNA as well ( Fig 3A ) . Similar induction by two strains of SeV was observed in mouse primary bone marrow-derived macrophages ( BMDMs ) and mouse embryonic fibroblasts ( MEFs ) ( S2C–S2E Fig ) . The viral induction of Tdrd7 in MEFs was triggered by IRF3-mediated induction of IFN , because it was not observed in Irf3-/- or Stat1-/- cells ( S2D and S2E Fig ) . Importantly , Tdrd7 mRNA was strongly induced in SeV-infected mouse lungs ( Fig 3B ) . These results clearly demonstrate that Tdrd7 , an ISG , is induced in virus-infected cells and tissues . To investigate the antiviral action of TDRD7 , we took two approaches: knockdown or knockout of the endogenous TDRD7 gene and ectopic expression of exogenous TDRD7 in multiple human and mouse cell types . In our test cells , we examined the endogenous levels of TDRD7 and as expected , the protein expression of TDRD7 varied between various cell types ( S2F Fig ) . In mouse lung epithelial cells , LA4 , the natural target cells of respiratory viruses , knockdown of endogenous Tdrd7 mRNA by two independent shRNAs enhanced the expression of SeV C protein ( Fig 3C , S3A Fig ) . In human retinal epithelial cells , ARPE19 , which express relatively higher levels of endogenous TDRD7 compared to LA4 , knockdown of TDRD7 ( Fig 3D , lower panel ) elevated the expression of SeV C protein ( Fig 3D ) . We confirmed these results in mouse fibroblasts , L929 , in which the knockdown of endogenous Tdrd7 ( S3B Fig ) also enhanced SeV C protein expression ( S3C Fig ) . Similar to HeLa , the stable knockdown of Tdrd7 in L929 cells had no impact on the cell viability ( S3D Fig ) . We further examined the antiviral activity of Tdrd7 in non-transformed immortalized MEFs , in which Tdrd7 was transcriptionally induced by IFN-treatment ( S3E Fig ) and its stable knockdown ( S3E Fig ) enhanced the SeV C protein expression ( S3F Fig ) . Using CRISPR/Cas9 system , we generated TDRD7 knockout ( TDRD7-/- ) human HT1080 cells ( Fig 3E , lower panel and S3G Fig ) . SeV C protein expression was elevated in TDRD7-/- cells ( Fig 3E , upper panel ) . In the reciprocal strategy , we ectopically expressed TDRD7 ( untagged or V5-tagged ) in cells that express low levels of endogenous TDRD7 , a scenario that mimics IFN-β-induced synthesis of ISGs . TDRD7 was ectopically expressed in HEK293T cells and confocal analyses showed its cytoplasmic distribution in uninfected cells ( S3H Fig ) . In these cells , TDRD7 strongly inhibited viral protein ( SeV C ) ( Fig 3F ) and mRNA ( SeV P mRNA , S3I Fig ) expression . Similarly , ectopic expression of Tdrd7 inhibited viral protein ( SeV C ) expression in mouse L929 ( S3J Fig ) and LA4 ( S3K Fig ) cells . We investigated whether the TDRD7-mediated inhibition of viral mRNA and protein leads to the reduction of infectious virion production . Indeed , the production of infectious virus particles was inhibited by ectopic expression of Tdrd7 in L929 cells ( Fig 3G ) . In subsequent experiments , we investigated the mechanism of anti-SeV action of Tdrd7 . Because the role of autophagy in SeV replication was not clear , we examined various stages of the autophagy pathway ( Fig 4A ) in SeV-infected cells . In L929 cells , SeV infection triggered robust induction of autophagy , which was analyzed by the degradation of p62 ( Fig 4B , upper panel ) and the generation of LC3-II ( Fig 4C ) , the indicators of two late stages of the autophagy pathway . The induction of autophagy was correlated with the expression of viral protein ( SeV C , Fig 4B , middle panel ) . In another cell type ( LA4 ) , SeV infection similarly triggered autophagy pathway , which was examined by LC3-II generation and p62 degradation ( S4A Fig ) . The molecular signatures of the early stages of autophagy , the dephosphorylation of Ser757 and phosphorylation of Ser317 of ULK1 , were also detected in SeV-infected cells ( Fig 4D ) . These results clearly indicate that SeV infection triggers different stages of the autophagy pathway in multiple cell types . To investigate whether ‘virus-induced autophagy’ is required for SeV replication , we took pharmacological and genetic approaches . A chemical inhibitor of autophagy , 3-MA , inhibited , whereas an activator of autophagy , rapamycin , promoted SeV mRNA synthesis ( Fig 4E ) . Furthermore , chemical inhibitors of various stages of autophagy also significantly suppressed the expression of SeV C protein ( S4B Fig ) . Similarly , knockdown of ATG5 ( Fig 4F , lower panel ) , a key component of autophagy pathway , significantly reduced viral protein expression in human cells ( SeV C , Fig 4F ) . As expected , the SeV-induced autophagy , examined by p62 degradation , was inhibited in these cells ( Fig 4F , p62 levels ) . Collectively , our results clearly indicate that SeV-induced autophagy pathway is required for its replication . The above results led to the hypothesis that TDRD7 might interfere with ‘virus-induced autophagy’ to inhibit SeV replication . To test this , we examined various stages of SeV-induced autophagy in cells , in which TDRD7 expression had been modulated either by ectopically expressing exogenous TDRD7 or by ablating the endogenous TDRD7 levels . In L929 cells , ectopic expression of Tdrd7 strongly inhibited the degradation of p62 ( Fig 5A ) , and the accumulation of LC3-II ( Fig 5B ) . As expected , the knockdown of endogenous TDRD7 in ARPE19 cells , triggered increased accumulation of LC3-II ( S4C Fig ) . Similar results were also obtained in Tdrd7-ablated murine macrophages ( RAW264 . 7 ) ( S4D and S4E Fig ) . In these cells , Tdrd7 knockdown also elevated SeV C expression ( S4E Fig , middle panel ) . We further investigated whether the antiviral action of TDRD7 is mediated by direct inhibition of autophagy or its ability to modulate IFN and ISG induction . The inhibition of autophagy pathway by knockdown of ATG5 in TDRD7-/- cells suppressed SeV C protein expression ( Fig 5C ) . As expected , these cells restored the ability of IFN to inhibit SeV C expression ( Fig 5C ) . Furthermore , Tdrd7 knockdown cells did not exhibit any significant difference in induction of IFN by SeV ( S4F Fig ) and ISG ( Ifit1 ) by IFN ( S4G Fig ) . These results demonstrated that the antiviral ISG , TDRD7 , inhibits ‘virus-induced autophagy’ to control SeV replication . Because autophagy is required for normal cellular homeostasis , it is also induced in response to many cellular stresses , other than virus infection . In the next series of experiments , we examined whether TDRD7 could inhibit autophagy induced by nutrient deprivation ( e . g . serum starvation , SS and Hank’s Balanced Salt Solution , HBSS ) or rapamycin , known activators of the autophagy pathway . As indicated by the LC3-II level , SS-induced autophagy was inhibited by ectopic expression of TDRD7 in human HEK293T cells ( Fig 5D ) . Similarly , in mouse L929 cells , ectopic expression of Tdrd7 inhibited degradation of p62 ( Fig 5E ) and enhancement of LC3-II levels ( S5A Fig ) . Accumulated LC3-II produces cytoplasmic puncta , which we analyzed by expressing a GFP-LC3 fusion protein in L929 cells . The cytoplasmic puncta formation by LC3-II was significantly reduced in Tdrd7-expressing cells ( Fig 5F ) . Similarly , rapamycin-induced LC3-II formation in L929 cells was inhibited by Tdrd7 ( Fig 5G ) . In Tdrd7-knockdown RAW264 . 7 cells , rapamycin induced a faster degradation of p62 ( S5B Fig ) and increased accumulation of LC3-II ( S5C Fig ) . These results demonstrated that Tdrd7 can inhibit autophagy , induced by both viral and non-viral agents , in various human and mouse cell types . Next , to identify the specific target of Tdrd7 , we biochemically analyzed the four stages of the autophagy pathway ( Fig 4A ) ; for this purpose we used SS of L929 cells to induce autophagy . Because we already knew that the ‘fusion’ ( p62 degradation ) and the ‘elongation’ ( LC3-II levels ) steps of the autophagy pathway were inhibited by Tdrd7 , we focused our attention to the further upstream pre-elongation steps of the autophagy pathway . As a readout of the ‘nucleation’ step , we analyzed PI3 kinase III activity by direct measurement of PI ( 3 ) P produced in virus-infected cells and by the generation of NADPH Oxidase ( phox ) [49 , 50] . In SeV-infected L929 cells , PI3K III was rapidly activated and its activity was inhibited by Tdrd7 expression ( Fig 6A , S5D Fig ) . We further expressed a GFP-conjugated p40 subunit of phox in L929 cells , which produced puncta structures upon autophagy stimulation [49] . Ectopic expression of Tdrd7 decreased the number of phox puncta structures ( Fig 6B , left panel ) , which was quantified by counting the number of puncta structures per GFP-expressing cell ( Fig 6B , right panel ) . These results established that the Tdrd7-mediated block is at the ‘nucleation’ step of the autophagy pathway or further upstream . We examined the effects of TDRD7 on the activation of ULK1 , a kinase essential for triggering of the induction stage . ULK1 is activated by phosphorylation of multiple Ser/Thr residues , among which Ser317 is a critical residue that is directly phosphorylated by the upstream kinase AMPK [51] . SS triggered robust phosphorylation of ULK1 on Ser317 , which was inhibited by ectopic Tdrd7 expression ( Fig 6C ) . Activated AMPK inhibits the activity of mTOR , another Ser/Thr kinase , which phosphorylates ULK1 on Ser757 to inhibit the autophagy pathway [51] . Therefore , dephosphoryation of Ser757 of ULK1 is a positive trigger of the autophagy pathway [51] . SS-induced dephosphorylation of ULK1 ( Ser757 ) was also strongly inhibited by Tdrd7 ( Fig 6D ) . These results pointed to AMPK as the target of TDRD7 . AMPK was required for SeV replication: in HeLa cells , knockdown of endogenous AMPK suppressed the expression of SeV C protein ( Fig 6E ) and ectopic expression of AMPK strongly enhanced SeV C protein expression in L929 cells ( Fig 6F ) . To determine whether only the physical presence or the kinase activity of AMPK is required for virus replication , we used a small molecule chemical inhibitor of AMPK kinase activity , Compound C ( CC ) . Treatment of cells with CC inhibited SeV C protein expression in both human and mouse cells ( Fig 6G and 6H ) , demonstrating that AMPK enzyme activity is required for SeV replication . As expected , CC treatment inhibited phosphorylation of AMPK ( on Thr172 ) upon SeV infection ( Fig 6G and 6H ) . We investigated the effect of Tdrd7 on the activation of AMPK , by monitoring the phosphorylation of its Thr172 . AMPK was rapidly phosphorylated by SS of cells , but Tdrd7 expression inhibited this phosphorylation ( Fig 6I ) . Similarly , HBSS-induced phosphorylation of AMPK on Thr172 was also inhibited by Tdrd7 ( S5E Fig ) . Importantly , SeV infection strongly activated AMPK and this step was inhibited by Tdrd7 ( Fig 6J ) . Together , our results clearly demonstrated that Tdrd7 is an inhibitor of the autophagy-inducing kinase AMPK and the antiviral action of Tdrd7 is mediated by its ability to inhibit AMPK activation . Next , we investigated whether TDRD7 can inhibit the replication of other paramyxoviruses by inhibiting autophagy . We chose two clinically important human paramyxoviruses , HPIV3 ( family: Paramyxoviridae , sub-family: Paramyxovirinae , genus: Respirovirus ) and RSV ( family: Paramyxoviridae , sub-family: Pneumovirinae , genus: Pneumovirus ) , to examine the generality of TDRD7 action . HPIV3 infection triggered robust autophagy in the infected cells , as examined by the increased LC3-II levels and p62 degradation ( Fig 7A ) . In ATG5 knockdown cells , HPIV3 replication was strongly inhibited , as manifested by the expression of virus-encoded GFP ( Fig 7B ) and the viral structural protein , HN ( Fig 7C ) . Similar to SeV and HPIV3 , RSV infection also triggered autophagy ( S6A Fig ) and its replication was inhibited in ATG5-knockdown human cells ( S6B Fig ) . To investigate whether TDRD7 inhibits the replication of HPIV3 and RSV , we used TDRD7-expressing human cells . Ectopic expression of TDRD7 inhibited HPIV3 replication , which was examined by both GFP and viral HN expression ( Fig 7D and 7E ) ; IFN-β-treatment , a known inhibitor of HPIV3 replication , was used as a positive control . Similar to HPIV3 , ectopic expression of TDRD7 strongly inhibited RSV replication , examined by the expression of viral proteins in the infected cells ( Fig 7F ) . As anticipated , AMPK was required for the replication of both HPIV3 and RSV . AMPK-knockdown cells expressed reduced levels of HPIV3-encoded viral protein ( HN ) ( Fig 7G ) and GFP ( Fig 7H ) . They also exhibited reduced RSV replication , as indicated by virus-encoded red fluorescent protein expression ( Fig 7I ) . Similar to SeV , HPIV3 infection caused phosphorylation of AMPK and as expected , this was inhibited by CC treatment ( Fig 7J ) . The treatment with CC caused strong reduction of HPIV3 HN protein expression ( Fig 7J ) . In addition to the inhibition of viral protein expression , CC strongly inhibited the production of infectious HPIV3 particles ( Fig 7K ) and infectious RSV particles ( Fig 7L ) . Our results clearly demonstrated that the kinase activity of AMPK was required for the replication of paramyxoviruses and TDRD7 inhibited their replication by inhibiting AMPK activation . Finally , to examine the specificity of TDRD7 action , we chose a member of another virus family , encephalomyocarditis virus ( EMCV; family: Picornaviridae , genus: Cardiovirus ) . EMCV replication , analyzed by the expression of viral RNA polymerase ( 3DPol ) , was not inhibited but enhanced by the ectopic expression of TDRD7 ( S7 Fig ) . These results clearly established the specificity of antiviral action of TDRD7 against viruses from different families .
Here , we report a novel mechanism by which the interferon system provides an antiviral response against paramyxoviruses ( Fig 8 ) . Using a high throughput genetic screen , we have identified a viral restriction factor , TDRD7 , which inhibited paramyxovirus-induced autophagy , a critical step of viral life cycle . Our mechanistic studies revealed that TDRD7 blocks the activation of AMPK , the enzyme that triggers autophagy . As expected , a chemical inhibitor of AMPK’s downstream activities also restricted the replication of paramyxoviruses . Because TDRD7 is a newly discovered viral restriction factor and its antiviral action is novel , we validated the critical results in multiple human and mouse cell types ( S2 Table ) . We , therefore , present a new mechanism by which the IFN system not only provides antiviral protection , but also controls cellular metabolic activity . In search for a common cellular mechanism that the paramyxoviruses utilize , we uncovered a role of autophagy , which was robustly induced in the early phase of viral life cycle and was required for a stage prior to the transcription of viral mRNA . As a model paramyxovirus , we used SeV , also known as mouse parainfluenza virus type I , because of its wide range of infectivity in vitro . SeV triggered a pro-viral autophagy pathway in the infected cells to facilitate its replication . Chemical inhibitors of autophagy blocked , whereas an activator of autophagy promoted , SeV replication . Genetic deficiency of the autophagy pathway significantly inhibited SeV replication . Many RNA viruses use the autophagy pathway to promote their replication [38–42] . Hepatitis C Virus ( HCV ) , Dengue virus and Poliovirus directly use the autophagosome membranes to facilitate their replication [38–40] . Measles virus triggers autophagy to sequester RIG-I in the autophagosome , to inhibit the antiviral action of IFN [42] . In contrast , autophagy impairs the replication of some DNA viruses . In neurons , autophagy is considered an antiviral response against Herpes Simplex Virus ( HSV-1 ) replication [52] . HSV-1 neurovirulence factor , ICP34 . 5 inhibits autophagy to support virus replication and pathogenesis [53] . However , HSV-2 and Varicella Zoster Virus ( VZV ) , two other α-herpesviruses require basal autophagy to promote their replication [53–56] . The members of γ-herpesviruses , Kaposi’s Sarcoma-associated Herpesvirus ( KSHV ) , Epstein-Barr virus ( EBV ) antagonize cellular autophagy using viral homologue of Bcl-2 [57] . It will be interesting to investigate whether these viruses can induce TDRD7 and whether TDRD7 has any effect on their replication . We demonstrated that AMPK , the initiator kinase of the autophagy pathway , is involved in paramyxovirus replication . AMPK was activated by SeV and HPIV3 infection and its activity was required for virus replication . AMPK has multiple cellular functions in addition to initiating autophagy . These include cell growth , mitochondrial biogenesis , and lipid and glucose metabolism [34] . By activating AMPK , the paramyxoviruses may also activate the autophagy-independent activities whose contributions to virus replication remain to be explored . However , our results clearly established that the autophagy-inducing effect of AMPK was critical for virus replication , because ablation of downstream components of the autophagy pathway had similar inhibitory effects . Because paramyxoviruses utilize the autophagy pathway , the role of AMPK in virus replication may be restricted to the activation of autophagy . Whether a specific component of the autophagy pathway , downstream of AMPK , is utilized by the paramyxoviruses , will be investigated in the future . Previous studies have indicated a role of lipophagy and macropinocytosis in promoting AMPK-induced virus replication . Dengue virus promotes AMPK-mTOR signaling pathway to promote lipophagy , a selective autophagy that targets lipid droplets [58] . Vaccinia virus activates AMPK to promote macropinocytosis and actin dynamics , which are required for viral entry into the cells [59] . A kinome screen has revealed that AMPK activity is required for HCMV replication; however , the exact mechanism is currently unknown [60] . KSHV directly interacts with AMPK via its K1 viral protein to promote cell survival [61] . In contrast to these , the picornavirus EMCV replication was promoted in the presence of TDRD7 . The exact mechanism behind this will require further investigation . How the IFN system regulates the autophagy pathway is largely unexplored; however , IFN treatment of cancer cells triggers autophagy via PI3K/mTOR signaling [62] . Some ISGs , e . g . PKR and RNase L promote autophagy in virus-infected cells as antiviral defense mechanisms [63–65] . Several studies indicate crosstalk between autophagy and innate immune signaling pathways; components of autophagy pathway are required for RIG-I signaling [66] . Autophagy is required for TLR7-induced type I IFN production by VSV-infected plasmacytoid dendritic cells [67] . Our study provides a new mechanism of IFN-mediated control of virus replication via inhibition of the autophagy pathway . Because TDRD7 prevents AMPK activation by non-viral stresses , such as nutrient deprivation , it may have broader effects on AMPK-dependent cellular processes in uninfected cells . For example , IFN is expressed in a low amount in uninfected immune cells , e . g . dendritic cells . The development and biological functions of these cells may be regulated by TDRD7’s anti-AMPK activity . Autophagy is required for cellular homeostasis and unregulated autophagy may lead to disease conditions [68] . In these scenarios , TDRD7-mediated autophagy inhibition would be beneficial . IFN signaling has been shown to inhibit AMPK activation [69]; however , the mechanisms or the involved ISGs are unknown . Our results indicate that TDRD7 is one of the executioner ISGs for the anti-AMPK activity of IFN . AMPK is involved in multiple cellular functions and is activated when cellular ATP levels are low , a scenario that mimics virus infection , which requires high metabolic activity of the infected cells . IFN signaling also regulates the activity of mTOR , a kinase that is involved in protein synthesis . AMPK directly inhibits the activity of mTOR by interacting with intermediate proteins [51 , 70] . Therefore , the IFN-mediated inhibition of AMPK may further activate mTOR signaling to enhance the synthesis of desired proteins in the virus-infected cells . Because TDRD7 is known to interact with cellular RNAs [44] , this activity may be required for its anti-AMPK functions . AMPK can be activated by long non-coding RNA ( lncRNA ) [71] , and it is speculative that sequestration of the lncRNA activator by TDRD7 may give rise to its AMPK-inhibitory action . Future investigation will be required to explore this possibility . How paramyxoviruses activate AMPK is an interesting question . To determine whether paramyxoviruses directly trigger AMPK activation by its interaction with viral proteins or by activation of upstream signaling pathways , will require additional investigation . Answers to these questions will lead to therapeutic potential by targeting the virus-AMPK interaction . Because temporary AMPK inhibition is not toxic , the chemical inhibitor is a potential candidate for antiviral therapy . However , AMPK is involved in modulating the functions of both innate immune cells , e . g . macrophages , and the adaptive immune cells , e . g . T cells [72 , 73] . Therefore , the use of AMPK inhibition as an antiviral strategy will require in-depth investigation of both innate and adaptive immune responses in virus-infected host . Because type I IFN is also involved in regulating the functions of these cells , TDRD7 may also contribute to the regulation of immune cell functions . TDRD7 knockout mice have only been examined for its role in lens and germ cell development [44 , 45] . Because TDRD7 is not present in high amount in majority of the cell types , the transcriptional induction by viruses or IFN exposure will uncover its new roles in other cell types as well . Whether the previously identified functions of TDRD7 can also be regulated by the IFN system will require further investigation .
Human cell lines HeLa , HT1080 , ARPE19 , HEK293T , and mouse cell lines L929 , LA4 , MEFs , RAW264 . 7 were maintained in DMEM containing 10% FBS , penicillin and streptomycin . All cell lines used in this study were maintained in the laboratory . Expression vectors of human and mouse TDRD7/Tdrd7 gene ( untagged ) were obtained from Origene and sub-cloned into lentiviral vector ( pLVX-IRES-puro , V5-tagged ) . AMPK expression plasmid was obtained from Addgene and was sub-cloned into lentiviral vector ( pLVX-IRES-puro , HA-tagged AMPK ) and Flag . VPS34 plasmid was obtained from Addgene . Autophagy inhibitors ( 3-MA , chloroquine , quinacrine , bafilomycin-1 ) or activators ( rapamycin ) , AMPK inhibitor ( Compound C ) were obtained from Sigma-Aldrich , human and mouse IFN-β were obtained from R&D , MTT was obtained from Fisher Scientific , and Lipofectamine 2000 was obtained from Thermo Fisher Scientific . The antibodies against the specific proteins were obtained as indicated below: anti-SeV C: raised in the authors’ laboratory [74] , anti-whole SeV antibody was a gift from John Nudrud ( Case Western Reserve University ) , anti-HPIV3 HN: Abcam , anti-RSV: Abcam , anti-3DPol: Santa Cruz , anti-TDRD7: Sigma-Aldrich , anti-LC3: Cell Signaling , anti-p62: Fitzgerald , anti-pULK1 ( Ser757 ) /anti-pULK1 ( Ser317 ) /anti-ULK1: Cell Signaling , anti-pAMPK ( Thr172 ) /anti-AMPK: Cell Signaling , anti-ATG5: Cell Signaling , anti-Actin: Sigma , anti-V5: Thermo Fisher Scientific , anti-Flag: Sigma-Aldrich , anti-Flag-agarose beads: Sigma-Aldrich . A custom generated , lentivirus-based shRNAmir library against human ISGs was obtained from Open Biosystems . The seed sequences for shRNA targeting each ISG have been described before [46] . The vector was designed to co-express the shRNA and GFP by a cytomegalovirus ( CMV ) promoter . The shRNA plasmids were packaged into lentiviral vectors in 96-well plates using the manufacturer’s instructions . HeLa cells were transduced with these lentiviruses for 48 h , when the cells were treated with 1000 U/ml of IFN-β . Twenty four hours later , the cells were infected with SeV ( Cantell ) at an MOI of 10 . After 16 h , cells were harvested , fixed with 1% paraformaldehyde in phosphate-buffered saline ( PBS ) for 10 min , permeabilized with 0 . 1% ( wt/vol ) saponin , and incubated with an anti-SeV polyclonal antibody and an Alexa Fluor 647-conjugated goat anti-rabbit secondary antibody . Cells were analyzed using a BD LSRFortessa flow cytometer ( BD Biosciences ) and the data were analyzed by FlowJo . Viral infection was determined based on the percentage of SeV-positive cells in shRNA-transduced ( GFP ) populations ( % infectivity , as illustrated in Fig 1C ) . The relative infectivity in each well was normalized to the wells containing a shRNA against IRF9 sequence to obtain z-scores [47] . Independent lentivirus stocks were used to validate the primary screen results in three independent replicates . Primary hits were defined as those z scores greater than 1 . 9 . The primary hits obtained from the high throughput screen were validated by stably expressing the respective ISG shRNAs ( which were positive in the primary screen ) in HeLa cells . We used non-targeting ( NT ) and IRF9-specific shRNAs as controls . The HeLa cells , stably expressing the shRNAs against the shortlisted ISGs were pre-treated with IFN-β followed by infection with SeV Cantell ( moi: 10 ) and viral protein ( SeV C ) expression was analyzed by immunoblot . For generating stable knockdown of TDRD7/Tdrd7 genes in human and mouse cells , the respective shRNAs [from the shRNA library , Open Biosystems ( TDRD7: GATCGCACATGTTTATTTA , used in all human cells of the study ) , ( Tdrd7#1: CAGGATTTGCCTCAGATTA , used in all mouse cells of the study ) or Sigma ( Tdrd7#2 , SHCLNG-NM_146142 , TRCN0000102515 , used only in LA4 cells ) ] were lentivirally expressed and the transduced cells were selected in puromycin containing medium . The stable knockdown cells were evaluated for levels of TDRD7/Tdrd7 by qRT-PCR in the absence or the presence of IFN-treatment or immunoblot . These and the control ( NT ) cells were evaluated for viral replication . ATG5-specific shRNAs ( Sigma # SHCLNG-NM_004849 ) were stably expressed lentivirally and the transduced cells were selected in puromycin containing medium . AMPK knockdown cells were generated using lentiviral shRNA plasmids ( Sigma # SHCLNG-NM_006251 ) and the transduced cells were selected in puromycin containing medium . Stable cell lines ectopically expressing epitope-tagged human and mouse TDRD7/Tdrd7 using lentiviral delivery systems ( pLVX-IRES-puro ) and were selected in puromycin containing medium . The stable cells were used for viral infection and other biochemical analyses . Wherever indicated , the stable cell lines were also generated by transfecting the untagged TDRD7/Tdrd7 plasmids ( from Origene ) and selecting the transfected cells with puromycin . Cells ectopically expressing AMPK were generated by lentivirally transducing HA . AMPK using pLVX-IRES-puro and selecting the cells in puromycin containing medium . HT1080 cells were transfected with either control ( sc-418922 ) or TDRD7-specific ( sc-407210 ) CRISPR/Cas9 plasmids . Transfected cells were sorted for high GFP-expressers using flow cytometry , and the GFP-expressing cells were expanded to isolate individual clones . These clones were examined for TDRD7 mRNA levels by qRT-PCR analyses and protein levels by immunoblot . SeV Cantell ( VR-907 ) and 52 ( VR-105 ) strains were obtained from Charles River , and the infection procedure has been previously described [74 , 75] . Briefly , the cells were infected by the viruses ( at an MOI specified in the figure legends ) in serum-free DMEM for 1 . 5 h , after which the cells were washed and replaced with normal growth medium . The virus-infected cells were analyzed at the indicated time for viral protein expression or as described in figure legends . For quantification of infectious SeV particles in the culture medium , standard plaque assays were performed , as described previously [74 , 75] . Recombinant RSV ( rrRSV ) and HPIV3 ( rgHPIV3 ) infections were carried out in serum-free DMEM at the indicated MOI . Infectious rrRSV and rgHPIV3 particles were analyzed by quantification of fluorescent foci forming units or the virus-infected cells were photographed using fluorescence microscope . EMCV infection was carried out using previously published procedure [76] , and the infected cells were analyzed by the expression of viral protein , as indicated in the figure legends . For analyses of nutrient starvation induced autophagy , the cells were washed ( three times ) with and then incubated in serum-free DMEM or HBSS ( Lerner Research Institute Cell Culture Core ) for the time period indicated in the figure legends . At the end of the incubation period , the cells were harvested and the lysates were analyzed for pAMPK ( Thr172 ) , pULK1 ( Ser317 and Ser757 ) , LC3 , p62 . Similarly , for the analysis of puncta formation by GFP-LC3 or GFP . p40 . PHOX , the cells were transiently transfected with these plasmids and then serum starvation was carried out . At the end of the exposure , the cells were fixed and confocal microscopy was performed . The puncta structures were manually counted in GFP-expressing cells from multiple fields . Immunoblot was performed using previously described procedures [74 , 75] . Briefly , cells were lysed in 50 mM Tris buffer , pH 7 . 4 containing 150 mM of NaCl , 0 . 1% Triton X-100 , 1 mM sodium orthovanadate , 10 mM of sodium fluoride , 10 mM of β-glycerophosphate , 5 mM sodium pyrophosphate , protease and phosphatase inhibitors ( Roche ) . Total protein extracts were analyzed by SDS-PAGE followed by immunoblot . The density of protein bands on the immunoblots was quantified using Image J program . Total RNA was isolated using RNA isolation kit ( Roche ) and cDNA was prepared using ImProm-II Reverse Transcription Kit ( Promega ) . For qRT-PCR , 0 . 5 ng of cDNA was analyzed using Applied Biosystem's Power SYBR Green PCR mix in Roche LightCycler . The expression levels of the mRNAs were normalized to 18S rRNA . To investigate in vivo gene expression , lungs were harvested from the SeV-infected mice and quickly frozen on dry ice . Total RNA was isolated from frozen lungs using Trizol extraction and the cDNAs were prepared using ImProm-II Reverse Transcription Kit and then subjected to qRT-PCR analyses as described above [74] . For the qRT-PCR analyses of the respective genes , the following primers were used: TDRD7-fwd: CGAGCTGTTCTGCAGTCTCA , TDRD7-rev: GCCATGGCATAGCAGGTAAT , Tdrd7-fwd: CTAAGGGCTGTCCTGCAGTC , Tdrd7-rev: AGAGTTGCCTTTGGCTTT , SeV P-fwd: CAAAAGTGAGGGCGAAGGAGAA , SeV P-rev: CGCCCAGATCCTGAGATACAGA , Ifnb-fwd: CTTCTCCGTCATCTCCATAGGG , Ifnb-rev: CACAGCCCTCTCCATCAACT , Ifit1-fwd: CAGAAGCACACATTGAAGAA , Ifit1-rev: TGTAAGTAGCCAGAGGAAGG , 18S-fwd: ATTGACGGAAGGGCACCACCAG , 18S-rev: CAAATCGCTCCACCAACTAAGAACG . HEK293T cells expressing V5 . TDRD7 were grown on coverslips , fixed in 4% paraformaldehyde , permeabilized in 0 . 2% Triton X-100 and subjected to immunostaining by anti-V5 antibody followed by Alexa Fluor-conjugated secondary antibody . The objects were mounted on slides using VectaShield/DAPI and analyzed by confocal microscopy . For GFP . LC3 and GFP . p40 . PHOX analyses , the cells expressing these plasmids were analyzed by confocal microscopy . The images were further processed and analyzed using Adobe Photoshop software . Multiple culture fields ( at least 100 cells from more than 20 fields ) were analyzed to select representative images and for quantification . L929 cells , transfected with Flag . VPS34 and V5 . Tdrd7 , were infected with SeV for the indicated time ( in figure legends ) , when the cell lysates were immunoprecipitated with Flag-agarose beads . The immunoprecipitates were analyzed for PI3K III activity by measuring the PI ( 3 ) P levels using Class III PI3K ELISA kit ( Echelon ) following manufacturer’s instructions . The PI ( 3 ) P levels in the mock-infected VPS34-expressing cells was expressed as 100 and all other values were normalized to this . Cells at the density of 10 , 000/well were cultured in 96 well plates for 2 days followed by addition of 10μl of 5mg/ml MTT solution in PBS and additional culturing for 4 hours . The water insoluble formazan was dissolved in DMSO and the absorbance was measured at 570nm [77] . The absorbance in control ( NT ) cells was expressed as 100 and all other values were normalized to this . C57BL/6 Wt mice , obtained from Taconic , were either mock-infected ( PBS ) or intranasally infected with SeV ( 52 strain , 120 , 000 pfu ) , as described previously [74 , 78] . The lungs were harvested after 2 days of infection , total RNA was isolated and analyzed by qRT-PCR . The statistical analyses were performed using GraphPad Prism 5 . 02 software . The ‘p’ values were calculated using two-tailed , un-paired Student’s t tests and are shown in the relevant figures . The results presented here are the representatives of at least three biological repeats .
|
The antiviral functions of interferons ( IFNs ) are mediated by the IFN-induced proteins , encoded by the IFN Stimulated Genes ( ISGs ) . Because ISGs are virus-specific , we performed a high throughput genetic screen to identify novel antiviral ISGs against Sendai virus ( SeV ) , a respirovirus of the Paramyxoviridae family . Our screen isolated a small subset of anti-SeV ISGs , among which we focused on a novel ISG , Tudor domain containing 7 ( TDRD7 ) . The antiviral activity of TDRD7 was confirmed by genetic ablation of the endogenous , and the ectopic expression of the exogenous , TDRD7 in human and mouse cell types . Investigation of the mechanism of antiviral action revealed that TDRD7 inhibited ‘virus-induced autophagy’ , which was required for the replication of SeV . Autophagy , a cellular catabolic process , was robustly induced by SeV infection , and was inhibited by TDRD7 . TDRD7 interfered with the ‘induction’ step of autophagy by inhibiting the activation of AMP-dependent Kinase ( AMPK ) . AMPK is a multifunctional metabolic kinase , which was activated by SeV infection , and its activity was required for virus replication . Genetic ablation and inhibition of AMPK activity by physiological ( TDRD7 ) or chemical ( Compound C ) inhibitors strongly attenuated SeV replication . The anti-AMPK activity of TDRD7 was capable of inhibiting other members of Paramyxoviridae family , human parainfluenza virus type 3 and respiratory syncytial virus . Therefore , our study uncovered a new antiviral mechanism of IFN by inhibiting the activation of autophagy-inducing kinase AMPK .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"cell",
"death",
"autophagic",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"molecular",
"probe",
"techniques",
"hela",
"cells",
"pathogens",
"cell",
"processes",
"biological",
"cultures",
"microbiology",
"immunoblotting",
"viruses",
"protein",
"expression",
"rna",
"viruses",
"cell",
"cultures",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"l929",
"cells",
"paramyxoviruses",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"cell",
"lines",
"viral",
"replication",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gene",
"expression",
"and",
"vector",
"techniques",
"biochemistry",
"cell",
"biology",
"virology",
"viral",
"pathogens",
"interferons",
"biology",
"and",
"life",
"sciences",
"cultured",
"tumor",
"cells",
"organisms"
] |
2018
|
A new mechanism of interferon’s antiviral action: Induction of autophagy, essential for paramyxovirus replication, is inhibited by the interferon stimulated gene, TDRD7
|
As populations spread into new territory , environmental heterogeneities can shape the population front and genetic composition . We focus here on the effects of an important building block of heterogeneous environments , isolated obstacles . With a combination of experiments , theory , and simulation , we show how isolated obstacles both create long-lived distortions of the front shape and amplify the effect of genetic drift . A system of bacteriophage T7 spreading on a spatially heterogeneous Escherichia coli lawn serves as an experimental model system to study population expansions . Using an inkjet printer , we create well-defined replicates of the lawn and quantitatively study the population expansion of phage T7 . The transient perturbations of the population front found in the experiments are well described by a model in which the front moves with constant speed . Independent of the precise details of the expansion , we show that obstacles create a kink in the front that persists over large distances and is insensitive to the details of the obstacle’s shape . The small deviations between experimental findings and the predictions of the constant speed model can be understood with a more general reaction-diffusion model , which reduces to the constant speed model when the obstacle size is large compared to the front width . Using this framework , we demonstrate that frontier genotypes just grazing the side of an isolated obstacle increase in abundance , a phenomenon we call ‘geometry-enhanced genetic drift’ , complementary to the founder effect associated with spatial bottlenecks . Bacterial range expansions around nutrient-poor barriers and stochastic simulations confirm this prediction . The effect of the obstacle on the genealogy of individuals at the front is characterized by simulations and rationalized using the constant speed model . Lastly , we consider the effect of two obstacles on front shape and genetic composition of the population illuminating the effects expected from complex environments with many obstacles .
Populations expand into new territory on all length and time scales . Examples include the migration of humans out of Africa [1] , the recent invasion of cane toads in Australia [2] , and the growth of colonies of microbes . Although populations often persist long after invading [3] , events during their spread can have long-lasting effects on their genetic diversity [4 , 5] . Considerable effort has been undertaken to understand the role of the invasion process on the evolutionary path of the population: The small population size at the edge of the advancing population wave amplifies genetic drift , reducing genetic diversity , which can culminate in the formation of monoclonal regions [4] . The fate of mutations—deleterious , neutral , or beneficial—occurring in the course of the expansion depends on the location of their appearance with respect to the edge of the wave [4 , 6–10] . While the genetic consequences of such range expansions have been studied in the field [11 , 12] , the complexity of natural populations makes it difficult to draw general conclusions . Laboratory expansions of microbes have thus become a useful tool to illustrate , test , and inspire theoretical predictions [13–16] . The majority of theoretical and experimental work on range expansions has focused on homogeneous environments while habitats in nature are often spatially heterogeneous with regard to dispersal or population growth , the two processes that lead to the expansion . Incorporating environmental heterogeneity into models of spreading populations [3 , 4 , 17 , 18] raises complex problems . Heterogeneity can affect any parameter that controls population dispersal or growth and there can be many spatial patterns of heterogeneity . Ecologists and population geneticists often focus on different consequences of environmental heterogeneity . Work in population dynamics and ecology typically concentrates on the effect of heterogeneity on invasibility and the speed and impact of an invasion in such environments [3 , 17 , 19–22] and is closely linked to the mathematical and physical aspects of front propagation [23 , 24] . In contrast , population genetics studies usually assume a successful invasion and ask how environmental heterogeneities affect the population’s genetic composition [4] . Although heterogeneous carrying capacities [25] , fragmented environments [26] , single corridors or obstacles [8 , 27] , and environmental patterns found on earth [7 , 8 , 28] have been addressed from a theoretical perspective , a systematic understanding is still missing . In this work , we study the population dynamics and relate the dynamics of the population front to the consequences on the genetic composition of the spreading population , thereby linking the evolutionary and ecological consequences of range expansions . We want to understand what happens when expanding populations confront environmental heterogeneities . For simplicity , assume that at each point the environment is a high quality habitat ( large growth rate at population density well below carrying capacity ) or a low quality habitat ( very small or zero growth rate ) . What constitutes low quality habitats depends on the population: For a macroscopic expansion of a terrestrial animal or a plant , lakes and mountain ranges are examples of low quality habitat . For microbes , regions with poor nutrients may represent an obstacle to colony growth . If ρ is the fraction of the environment that allows growth , we can distinguish between two scenarios: For 0 < ρ ≪ 1/2 , the ‘island scenario’ , a largely inhospitable environment is interrupted by islands or oases of growth; in contrast , for 1/2 ≪ ρ < 1 , the ‘lake scenario’ , a largely hospitable environment is punctuated by obstacles that impede growth ( Fig 1A ) . The island scenario , reminiscent of stepping stone models of population genetics [29] , with a weak coupling between nearby islands by migration , is a situation where genetic drift can lead to genetic uniformity on individual islands due to founder effects [30] . Here , we address the lake scenario in the context of spatial expansions . In addition to the fraction of the habitat not occupied by obstacles , ρ , we must also consider the number of obstacles , N , in the new habitat to be invaded . When N ≫ 1 , i . e . , when many ( non-overlapping ) obstacles are engulfed as the range expansion progresses , we expect that the principal effect of interest from a population dynamics perspective is the speed of the invasion and the roughness of the population front ( top of Fig 1A ) . If , at the other extreme , the expansion domain only includes one obstacle with a size comparable to the size of the habitat invaded , the obstacle’s size , shape , and spatial arrangement are expected to greatly influence the shape of the front at the length scale of the habitat ( bottom of Fig 1A ) . We study the effect of isolated obstacles on the spread of populations . Using a combination of experiment , theory , and simulation , we characterize the obstacle’s effect in a regime of sizes where the shape of the front is well-described by a phenomenological model of expansion with constant speed . The constant speed model reveals general effects which hold independently of the mechanisms for population spread: The perturbation in the population front induced by the obstacle is determined by the obstacle’s width , but not by its precise shape . The front shape , induced by the obstacle , governs the effect on the genetic composition of the expanding population . Expanding past obstacles reduces genetic diversity and privileges genotypes that just miss an obstacle’s edges , an example of ‘geometry-enhanced genetic drift’ , effects which are reflected in the genealogy of individuals at the front . In addition to the phenomenological model of front shape , we study a reaction-diffusion model , which enables us to compare experiments to a theoretical description in more detail and to understand the utility of the constant speed model in situations that extend beyond the experimental system studied here . To derive these findings , we combine an analytical model , simulations , and experiments . While the experiments are the basis for theoretical work , they also allow us to test theoretical predictions . The analytical model provides the opportunity to make predictions for a variety of environments and length scales while simulations are used to explore regimes not accessible to analytical solutions . In addition to using established theoretical and experimental methods to study expanding populations , we present a new laboratory model system which allows us to quantitatively study population spread in heterogeneous environments: the expansion of bacterial viruses ( bacteriophage ) on a lawn of sensitive and resistant bacteria . Patches of resistant bacteria represent obstacles to the spread of the phage and can be generated using a printing technique , allowing us to quantitatively test predictions . The growth dynamics of the phage system with phage and bacterial host differs from the one-species system with logistic growth , the FKPP equation [43] , often used to study population expansions theoretically and used as the basis for our reaction-diffusion model . The dynamics of phage spread is governed by the density of phage , the density of bacteria , and the density of uninfected bacteria . However , at long times , the profiles of infected and uninfected bacteria are slaved to the profile of a traveling population wave of phage with a constant speed , and its motion mimics the dynamics of the simpler FKPP model . This similarity makes sense , because it is well-known that under broad conditions the solution to reaction-diffusion equations produces traveling waves with constant velocity whose speed is determined by linearization at the foot of the wave [43] . Hence , we expect that our phage system reflects well aspects of range expansions that depend on the biology at the leading edge of the front and believe that it offers the prospect of studying demographic and evolutionary processes in complex , yet well-defined environments .
To explore the effects of obstacles on the population front dynamics , we employed a microbial model system , bacteriophage T7 spreading on a lawn of E . coli cells . Phage T7 , a virus of E . coli , infects bacterial cells and lyses them , releasing a large number of new phage particles which undergo passive dispersal and can infect nearby cells , a cycle of growth and replication that leads to an advancing population front . Phage T7 must kill the bacteria it infects [31] and its spread on a bacterial lawn is revealed by the growth of plaques ( clearings in the lawn due to lysis of bacteria ) , a fast process easily visualized by bright-field or fluorescence microscopy . We produce a heterogeneous environment for phage spread by incorporating regions which do not support propagation of the population wave: While a wild-type bacterial region ( marked by a constitutively expressed yellow fluorescent protein ) corresponds to a region supporting phage production , a resistant bacterial patch , an obstacle ( similarly marked by a red fluorescent protein ( mCherry ) ) , does not , see Fig 1B and Materials and Methods . A lawn with regions of susceptible and resistant bacteria represents a static , heterogeneous environment that the phage population travels across during its expansion and that can be easily visualized . We designed an assay that allowed linear fronts of expanding phage populations to encounter obstacles of defined shape . We modified a method that used a consumer inkjet printer to print sugar solutions [32] to deposit bacteria in defined patterns on agar surfaces ( such as was done using custom-made equipment [33] ) . The printer produces a field of bacteria on a rectangular ( 3 . 5 × 2 cm2 ) agar patch at sub-mm resolution ( Fig 1C , Materials and Methods , S1 Protocol and S1 Fig ) . The printed founder cells grow into a lawn , which is inoculated with a linear front of phage T7 close to or at the region with resistant bacteria ( Fig 1D ) . The phage population spreads on this heterogeneous lawn , with repeated cycles of infection and lysis of the susceptible bacteria leading to the loss of fluorescence and the expanding dark region . Fig 1D shows such a printed pattern at different stages of the phage invasion ( see also S1 Video ) : A linear population wave of phage encounters the region of resistant bacteria , the obstacle . The front curves as it passes the widest part of the obstacle and the two curved regions move along the far side of the object until they unite with each other , giving rise to a kink that disappears with time as the front moves beyond the obstacle . We used the difference between two consecutive images to detect the front of the plaque ( Materials and Methods ) , and studied the front position as a function of time . We define the unperturbed front position d ( t ) as the position of the plaque’s edge at a horizontal distance of ±3 mm away from the obstacle center as displayed in Fig 1D . Fig 2A , displaying front position as a function of time , shows that the plaque grows at an approximately constant speed , but slows down slightly over time , presumably due to E . coli entering stationary phase [34] . The varying slope illustrates variation in front speed among replicates . Overall , the plaque extends with an approximately constant speed of 0 . 2 mm/h . The profile of the fluorescence signal in direction of the moving front is constant in time as shown in S2 Fig . Fig 2B shows the front shape over time for multiple replicates . The evolution of the front is very similar across replicates , despite small variations in front speed and initial conditions . While the perturbation of the front by the obstacle and the formation of a kink is intuitive at first , we aimed for a quantitative model which can describe front shape and make predictions which can be tested experimentally . An arguably most minimal model assumes that the front moves with constant speed in direction normal to the front and ignores the microscopic details of how phage replicate inside bacteria and diffuse outside them . We dubbed this the ‘constant speed model’ . Fig 2C–2G illustrates the dynamics of a front encountering a rhombus-shaped obstacle: ( C ) An initially linear front moves forward uniformly until the obstacle is encountered . ( D ) When it encounters the obstacle , the front stays linear , but is interrupted in the interval where it would overlap with the obstacle . ( This is different from scenarios where a front of material encounters an obstacle and the obstacle “pushes” the material to the sides . ) ( E ) Beyond the obstacle’s widest points , propagation with constant speed creates circular arcs in the shade of the obstacle that are connected to a linear front on either side of the obstacle . The circular elements span a region given by the obstacle’s width and encounter the obstacle at a 90° angle . ( F ) For a rhombus with height 2h and width 2w , the arcs from the two sides meet and a kink forms after the front has traveled a distance w 2 + h 2 beyond the point of maximum width . ( G ) The kink then heals due to the increasing radii of the circular segments , i . e . , the size of the indent Δ decreases ( Δ ( d ) ∼ 1/d for large “downstream” distances d , where d is the distance perpendicular to the front from the widest point of the obstacle to the unperturbed portion of the front , see below ) . Fig 2H shows that the height of the rhombus-shaped obstacle does not play a role in determining front shape and thus the size of the indent while the kink heals: For an obstacle which is taller ( light red rhombus ) , the kink forms later and the circular arc where it forms is correspondingly shorter , but the shape of the downstream kink is independent of the obstacle height . Moreover , the circular shapes of the front on the far side of rhombus-shaped obstacles all fall onto the same master curve when plotted in units of w . A calculation shows that the indent size Δ as function of position d is indeed independent of h for rhombus-shaped obstacles and shows the same functional behavior if all lengths are expressed in units of w ( see also S1 Appendix ) : Δ ( d ) w = d w 1 - 1 - w 2 d 2 ≈ d ≫ w w 2 d . ( 1 ) Counterintuitively , the width of the obstacle thus is a more important predictor of downstream front shape than obstacle height . For rhombus-shaped obstacles , obstacle height determines where the kink forms , but not the shape of the front after formation of the kink . Below , we will discuss more general obstacle shapes and the influence of obstacle size on the applicability of the constant speed model . The constant speed model predicts the shape of the front at a given front position relative to the obstacle , this way allowing direct comparison with the experimentally determined front shapes in Fig 2B . While the constant speed model captures overall front shape and the transient character of the perturbation , the details of the predicted front ( black line ) differ from the experimental data ( colored lines ) . The experimental profiles consistently lag behind the predicted front . The constant speed model also predicts that the shape of the front , scaled by the obstacle’s width , is identical for all rhombus-shaped obstacles . To test this prediction , we repeated the experiment for three more obstacles , in total combining two different widths with two different heights . Fig 3A and 3B display front shapes and the indent sizes as measured for all four obstacle shapes . As predicted , the data collapse very well onto single lines if lengths are divided by w . ( This is not the case for other scalings , see S3 Fig . ) Although the constant speed model successfully predicts how the experimentally determined front shape ( colored lines ) scales with the obstacle’s width and the front’s distance from the obstacle , the experimental profiles display a lag relative to the predicted front ( black line ) for all four different obstacles and , equivalently , have a larger indent size than predicted ( Fig 3A and 3B ) . However , this quantitative disagreement does not affect the scaling behavior of Δ ( d ) ( Fig 3B , see S4 Fig for the same data on linear scale ) . The constant speed model captures the general features of the front dynamics observed in the phage experiment , but the deviation prompted us to study a more detailed model which considers more of the details of phage propagation . In addition to understanding the deviation , the more detailed model will shed light on the range of applicability of the constant speed model . The dynamics of plaque growth on homogeneous lawns has attracted considerable interest in the past [35–39] . A reaction-diffusion model , which captures the phage-bacterial interaction , the phage life cycle , and focuses on bacteriophage T7 , has been suggested by Yin and McCaskill [36]: phage bind bacteria to form infected cells , and these , with a rate constant , burst to release more phage . More complex successor models focusing on the delay between infection and release of progeny phage have been published [38] . We decided not to generalize these models to heterogeneous environments for two reasons: ( i ) The appropriate parameters are not known for our experiments and ( ii ) we aimed for a general description that allows us to reach conclusions that extend beyond the infection of E . coli by bacteriophage T7 . We therefore employed a coarse-grained reaction-diffusion model in which a species disperses by diffusion and replicates locally with logistic growth ( the local reproduction rate increases linearly with population density , then decreases and reaches zero at the carrying capacity of the environment ) except inside of obstacles . In the absence of obstacles , this model produces a traveling population wave with an exponentially shaped leading edge that moves at constant speed like the population wave resulting from the model by Yin and McCaskill [36] ( see Materials and Methods for a brief discussion of the differences and similarities between our model and phage population spread ) . Mathematically , it is a version of the Fisher-Kolmogoroff-Petrovsky-Piscounoff equation ( FKPP equation ) [40–43] , which captures the two processes underlying a range expansion , dispersal and growth . In our generalized version , the growth function depends on location to include the effect of obstacles . The time evolution of phage population density u ( x , t ) depending on location x and time t is given by: ∂ u ( x , t ) ∂ t = D eff ∂ 2 u ( x , t ) ∂ x 2 + k eff ( x ) u ( x , t ) K - u ( x , t ) , ( 2 ) where the first term describes dispersal by diffusion with an effective diffusion coefficient Deff . The second term captures local logistic growth with reproductive rate keff ( x ) and constant carrying capacity K . By rescaling the phage density u ( x , t ) , we can set K = 1 without loss of generality . In general , keff will depend on the bacterial density , the number of phage an infected bacterium releases ( the burst size ) , the adsorption kinetics of the phage , the rate for lysis of infected host , etc . [36] . We used our data to estimate the values for the phage’s effective diffusion coefficient and effective reproductive rate on the lawn of susceptible bacteria , D ^ eff and k ^ eff , respectively . For biologically relevant initial conditions , an unimpeded , linear population front moves forward with front speed v = 2 D ^ eff k ^ eff and front width parameter ξ = D ^ eff / k ^ eff [43] . The front propagation is governed by the dynamics at the leading edge , a behavior we expect for the phage system ( see Materials and Methods for a more detailed comparison to the phage system ) . From Fig 2A we find that the plaque front extends with a speed of about 0 . 2 mm/h . With a rough estimate of the diffusion coefficient of 0 . 0144 mm2/h ( Refs . [36 , 44] , Materials and Methods ) we can determine an effective growth rate of k ^ eff = 0 . 7 / h for the phage in our experiments . We assume that the phage’s diffusion coefficient inside the obstacle remains the same , but that no growth is possible due to the lack of susceptible bacteria , thus allowing us to set keff ( x ) = 0 inside the obstacle and k eff ( x ) = k ^ eff otherwise . With the diffusion coefficient to be the same inside and outside the obstacle , individuals can diffuse into the obstacle , reminiscent of an absorbing boundary . We think this is the case in the experimental system as well , although it is possible that the effective diffusion coefficient differs slightly in the region with resistant bacteria from the region with susceptible bacteria . We next numerically solved Eq 2 for the four different obstacles considered experimentally . Fig 3E displays two snapshots from the numerical solution of the wide obstacle ( see S2 Video ) . To quantify front shape at the leading edge , we defined front position as the boundary at which the population density is larger than 5% of the carrying capacity ( white line in Fig 3E , see Materials and Methods ) . Fig 3C displays the fronts . For the wide obstacle ( and the three other obstacles , S5 Fig ) we observe a lag of the front for the numerical solution ( colored line ) relative to the constant speed prediction ( black line ) , in qualitative agreement with the experiments . This lag also manifests itself in an increased indent size ( Fig 3D ) . To test sensitivity to the value of the diffusion coefficient , D ^ eff , we repeated the analysis for the wide obstacle with D ^ eff → 3 D ^ eff and D ^ eff → D ^ eff / 3 and found the lag to persist in both cases ( S6 Fig , Materials and Methods ) . As expected , for decreasing D ^ eff the lag , relative to the constant speed prediction , becomes smaller . Taken together , the numerical solution of the reaction-diffusion model produces a lag similar to that seen in experiments of the phage model system ( Fig 3 ) even though its parameters were not derived from the front’s shape . Where does the lag originate from and under which circumstances is the constant speed model a good description ? Both questions are closely related and can be explained by considering the relative importance of diffusion and movement of the front . While diffusion results in a mean distance traveled scaling with the square root of time , propagation of the front results in a position change of the edge of the front linear in time . In consequence , diffusion is the faster process at small length and time scales , while only propagation of the front leads to significant changes in population density at large length and time scales . The critical length dividing these two regimes is given by D ^ eff / v , the ratio of the diffusion coefficient D ^ eff to the speed of the advancing front v . Up to a prefactor , this ratio is given by the front width parameter ξ = D ^ eff / k ^ eff and is proportional to the width of the profile , perpendicular to the front , which is shown in Fig 3E and S2 Video [43] . Small kinks in the front will eventually be rounded and small bulges in the front smoothed out by diffusion . ( We disregard possible number fluctuations at the frontier and associated possible front instabilities [45] . ) The process of invasion , however , plays the major role in determining front shape on length scales much larger than ξ , justifying the use of the constant speed model as an approximate , but intuitively useful model for understanding how populations spread around obstacles . For our experimental system , ξ ≈ 0 . 1 mm which is considerably , but not strikingly , smaller than the scale determining the shape of the obstacle ( 1 − 2 mm ) . The simplicity of the reaction-diffusion model ( Eq 2 only has two free parameters . ) allows us to identify two mechanisms for the lag of the front relative to the constant speed model: a modified shape of the front close to the obstacle’s boundary ( S7 Fig , panel A ) and a slow-down of the front around the point of maximum width ( S7 Fig , panel B ) ; see also S1 Appendix . First , phage particles diffuse into the obstacle , recognizable by the obstacle in Fig 3E turning yellow at its boundaries . The obstacle is therefore partially absorbing and the phage sink leads to a reduced population density close to the boundary . This flux into the obstacle does not lead to a slow-down of the overall front , since the population extends far to the sides of the obstacle . Instead , a lagging boundary layer arises whose width is of the order of the only length scale , the front width parameter ξ , and which moves at the same speed as the unperturbed front ( S7 Fig , panel A ) . If the obstacle induces large perturbations to the front ( predicted by the constant speed model ) , this boundary layer will not be an important component of overall front shape . If the induced perturbation is small , however , the boundary layer becomes an important constituent of overall front shape . Because our obstacles are only one order of magnitude larger than ξ , we expect the lagging boundary layer to contribute to overall front shape and thus to the observed lag . ( We attribute the differing shapes of the front at the boundary layer between experiment ( Fig 3A ) and theory ( Fig 3C ) to the coarse-graining embodied in our model and differences in front detection . ) This effect will be modified if diffusion into the obstacle is not possible . Second , expansions of circular populations with radii smaller than ξ are significantly slowed down compared to linear population fronts or circular population fronts with radii much larger than ξ [43] . The constant speed model predicts that a circular segment arises with a radius initially smaller than ξ when the front passes around the point of maximum width ( Fig 2H , S7 Fig , panel B ) . A temporary slow-down is therefore expected until the radius becomes significantly larger than ξ , leading to an apparent lag of the front close to the obstacle . In general , we expect a contribution to lagging of the front wherever the boundary of the obstacle is kinked or curved ( i . e . , many infinitely small kinks are present ) . Both effects depend on details of the obstacle’s shape , but are tied to the length scale ξ . The perturbations predicted by the model of constant speed , however , are tied to the size of the obstacle: doubling the size of the obstacle leads to a doubling of the size of the perturbation due to the obstacle . Both effects should therefore lead to only small corrections to the front shape predicted by the constant speed model in the limit of large obstacles ( large in all directions , linear size L ≫ ξ ) . Since we expect the constant speed model to successfully predict the front shape for large obstacles , we can construct the front shapes for more general obstacle shapes and infer general properties of front shape that are independent of the shape of the obstacle ( see below and S1 Appendix ) , which is not possible using experiments or numerical solutions alone . While for rhombus-shaped obstacles the front shape is particularly simple ( the front consists of two linear and two circular segments only , Fig 2H ) , we now consider general convex , mirror-symmetric obstacles . When the front encounters an obstacle ( as when it first envelops the tip of a rhombus or the front half of a circle ) , the shape of the front remains planar . As the obstacle starts to decrease in width , each point along the boundary is the source of a circular segment contributing to the front ( similar to Fig 2E ) and the front thus encounters the obstacle at a 90° angle . Eventually , a kink or a “cusp” ( a kink with infinite slope ) forms on the far side ( S8 Fig ) , which heals downstream from the obstacle . Note that when changing the size of the obstacle ( without changing its shape ) the front’s overall shape stays unchanged , but gets scaled by the same factor that the obstacle size increased or decreased . In addition , as the kink heals downstream from the obstacle , we eventually recover a scaling result similar to Eq 1 . In this respect , the front exhibits a universal behavior far away from the obstacle: the perturbation inherited by the front is determined by the obstacle’s width , but not by its precise shape . Some quantitative predictions of the constant speed model for isolated circular , elliptical and elongated tilted obstacles are found in S1 Appendix ( S8 and S9 Figs ) . For objects that are not convex , we expect a similar overall behavior . An obstacle with a complicated shape still results in a kink which gradually heals as the front moves beyond the obstacle ( S10 Fig , S3 Video ) . We next examine how genetic composition of a population is shaped by obstacles it encounters , first predicting the obstacle’s effects based on the constant speed model followed by examining an experimental model system and simulations . As populations expand , genetic drift leads to the local reduction of genetic diversity and the formation of monoclonal sections at the front [4] . Thus we consider a population front that contains different neutral genotypes at different positions along the front encountering an obstacle . Fig 4A displays a series of front shapes together with a simplified initial genotype distribution indicated by orange , green , cyan , blue , and red colors . In the spirit of the constant speed model , we focus on front shape dynamics alone . The front segment with the cyan genotype either cannot propagate within the obstacle or , in the case of bacteriophage T7 , slows down dramatically since only diffusive motion is possible . Hence , this genotype is lost and does not contribute to the range expansion at later times . After passing the point of maximum width , the circular arcs of the front in the ‘shadow’ of the obstacle grow due to inflation and therefore genotypes marked in green and blue occupy a larger part of the front . As the kink heals , the green and blue genotypes occupy the part of the frontier that lies in the shadow of the obstacle . The abundance of these genotypes , which were a small fraction of the initial front , stays elevated even after the kink has healed . Note , however , that part of the increase in genotype abundance is transient since the arc length of the circular segments gets reduced during healing of the kink , although the radius still grows and the front thus locally experiences inflation . Fig 4A depicts a special symmetric initial condition of genotype frequencies that guarantees that genotypes benefitting from the inflation in the wake of the obstacle ( green and blue genotypes ) will meet precisely at the top of the obstacle . However , selectively neutral , grazing genotypes will meet at the top for quite general initial conditions , i . e . , there is always a boundary that gets ‘pinned’ at the top of the obstacle . The constant speed model argues that genotypes that fail to encounter the obstacle will be unperturbed , those whose segment of the front entirely collides with the obstacle will be eliminated , and those that graze the obstacle will be privileged because they will fill in the region downstream of the obstacle . We tested this idea experimentally by using fluorescent proteins as labels for selectively neutral genotypes . Because we could not produce expansions with fluorescent phage , we used the expansion of three E . coli strains , which express different fluorescent proteins . Two of the strains have been characterized previously [13] and we constructed a third strain which behaves comparably for the purpose of the experiment . We created heterogeneous agar plates by adding a circular membrane with an impermeable region just below the top layer of agar . We then launched linear expansions of mixtures of the three marked strains and observed them as they grew past the circle that blocked access to nutrients ( Materials and Methods , S11 Fig ) . Fig 4B displays the range expansion after approximately 1 , 2 , 6 , and 10 days of growth ( see S4 Video for additional time points and Materials and Methods for a description of replicate experiments ) . Before the population meets the obstacle , genetic drift at the population front leads to separation into monoclonal regions of the three different colors [13 , 47] . After formation of these sectors , their boundaries wander which results in a coarsening of sectors [13] . Abstracting from this effect , we observe that the sectors encountering the obstacle head-on are lost but the two that just graze the obstacle grow in its shadow , increasing the abundance of the corresponding genotypes , and meet at the top of the obstacle . These features are experimentally reproducible and verify the predictions of the constant speed model ( Materials and Methods ) . Next , we performed stochastic simulations , in which individuals reproduce on a lattice . In each step , a site along the front is randomly chosen and is copied onto one of the unoccupied neighbored sites thus propagating the front [5] ( a variant of the Eden model [48 , 49] extended here to track genotypes , see S12 Fig and Materials and Methods ) . Individuals never die , i . e . , occupied sites never change . The obstacle is a set of lattice sites which cannot be occupied . Fig 4C shows an initially linear front encountering an obstacle . The obstacle leads to dynamics that are qualitatively similar to the bacterial range expansion described above ( S5 Video ) . Fig 4C illustrates that individual genotypes can go extinct by two processes: the wiggling of sector boundaries caused by genetic drift [5 , 13 , 47] and collision with the obstacle ( light green to light blue sectors , Fig 4C ) . The genotypes that graze the corners that define the obstacle’s width dominate the curved part of the front during the subsequent inflationary phase ( green and purple sectors , Fig 4C ) and meet at the top of the obstacle . The founder effect of individuals near the point of maximum width also dominates the population’s genealogy downstream of the obstacle . Black lines in Fig 4C represent lineages of individuals at the front . As already evident from the labeling of genotypes with colors , none of the lineages pass through the area in front of the obstacle . In addition , most of the individuals at the curved part of the front originate from a small number of ancestors near the point of maximum width . Strikingly , none of these lineages pass through the point where the two populations meet behind the obstacle . Despite the expansion of the green and purple sectors right before they encounter each other , the parts of the population which meet at the top of the obstacle have no descendants at the front at late times . This effect arises because the two sectors encounter each other ( almost ) head-on just behind the obstacle ( Fig 4C ) . Although this effect does not manifest itself in the sectoring pattern we deduced from the constant speed model ( Fig 4A ) , it can be understood within the framework of the model: In Fig 2H , blue lines indicate the position of a virtual marker at the front coinciding with the overall shape of lineages behind the population front . This suggests that the constant speed model may also be used to predict the evolutionary dynamics of a spreading population in more complex environments . In summary , we found that the constant speed model used to describe the front shape of an expanding population can be used to understand the effects of an obstacle on the diversity of neutral genotypes in an expanding population . These include the loss of genotypes encountering the obstacle head-on and a founder effect from individuals present at the point of maximum width . Since the obstacle does not affect fitness of individuals carrying specific genotypes , but in an intricate way increases random fluctuations , these effects are an example of ‘geometry-enhanced genetic drift’ . In the regime in which the constant speed model is valid , the effect of the obstacle on front shape is limited to a downstream region as wide as the obstacle and is transient due to healing of the kink ( Fig 2H ) . If the habitat is much larger than a single obstacle , the overall front speed and shape is therefore not influenced by the presence of a single obstacle . What is the effect of many such obstacles introduced in Fig 1A ? Insight can be gained by considering two obstacles which are offset and placed behind each other as displayed in Fig 5A . We focus on the population front between both obstacles arguing that in the presence of many obstacles the population encounters such pairs of obstacles subsequently . Instead of displaying the front at different time points , a blue arrow is used to indicate the path of an imaginary marker at the front which propagates with constant speed ( compare to Fig 2H; the path of the marker can be derived from minimizing path length as explained in S1 Appendix on the ‘Analogy to geometrical optics’ ) . The dashed gray arrow indicates the path of that marker in the absence of the second obstacle illustrating that the presence of the second obstacle lengthens the path and thus slows down the front between both obstacles . This effect is more readily visible in a regular pattern of rhombus-shaped obstacles ( Fig 5B ) ; the path of the virtual marker repeatedly changes direction , the speed of the front in normal direction is lower . To address the same scenario using the reaction-diffusion model , we extended our analysis of Eq 2 using the parameters employed to study the case of a single obstacle ( Fig 3E ) . Fig 5C displays two snapshots of the numerical solution ( S6 Video , Materials and Methods ) . Both obstacles transiently perturb the front , but not independently . Due to the first obstacle , the front reaches the right side of the second obstacle after it reaches the left side , resulting in the formation of a kink which is asymmetric and slightly shifted to the right . The front lags the unperturbed part of the front ( dashed white line indicating front position at the boundary of the channel ) in the wake of both obstacles , effectively resulting in a slow-down . As discussed above , for the obstacle size considered , the constant speed model is not a perfect description of front shape . The lag observed relative to the unperturbed front therefore originates from a combination of the geometrical slow-down in Fig 5A and a slow-down for reasons discussed above . Extending our qualitative analysis of ‘geometry-enhanced genetic drift’ we repeated the stochastic simulation ( Fig 4C ) with two obstacles . In Fig 5D two snapshots of one realization of the simulation are displayed . Due to the stochastic nature of genetic drift , rigorously analyzing the effects of multiple obstacles on genetic diversity is not possible without detailed quantification . However , we observe two effects expected from our understanding of single obstacles . First , there is a sector boundary at the top of the second obstacle with two sectors encountering each other from opposite sides of the obstacle . Second , there is a lineage passing the first obstacle on the left and the second obstacle on the right and just grazes both obstacles ( see Materials and Methods for a discussion of other instances of the simulation ) . These two observations illustrate two of the effects we expect many obstacles to have on genetic diversity . First , a subset of sector boundaries will be created or pinned by obstacles introducing an effective wandering of sector boundaries not arising from genetic drift at the front ( the mechanism for wandering of sector boundaries in the absence of obstacles ) . Second , if lineages preferentially graze sides of obstacles , an effective description of the genealogy in a complex environment may be possible by considering a small subset of possible paths through the maze of many obstacles .
Organisms rarely spread across featureless habitats . Instead , they must find ways to survive and reproduce in the presence of environments that are heterogeneous in space and time . To investigate the effects of spatial heterogeneities on the dynamics and genetics of a spreading population , we combined experimental and theoretical approaches to understand the effect of single obstacles , of defined geometry , where organisms could not reproduce . When bacteriophage T7 encounters resistant E . coli the bacteriophage population front is perturbed in the wake of the obstacles by a sharp kink that slowly heals as the front moves on . A constant speed model gives an intuitive understanding of this perturbation , and a more detailed reaction-diffusion model rationalizes the deviation between experiment and the constant speed model’s predictions . In addition , the constant speed model explains that in a genetically diverse population , genotypes that run into the obstacle are eliminated and those that graze its sides increase in abundance , an example of ‘geometry-enhanced genetic drift’ . A mathematically rigorous analysis by Berestycki et al . predicted transient perturbations of planar waves encountering a single compact obstacle [50] . From a physical perspective , when the obstacle’s linear size L satisfies L ≫ ξ , where ξ characterizes the front width , considerable understanding of the perturbation is possible using a model based on front propagation locally and with constant speed . In this limit , the shape of the front can be found using a straightforward geometric construction that has an analogy in geometrical optics ( S1 Appendix ) . Interestingly , in this regime , a linear front stays unperturbed while it envelops the obstacle , in contrast to a first intuition based on a front of fluid material encountering an obstacle such as lava flow encountering a barrier [51] . However , a front of forest fire resembles the situation of phage propagating on a lawn of bacteria; indeed , ideas very similar to the model of constant speed are used to predict forest fires [52] . Our analysis of the front predicted by the constant speed model shows that the width of the obstacle , and not its precise shape , determines the long-term dynamics of the perturbation caused by the obstacle . The study of two obstacles placed behind each other and offset suggests an overall slow-down of the front in the presence of many obstacles . This effect is expected to depend on the density of obstacles . If obstacles are sparse , the healing of the perturbations implies that the front speed should be only marginally reduced compared to expansion in the absence of any obstacles . If obstacles are close enough to each other that the perturbation from the preceding obstacle has not healed much before the next obstacle is encountered , the perturbations will add up faster and an ensemble of obstacles will reduce front speed more . Obstacles regularly placed on a lattice are a special case: the existence of open channels , unobstructed by obstacles and much wider than the front width parameter , will allow the front to travel as fast as it would without obstacles; the remaining territory will then be explored in the wake of the front . If the density of obstacles is so high that no free paths connect the different boundaries of the environment , the traveling wave cannot propagate around obstacles . When dispersal within obstacles is possible , the population can nevertheless expand via migration between regions with good growth conditions , which is essentially the island scenario depicted in Fig 1A . Invasion is not possible in a scenario where population spread is hindered by a connected set of impermeable obstacles ( compare to the percolation threshold concept [53] ) . When the size of the obstacle approaches the parameter that sets the width of the population front , the constant speed model breaks down . This regime can be understood by numerically solving a two-dimensional reaction-diffusion system ( a generalized FKPP equation ) , which rationalizes the lag between the experimentally observed phage front and the constant speed model prediction , and bridges the gap to the regime where the length scale of the heterogeneities in the environment is much smaller than ξ and perturbations in front shape are therefore not expected . Following these ideas will complement recent studies using reaction-diffusion models to study invasion in heterogeneous environments [20] . From the experimental side , extending the printing assay to environments with many obstacles or creating random environments by spotting a mixture of bacteria susceptible and resistant to phage onto an agar surface ( S8 Video , S13 Fig , Materials and Methods ) might shed light on this question in the future . The models we used to describe the spread of phage populations were successful , even though they ignored the details of the bacteriophage life cycle . We found that for large obstacles the constant speed model is a good description for the front shape and expect in consequence the effects of ‘geometry-enhanced genetic drift’ to hold . How do these results apply to organisms whose spreading mechanism is very complex or even not well characterized ? In general we expect that for other population waves than those considered here , similar considerations hold . Specifically , we expect a length scale to exist beyond which a constant speed approximation results in a good description of front shape . Thus , our findings based on the constant speed model such as universality of the shape of the population front and the genetic consequences should be applicable to population fronts with a differing underlying dynamics , including pushed fronts ( fronts where the bulk of the wave and not the leading edge sets the dynamics [24] ) . Upon decreasing the obstacle size , we expect the constant speed model to break down and front shape and population spread to depend on the details of the biological system considered . Similar considerations hold when the nature of the heterogeneities is changed . We here considered obstacles with vanishing growth rate . If , however , the obstacle was a region with reflecting boundary conditions , i . e . , diffusive dispersal into the obstacle was not possible , we expect the behavior on large scales for large obstacles to be described by the constant speed model , while the behavior at small scales and near the boundary of the obstacle would be different . When the obstacle strongly perturbs the shape of the population front , we predicted that these perturbations affect the fate of genotypes and lead to ‘geometry-enhanced genetic drift’ . Analyzing the fate of lineages shows that the descendants of individuals trapped in front of the obstacle or born right behind it are lost in the long term . Our results are in qualitative agreement with a simulation study that demonstrated a decreased probability of survival of neutral ( and deleterious ) mutations occurring just in front of and right behind an obstacle [27] . Taken together , our results show that the long-term reproductive success of an individual depends on its position relative to the obstacle the population encounters as well as the random sampling that drives genetic drift , expanding the list of factors that contribute to ‘survival of the luckiest’ [54] . In addition to these effects , obstacles separating two genotypes for a considerable amount of time could also help preserve genetic diversity , similar to the mechanism of allopatric speciation . Theory and simulation , including a more detailed description of the evolutionary dynamics on top of the population dynamics is needed to disentangle these effects . More work is also needed to understand the effects of many obstacles: Considering the effects of two obstacles placed behind each other and offset illustrates that obstacles can shape the genetic composition of a population by creating transition zones between two genotypes and constraining the spatial structure of lineages . More research is needed to illuminate how genetic drift at the population front and ‘geometry-enhanced genetic drift’ due to obstacles together shape the genetic makeup of a population . Single obstacles could have pronounced effects on evolution beyond shaping the abundance of neutral genotypes . Because the small subpopulation that grazes the obstacle expands spectacularly , obstacles could make it easier for deleterious mutations to survive . This expansion protects deleterious mutations from extinction [55] and could establish a subpopulation which is large enough to survive for a considerable amount of time . This time span might be long enough for a second , beneficial mutation to occur , which has implications for the crossing of fitness valleys , similar to the effects due to genetic drift at the front [56] . Because the obstacle is not a population bottleneck , failure to acquire such a second mutation does not lead to a reduced fitness of the population in the long-term: genotypes that passed further away from the obstacle would eventually spread sideways and extinguish the deleterious allele if it is not rescued by a second beneficial mutation . A true spatial bottleneck would have fixed the deleterious allele and thus reduced the ( absolute ) fitness of the whole population . The rapid evolution of phage should allow such questions to be addressed experimentally in the future . Higher organisms differ in two important aspects from the E . coli system and the stochastic simulation , they generally are diploid or polyploid and their population is dynamic behind the front . While the consequences of obstacles on diploid organisms undergoing recombination are an important area for future research , our results are relevant for the evolutionary dynamics of mitochondrial DNA carried by diploids . Gene flow behind the front will blur the sector boundaries which are frozen in both our experiments and the stochastic simulation . However , this diffusive blurring is slow ( scaling as the square root of time since it is a diffusive process ) while the front advances more rapidly ( linearly with time ) . Hence , the boundaries remain well-defined for some distance behind the wave [57] . This study focused on a regime where the front dynamics can be described by a model of constant speed . In this regime , the results are insensitive to details of the expansions and details of the obstacle shape . Assuming that the population front is subdivided into monoclonal regions , an effect of ‘geometry-enhanced genetic drift’ can be described which is closely connected to the dynamics of front shape . We believe that these findings carry over to a wide variety of population expansions and beyond the neutral evolutionary dynamics considered here . Finally , although our analysis focused on single obstacles , we believe that our findings can be extended to natural environments , which typically display more complex heterogeneities . As a first step , by using the findings for isolated obstacles we expect to be able to describe observables such as the effective front speed and an ‘effective genetic drift’ in environments with obstacles such as those displayed in Fig 1A .
A semi-automated image analysis pipeline was used to extract front shapes , front positions , and indent sizes ( such as in Figs 2A , 2B , 3A and 3B ) from the fluorescence time-lapse information . First , the channel detecting YFP fluorescence was used to define a front right after the plaque boundary got established and the channel detecting mCherry was used to identify the three upper corners of the obstacle . This information was used to define a coordinate system with the obstacle’s center at the origin and the front extending in y-direction ( referred to as ‘upper region’ in the following , e . g . , Fig 1D ) . The image was cropped ( 5 . 3 mm in direction of front movement , 0 . 7 mm in direction opposite to front movement , and 3 mm to either side of the obstacle ) . After normalization using the upper , uninfected region , the difference between two consecutive frames ( YFP channel ) was used to identify the front . In the difference image the extending front manifests itself as a bright region whose upper boundary was identified using thresholding . The algorithm was tested manually since the front is easily detectable by eye , although the decay in fluorescence extends to about 1 mm ( S2 Fig ) . A few frames were excluded from the analysis due to jumps in the front which could be detected automatically using a threshold for local slope of front shape . Finally , front position was determined from the curve of the front close to the boundaries of the cropped region . Indent size was derived as the distance between the most lagging part of the front and front position ( after the kink has formed , i . e . , curve of front was defined around the axis of bilateral symmetry ) . The corners of the obstacles identified were also used to identify the size of obstacles , which were slightly smaller than in the printing template . When displaying data for individual obstacles the median of all the obstacles included in the analysis was used; for collapse plots , the size of each single obstacle was used to rescale data . For analysis , front detection was limited to frames obtained within 22 h even if the experiment lasted longer . After this time front detection becomes more challenging most likely due to the bacterial lawn transitioning into stationary phase .
|
Geographical structure influences the dynamics of the expansion of populations into new habitats and the relative importance of the evolutionary forces of mutation , selection , genetic drift , and gene flow . While populations often spread and evolve in highly complex environments , simplified scenarios allow one to uncover the important factors determining a population front’s shape and a population’s genetic composition . Here , we follow this approach using a combination of experiments , theory , and simulations . Specifically , we use an inkjet printer to create well-defined bacterial patterns on which a population of bacteriophage expands and characterize the transient perturbations in the population front caused by individual obstacles . A theoretical understanding allows us to make predictions for more general obstacles than those investigated experimentally . We use stochastic simulations and experimental expansions of bacterial populations to show that the population front dynamics is closely linked to the fate of genotypes at different parts of the front . We anticipate that our findings will lead to understanding of how a wide class of environmental structures influences spreading populations and their genetic composition .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
How Obstacles Perturb Population Fronts and Alter Their Genetic Structure
|
Leprosy is a chronic inflammatory disease caused by Mycobacterium leprae that mainly affects the skin and peripheral nervous system , leading to a high disability rate and social stigma . Previous studies have shown a contribution of genes encoding products of the lectin pathway of complement in the modulation of the susceptibility to leprosy; however , the ficolin-3/FCN3 gene impact on leprosy is currently unknown . The aim of the present study was to investigate if FCN3 polymorphisms ( rs532781899: g . 1637delC , rs28362807: g . 3524_3532insTATTTGGCC and rs4494157: g . 4473C>A ) and ficolin-3 serum levels play a role in the susceptibility to leprosy . We genotyped up to 190 leprosy patients ( being 114 ( 60% ) lepromatous ) , and up to 245 controls with sequence-specific PCR . We also measured protein levels using ELISA in 61 leprosy and 73 controls . FCN3 polymorphisms were not associated with disease , but ficolin-3 levels were higher in patients with FCN3 *2B1 ( CinsA ) haplotype ( p = 0 . 032 ) . Median concentration of ficolin-3 was higher in leprosy per se ( 26034 ng/mL , p = 0 . 005 ) and lepromatous patients ( 28295 ng/mL , p = 0 . 016 ) than controls ( 18231 ng/mL ) . In addition , high ficolin-3 levels ( >33362 ng/mL ) were more common in leprosy per se ( 34 . 4% ) and in lepromatous patients ( 35 . 5% ) than controls ( 19 . 2%; p = 0 . 045 and p = 0 . 047 , respectively ) . Our results lead us to suggest that polymorphisms in the FCN3 gene cooperate to increase ficolin-3 concentration and that it might contribute to leprosy susceptibility by favoring M . leprae infection .
Leprosy is a chronic infectious disease caused by Mycobacterium leprae that mainly affects the skin and peripheral nerves [1] and can cause progressive and permanent damage , if untreated . Despite the disease elimination in 119 of the 122 countries where it was considered a public health problem , Brazil still ranked second in the world , behind India and accounts for 92% of leprosy cases in the Americas [2] . Upon exposure to M . leprae , most individuals are intrinsically resistant to infection . Among those who are susceptible , infection may progresses to a wide spectrum of manifestations , with two polar forms: the tuberculoid leprosy and the lepromatous leprosy . Tuberculoid leprosy is characterized by strong cell-mediated immunity , type 1 cytokine profile , low bacillary load and localized lesions . On the other hand , lepromatous leprosy is characterized by low cellular response , type 2 cytokine profile , high bacillary load and disseminated lesions [3] . There is enough evidence to suggest that susceptibility to leprosy and to different clinical manifestations is markedly influenced by host genetic factors [3–6] . Ficolins ( Ficolin-1 or M-Ficolin , Ficolin-2 or L-Ficolin and Ficolin-3 or H-ficolin ) are soluble molecules of the innate immune system that recognize a wide range of pathogen-associated molecular patterns ( PAMPs ) [7 , 8] . Ficolins form complexes with MASPs ( MBL-associated serine proteases or MASPs ) and activate complement through the lectin pathway , leading to opsonization and phagocytosis of pathogens , and stimulating the production of inflammatory cytokines and nitric oxide [9] . Most active ficolins are composed of four trimeric subunits . Each monomer is formed by an N-terminal region , a collagen-like domain and a fibrinogen-like domain; important in the oligomerization process , in the MASP/phagocyte interaction and in the recognition of specific PAMPs in pathogens , respectively [9] . Ficolins 1 , 2 and 3 are encoded by FCN1 and FCN2 genes on 9q34 and FCN3 on 1p36 . 11 , respectively [10] . In previous studies , we demonstrated that FCN2 gene haplotypes associated with normal ficolin-2 levels have a protective effect against leprosy [11] and that FCN1 gene -271DelT , -399A , -542G , -1981A polymorphisms were associated with susceptibility to leprosy [12] . FCN3 comprises eight exons , one of them being an alternative exon ( exon 4 ) . Both FCN3 transcripts , with and without exon 4 , occur especially in the lung , but also in the liver , heart , kidney , adrenal gland , breast , spleen , thyroid and visceral adipose tissue , but the shorter transcript is less abundant [13] . Exons are highly conserved: although 164 polymorphic noncoding variants are currently listed in Ensembl , all coding DNA variations ( including those synonymous ) occur at global frequencies below 1% . The g . 1637delC variant ( rs532781899 ) in exon 5 is actually the only one reported to be polymorphic at the global scale [14] . It causes a frameshift , leading to premature termination of the translation product . This generates a truncated protein , unable to perform PAMP recognition and complement activation , which may be associated with repetitive infections in some individuals [15–18] . Ficolin-3 has 299 amino acids and is the most abundant ficolin in serum , with a median concentration of ~19500 ng/mL ( range 3000–60300 ng/mL ) [17 , 19] . Low ficolin-3 concentration in serum has already been associated with the pathophysiology of sarcoidosis [20] , chemotherapy-related infections in children [21] , Crohn's disease [22] and heart failure [23] . On the other hand , high ficolin-3 levels were associated with Systemic Lupus Erythematosus [24] , ovarian tumors [25] and seem to be a risk factor for shorter graft survival in kidney transplantation [26] . The impact of FCN3 polymorphisms in other diseases has yet to be explored . In this work , we investigated whether FCN3 polymorphisms and ficolin-3 serum levels play a role in the susceptibility to leprosy and observed an association between high ficolin-3 levels in serum and the disease .
The study was approved by Human Research Ethics Committee , Health Sciences Sector at Federal University of Parana , Brazil ( approval number: 218 . 104 ) . Study subjects comprised consecutive outpatients from the Hospital de Clínicas , Federal University of Paraná , State Health Department of Paraná , and inpatients from the Sanitary and Dermatologic Hospital of Paraná , both located in Curitiba , southern Brazil . For all 190 patients ( 38 . 4% female; 82 . 3% Euro-Brazilian , 17 . 7% Afro-Brazilian; average age of 51 . 5 years , range 18–94 ) , leprosy was diagnosed on the basis of the clinical and histopathological features of affected lesions and classified according to the criteria of Ridley and Jopling [27] . The initial diagnosis was lepromatous leprosy for 114 ( 60% ) , tuberculoid leprosy for 15 ( 7 . 9% ) , and borderline leprosy for 28 patients ( 14 . 7% ) ; 10 patients ( 5 . 3% ) had an undetermined form of leprosy and 23 ( 12 . 1% ) were unspecified . As control subjects , 245 , unrelated , symptom-free blood donors from HEMEPAR ( Centro de Hematologia e Hemoterapia do Paraná ) were assessed ( 53% female; 80% Euro-Brazilian , 15% Afro-Brazilian; average age of 37 . 7 years , range 18–61 ) . Patients and control subjects had a similar socioeconomic status , were from the same geographical area , and shared the same ethnic background . All patients and control subjects provided written informed consent . DNA extraction was performed using QIAamp DNA extraction kits ( Qiagen ) according to the manufacturer’s instructions . Three FCN3 SNPs were assessed by sequence-specific amplification method ( PCR-SSP ) , being: g . 1637delC ( rs532781899 ) in exon 5; g . 3524_3532insTATTTGGCC ( rs28362807 ) in intron 5 and g . 4473C>A ( rs4494157 ) in intron 7 ( Table 1 ) . Although there are other noncoding polymorphisms not in LD with those selected , they do not tag a haplotype block in the Iberian population ( data from the 1000 Genomes project ) , which is representative for most Euro-Brazilians , as do the intronic polymorphisms chosen for this study . FCN3_Ex5_1637del_R or FCN3_Ex5_1637C_R were conjugated with FCN3_Ex5_F primer to generate a fragment of 748 bp . An amplification control fragment of 500 bp of FCN2 gene was simultaneously generated . FCN3_In5_3524_3532del_F or FCN3_In5_3524_3532ins_F were conjugated with FCN3_In7_4473A_R or FCN3_In7_+4473C_R primer to generate a fragment of 984 bp . A control fragment of 431 bp of HGH gene was simultaneously generated . The intron 5_intron 7 haplotypes were determined without having to infer their phase on the chromosomes due to the PCR-SSP approach with primers annealing on two different SNPs ( S1 Fig ) . PCR was carried out in a final volume of 15 μl in a T100TM thermocycler ( BioRad ) . PCR conditions were as follows: 0 . 7 μM for exon 5 and 0 . 2 μM for intron 5 and 7 SSP primers and 0 . 1 μM control primers , 1 × Coral Load PCR buffer ( Qiagen , Hilden , Germany ) , 2 . 0–1 . 75 mM MgCl2 ( only for exon 5 primers reaction , Qiagen , Hilden , Germany ) , 1 . 5% glycerol , 0 . 2 mM deoxyribonucleotide triphosphate ( dNTP ) ( Invitrogen , São Paulo , Brazil ) , 0 . 5% Q Solution ( only for intron 5 and 7 primers reaction , Qiagen , Hilden , Germany ) , 0 . 03 U/μl of Taq polymerase ( Invitrogen , São Paulo , Brazil ) , 20 ng/μl DNA and water to complete the final volume . The amplification protocol starts with a 3 min denaturation step at 96°C , followed by 35 cycles of 15 sec at 94°C , 30 sec at the specific annealing temperature and 30 sec at 72°C , concluding with 5 min at 72°C in the final DNA extension step . Annealing temperature decreased every 10 cycles ( 64°C , 62°C and 60°C; 60°C , 58°C and 56°C for exon 5 and intron5_7 primers reaction , respectively ) , according to a previously published “touch-down” strategy which assures higher specificity to the amplification , while providing a larger amount of the desired PCR product [28] . The haplotypes defined by two SNPs , amplified by a pair of SSPs , were identified by the presence or absence of specific bands after agarose gel electrophoresis . Control bands informed on the quality of the reactions . We measured ficolin-3 concentrations in 1:250 diluted sera ( 1:150 or 1:50 when necessary ) of 61 patients and 73 controls with the same proportion of selected FCN3 genotypes , using the enzyme-linked immune sorbent assay HK 340 ( Hycult Biotechnology , Uden , The Netherlands ) . Relative low ficolin-3 concentration was defined as <10368 ng/mL , which corresponded to the 20th percentile , and high levels as >33362 ng/mL , corresponding to the 80th percentile among controls . Genotype and allele frequencies were obtained by direct counting . The hypothesis of Hardy–Weinberg equilibrium was verified using the approach of Guo and Thompson implemented in the ARLEQUIN software package version 3 . 1 ( http://anthro . unige . ch/arlequin/ ) . Tests of independence between patients and controls , as well as between patients with the lepromatous and non-lepromatous forms ( tuberculoid , borderline and undetermined form of leprosy ) , were performed using Fisher exact test . Ficolin-3 levels were compared between the groups using nonparametric Mann-Whitney/Kruskal–Wallis tests using GraphPad Prism 3 . 0 software package . Two-tailed P-values less than 5% were considered significant . Logistic regression models were used to adjust results for age , sex and ethnic group distribution , using STATA v . 9 . 2 ( Statacorp , USA ) . Due to the sample size , statistical analyzes were performed between lepromatous leprosy and non-lepromatous patients ( which included tuberculoid , borderline and undetermined leprosy ) . Clinical forms of leprosy was compared with healthy controls since this approach could reveal subtle differences not apparent when comparing them just with leprosy per se .
FCN3 genotype distribution was in Hardy-Weinberg equilibrium . The allelic frequencies in Euro-Brazilian and Afro-Brazilian patients and controls did not differ from those reported in the HapMap project for CEU ( North-Americans of Northern and Western European ancestry from Utah ) and YRI ( Yoruba in Ibadan , Nigeria ) populations [29] . There was no difference in the allelic and genotypic frequencies between controls and leprosy patients , as well as lepromatous and non lepromatous groups ( Table 2 ) . Importantly , due to the very low frequencies of g . 1637del , our study was underpowered in detecting associations with this SNP . Euro-Brazilians and Afro-Brazilians as well as males and females also had similar allelic and genotypic frequencies . There was strong linkage disequilibrium ( LD ) between the two non-coding SNPs rs28362807 ( g . 3524_3532insTATTTGGCC , intron 5 ) and rs4494157 ( g . 4473C>A , intron 7 ) , as indicated by the correlation coefficient values ( r2 ) ( Fig 1 ) . The low r2 values observed for g . 1637delC SNP reflect the completely discrepant frequencies of this SNP ( an uncommon deletion ) in comparison to the other two SNPs . Different combinations of investigated polymorphisms ( g . 1637delC , g . 3524_3532insTATTTGGCC and g . 4473C>A ) resulted in 5 observed haplotypes , one of them possibly recombinant . According to the degree of sequence identity with the Pan troglodytes FCN3 gene sequence , the most probable ancestral haplotype ( named as *1 ) is formed by g . 1637C , g . 3524_3532ins and g . 4473C alleles ( for short , CinsC ) ( Fig 2 ) . The *2A ( CdelC ) haplotype was the most frequent , with 69–74% frequency in all investigated groups followed by *2B1 ( CinsA ) haplotype ( 18–27% ) . Other haplotypes , including *2B2 ( delinsC ) , harboring the deletion in exon 5 , were rather uncommon ( Table 2 ) . All the alleles found in our Afro-Brazilian sample were also found in all African population from NCBI and 1000 Genomes Project [14 , 29] , giving us clues about the likely African origin of these polymorphisms . The median level of ficolin-3 observed in the control group ( 18231 ng/mL [3129–60300 ng/mL] ) is in good agreement with published data in adults ( ~19500 ng/mL; [19] ) . The concentration of ficolin-3 in serum was higher in leprosy per se ( 26034 ng/mL , p = 0 . 005 , OR 6 . 80 [1 . 65–28] ) and lepromatous patients ( 28295 ng/mL , p = 0 . 016 , OR 6 . 77 [1 . 43–32] ) compared with controls ( 18231 ng/mL ) , even after correction for age , sex and ethnic group , but did not differ between lepromatous and non lepromatous groups ( Fig 3A ) . In addition , high ficolin-3 levels ( >33362 ng/mL ) were more common in leprosy per se ( 34 . 4% ) and in lepromatous patients ( 35 . 5% ) than controls ( 19 . 2%; p = 0 . 045 and p = 0 . 047 , respectively ) , while the frequency of low ficolin-3 concentrations ( <10368 ng/mL ) did not differ between the groups ( Table 3 ) . To assess whether the polymorphisms g . 1637delC ( rs532781899 ) ; g . 3524_3532insTATTTGGCC ( rs28362807 ) and g . 4473C>A ( rs4494157 ) of FCN3 gene were related to the variation of ficolin-3 serum concentration , they were evaluated separately and as haplotypes . The g . 1637del/1637C heterozygote controls had lower ficolin-3 median concentration ( 3762 ng/mL ) than g . 1637C/1637C homozygote controls ( 18382 ng/mL , p = 0 . 023 ) , corroborating previous published data [17 , 32] . Surprisingly , we did not observe the same difference between leprosy patients ( p = 0 . 143 ) , moreover , in both g . 1637del/1637C and g . 1637C/1637C genotypes , patients had higher ficolin-3 concentration than controls ( S1 Table ) . The effect of remaining polymorphisms in ficolin-3 levels was evaluated by removing all g . 1637del/1637C individuals of the analyses . In the dominant model , ficolin-3 levels were higher in leprosy per se and lepromatous patients with g . 4473A allele when compared to g . 4473C/4473C ( p = 0 . 043 and p = 0 . 028 , respectively ) . Moreover , the g . 3524_3532ins and g . 4473A alleles were associated with higher ficolin-3 levels in leprosy patients , compared to controls ( p = 0 . 042 , p = 0 . 040; respectively ) . Under a recessive model , homozygous genotypes for rarer alleles ( g . 3524_3532ins and g . 4473A ) seem to lead to increased ficolin-3 in patients when compared to controls . All results were adjusted for demographic factors by logistic regression ( S1 Table ) . The same pattern of associations was observed in the haplotypes , with ficolin-3 levels being higher in leprosy patients with the *2B1 haplotype ( CinsA ) , than in those without it ( 32795 vs . 21958 ng/mL , p = 0 . 033; excluding individuals with the deletion in exon 5 ) , and than in *2B1 controls ( 20790 ng/ml; p = 0 . 032 ) . Similarly , lepromatous patients with the *2B1 haplotype had higher ficolin-3 levels than lepromatous patients without it ( 35731 vs . 22294 ng/mL , p = 0 . 021 ) and than *2B1 controls ( p = 0 . 032 ) . There was a trend in the same direction , between controls with and without *2B1 ( p = 0 . 080; Fig 3B ) .
There are several evidences indicating an immunoregulatory role for the pattern recognition molecules ( PRMs ) of the lectin pathway in the susceptibility and clinical expression of leprosy [6 , 11 , 12 , 33] . This is the first study addressing FCN3 polymorphisms and ficolin-3 levels in leprosy . In previous studies , high MBL levels were shown to increase the susceptibility to lepromatous leprosy [6 , 34] , whereas MBL2 haplotypes/genotypes conferring low MBL levels and deficiency in complement activation , conferred resistance against the development of lepromatous and borderline leprosy [33] . Furthermore , FCN2 and FCN1 haplotypes have protective effects against the susceptibility to leprosy per se [11 , 12] . Thus , components of the lectin pathway seem to be good candidates as biomarkers to be associated with the host response against M . leprae . In this study , we observed that higher ficolin-3 levels were associated with the disease per se and with lepromatous leprosy . Higher ficolin-3 levels were also associated with a specific FCN3 haplotype , containing an insertion in intron 5 ( g . 3524_3532insTATTTGGCC ) and the A allele at position +4473 in intron 7 ( g . 4473A ) . Interesting , introns 5 and 7 contain CpG islands . They are also enriched for typical histone modifications , known to characterize active enhancers ( H3K27ac—H3 acetylated at lysine 27 and H3K4me1- H3 monomethylated at lysine 4 ) [35 , 36] . Different regulatory proteins ( such as CTCF—CCCTC-binding factor , SPI1—Spleen focus forming virus ( SFFV ) Proviral Integration 1 and EGR1—Early growth response protein 1 ) bind to these intronic regions , as seen by chromatin immunoprecipitation assay in different cell lines ( such as NHLF—lung fibroblasts , BJ—skin fibroblast and HMC—cardiac myocytes ) [37] . Variants within these sequences , as those investigated here ( or others strongly linked ) , may increase enhancer activity in response to inflammation signals , causing enhanced gene transcription and higher protein levels ( Fig 4 ) . This would explain why we only found clear evidence for an association in patients , which probably present an inflammatory response , observing only a trend in healthy individuals . On the other hand , we cannot dismiss the possibility that binding of regulatory proteins in these sites could modulate splicing of the alternative exon 4 , whose inclusion in the most abundant FCN3 transcript leads to a longer collagenous tail . High ficolin-3 levels have been previously reported in the sera of systemic lupus erythematosus patients [24] , children with acute leukemia [21] , ovarian cancer patients [25] and associated with graft loss in kidney transplant recipients [26] , indicating a probable pathogenetic role for high ficolin-3 concentration in these disorders . We hypothesize that high ficolin-3 levels in leprosy patients probably play an unfavorable role by facilitating M . leprae dissemination . M . leprae may explore complement activation and opsonization induced by PRMs as one of the invasion mechanisms of macrophages and consequent evasion of the immune system [38] . Indeed , lectin pathway PRMs , including MBL ( mannose-binding lectin ) and ficolin-2 , were shown to bind to mycobacteria leading to MASP2 ( MBL-associated serine proteases ) activation [38] . Although no direct binding of ficolin-3 on M . bovis BCG cell surface was found [38] , it is know that the mycobacterial cell walls comprise long polymers of N-acetyl glucosamine ( GlcNAc ) [39] , which is a ligand for ficolins [40] , and could therefore be a potential target for ficolin-3 . The effect of the g . 1637delC ( rs532781899 ) polymorphism reducing ficolin-3 serum concentration [17] was only evident in controls . This is most probably a sampling effect , because among the two heterozygous g . 1637del/1637C patients , one also carried the g . 3524_3532ins and g . 4473A alleles , associated with increased ficolin-3 concentration . Thus , whereas the FCN3 transcript with the g . 1637del allele would produce a non-functional protein in this individual , the g . 3524_3532ins and g . 4473A alleles , combined in a haplotype harboring the wild type allele ( g . 1637C ) , form a functional protein and lead to high ficolin-3 concentration in this patient ( 21337 ng/mL ) , elevating the ficolin-3 mean level in the heterozygous g . 1637del/1637C patients . In conclusion , we identified high concentration of ficolin-3 in leprosy patients , associated with FCN3 polymorphisms present in introns 5 and 7 . We suggest that high ficolin-3 levels increase the susceptibility to leprosy playing an unfavorable role in these patients by favoring M . leprae dissemination .
|
Leprosy is considered a neglected disease and still a public health problem in many countries where it was not yet eliminated , leading to a high disability rate and social stigma . The molecular mechanisms of M . leprae infection and immune evasion are still poorly known , raising the need for studies that may contribute to a better understanding of leprosy etiology , as well as improvement in diagnosis and treatment . Ficolin-3 is a soluble molecule of the innate immune system that recognizes a wide range of pathogen-associated molecular patterns leading to complement activation and phagocytosis . We observed high concentration of ficolin-3 in leprosy patients , likely caused by polymorphisms present in intronic regions of FCN3 gene , which may contribute to leprosy susceptibility by favoring M . leprae infection . This is the first study addressing FCN3 polymorphisms and ficolin-3 levels in leprosy , indicating it as a good candidate biomarker associated with the host response against M . leprae .
|
[
"Abstract",
"Introduction",
"Material",
"and",
"methods",
"Results",
"Discussion"
] |
[
"mycobacterium",
"leprae",
"medicine",
"and",
"health",
"sciences",
"variant",
"genotypes",
"tropical",
"diseases",
"alleles",
"genetic",
"mapping",
"bacterial",
"diseases",
"neglected",
"tropical",
"diseases",
"molecular",
"biology",
"techniques",
"bacteria",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"genome",
"complexity",
"artificial",
"gene",
"amplification",
"and",
"extension",
"proteins",
"actinobacteria",
"molecular",
"biology",
"genetic",
"loci",
"biochemistry",
"haplotypes",
"lectins",
"polymerase",
"chain",
"reaction",
"heredity",
"genetics",
"leprosy",
"biology",
"and",
"life",
"sciences",
"genomics",
"computational",
"biology",
"introns",
"organisms"
] |
2017
|
Association of a new FCN3 haplotype with high ficolin-3 levels in leprosy
|
Aedes control interventions are considered the cornerstone of dengue control programmes , but there is scarce evidence on their effect on disease . We set-up a cluster randomized controlled trial in Santiago de Cuba to evaluate the entomological and epidemiological effectiveness of periodical intra- and peri-domiciliary residual insecticide ( deltamethrin ) treatment ( RIT ) and long lasting insecticide treated curtains ( ITC ) . Sixty three clusters ( around 250 households each ) were randomly allocated to two intervention ( RIT and ITC ) and one control arm . Routine Aedes control activities ( entomological surveillance , source reduction , selective adulticiding , health education ) were applied in the whole study area . The outcome measures were clinical dengue case incidence and immature Aedes infestation . Effectiveness of tools was evaluated using a generalized linear regression model with a negative binomial link function . Despite significant reduction in Aedes indices ( Rate Ratio ( RR ) 0 . 54 ( 95%CI 0 . 32–0 . 89 ) in the first month after RIT , the effect faded out over time and dengue incidence was not reduced . Overall , in this setting there was no protective effect of RIT or ITC over routine in the 17months intervention period , with for house index RR of 1 . 16 ( 95%CI 0 . 96–1 . 40 ) and 1 . 25 ( 95%CI 1 . 03–1 . 50 ) and for dengue incidence RR of 1 . 43 ( 95%CI 1 . 08–1 . 90 ) and 0 . 96 ( 95%CI 0 . 72–1 . 28 ) respectively . The monthly dengue incidence rate ( IR ) at cluster level was best explained by epidemic periods ( Incidence Rate Ratio ( IRR ) 5 . 50 ( 95%CI 4 . 14–7 . 31 ) ) , the IR in bordering houseblocks ( IRR 1 . 03 ( 95%CI 1 . 02–1 . 04 ) ) and the IR pre-intervention ( IRR 1 . 02 ( 95%CI 1 . 00–1 . 04 ) ) . Adding RIT to an intensive routine Aedes control programme has a transient effect on the already moderate low entomological infestation levels , while ITC did not have any effect . For both interventions , we didn’t evidence impact on disease incidence . Further studies are needed to evaluate impact in settings with high Aedes infestation and arbovirus case load .
Dengue is a growing problem worldwide , and it is currently present in more than 100 countries [1] . Recently , chikungunya and Zika , two other Aedes-borne diseases , have also been spreading geographically , and they represent a growing public health threat [2 , 3] . The recent co-circulation of these three Aedes-borne diseases in several parts of the world represents a new challenge for healthcare systems and requires the scientific world to devise effective and efficient control tools and strategies [4 , 5] . Dengue control programmes have existed for decades , and they are now being extended to the control of the above-mentioned viral diseases . Routinely employed dengue vector control tools and strategies are costly , require labour-intensive delivery , have poor long-term sustainability and are failing to control local arbovirus transmission and its geographical spread [6–8] . One example of a routinely applied measure to control the adult Aedes mosquito during an outbreak is ultralow-volume insecticide application . However , in reality , this has a rather limited effect because lethal amounts of insecticides do not reach most indoor mosquitoes , attributable to rarely followed standardized implementation procedures [9 , 10] and because mosquitoes are increasingly developing insecticide resistance [11] . While vector control interventions are currently considered the cornerstone of dengue control programmes , there is unfortunately scarce evidence on their effectiveness on dengue incidence [12] . Decision-makers of national control programmes have to formulate policies based on only a few published studies; however , owing to publication bias and non-standardized designs , these may provide inaccurate guidance . Reliable evidence on Aedes control tools and strategies is needed urgently in view of the epidemiology of current Aedes-borne diseases and the contraindication of the use of Dengvaxia , the first dengue vaccine , for children younger than 9 years [13] . The international dengue research community has stressed the need for developing , evaluating and implementing innovative , integrated and synergistic interventions that combine the best vector control tools with recently commercialized dengue vaccines [14 , 15] . The use of indoor insecticide residual spraying ( IRS ) is being advocated for Aedes control by the CDC [16] and PAHO [17] , although it is unclear what IRS entails for Aedes control . IRS is also used for controlling other disease-transmitting vectors , making it an interesting option for integrated vector management [18] . In a recent meta-analysis on dengue vector control [12] , only two observational studies evaluating this method were reported [19 , 20]; unfortunately , their results regarding the impact on entomological infestation levels were contradictory . More recently , in Peru , a pilot study in 36 houses showed that deltamethrin caused Aedes mortality greater than 80% on treated surfaces for up to 8 weeks after IRS application [21] , indicating its potential for sustained Aedes control . Therefore , there is a need for a controlled experimental field study , evaluating the health impact in terms of entomological and dengue incidence indicators under real-world conditions [22] , to ensure the relevance of the study results for public health . The entomological effect of insecticide treated curtains ( ITC ) has been evaluated in several contexts [23–27] , but not yet the effect on dengue incidence . As these vector control interventions are implemented in a delimited geographical area comprising several houses or house-blocks , the unit of intervention in such a trial need to be at cluster level . We report on a cluster randomized controlled trial in which the effect of indoor and peri-domestic residual insecticide treatment ( RIT ) is compared to insecticide-treated curtains ( ITC ) and a routine Aedes control programme in Santiago de Cuba during epidemic and inter-epidemic periods . The setting has seasonal fluctuating low entomological infestation levels ( with average House Index of 2% ) and low clinical dengue incidence , with increasing outbreaks in the last decade and reaching 20 clinical cases/100 000 inhabitants over the last few years . Such settings—where Aedes infestation and dengue circulation are intensifying or re-emerging or , on the contrary , declining from high levels as result of intensified control efforts—are bound to become more prevalent worldwide . Within this trial , we provide evidence from such a setting and evaluate the factors influencing its effectiveness by applying Wilson et al . ’s measure [22] for vector control tool studies .
The study was conducted in Santiago in southeast Cuba . Aedes proliferation is favoured by various factors , including the presence of an average of four water-holding containers of different types in each house [28] , high population density , uncontrolled urbanization , deficient solid and liquid waste management , high temperatures ( 28–34°C ) and rainfall from June to September . Despite an intensive routine Aedes control programme ( ACP ) , Aedes infestation persists with an average house index ( HI ) of 2% for Santiago; this value may be much higher at the house-block level , which may explain the sporadic outbreaks seen since 1997 [29–31] . The standard control activities conducted by the ACP teams include entomological surveillance and source reduction through periodic inspection of houses , larviciding ( with temephos ) of water-holding containers , selective adulticiding ( fogging with cipermethrine and clorpiriphus or perifocal residual spraying with deltamethrin ) when Aedes foci or dengue cases are detected , providing health education , promoting community-based environmental management and enforcing mosquito control legislation through the use of fines ( S1 ) . We set up a cluster randomized controlled trial between April 2011 and April 2013 . This study was designed to be implemented and followed-up over a 2-year period . However , owing to the major destruction of houses caused in the study area by Hurricane Sandy , this study was interrupted at the end of October 2012 . The twelve urban health areas in the Santiago municipality were included in the study . Vector control interventions are implemented at the level of geographical areas , therefore a cluster trial is indicated for the evaluation of effectiveness . Sixty-nine house-blocks were selected based on their elevated Aedes infestation in the previous five years . Each house-block and 3–4 of its surrounding house-blocks formed a cluster , resulting in a total of 250 houses . The 69 clusters were then ranked based on the Aedes infestation levels between January and December 2009 , and they were mapped . During mapping , six clusters with common boundaries were excluded to control for a potential spill-over effect [32] , and they were replaced by the next cluster in the ranked list . The remaining 63 clusters were randomly allocated by JPR and PVdS to the control , RIT or ITC intervention groups blocked by health area and by using the random number selection function in Excel . The primary outcome measure ( as described in the study protocol ) was HI , however , the intensive dengue transmission during the study period permitted to evaluate disease incidence measures and the public health impact of the intervention . The sample size ( number of clusters ) was determined using Hayes and Bennett’s calculation [33]; it had a power of 80% to detect a 50% reduction in the HI ( on average , 2% ) at an alpha error level of 0 . 05 ( assuming a between-cluster coefficient of variation of 0 . 5 ) . We increased the number of clusters by 5% , resulting in a final sample size of 21 clusters of 250 houses/study arm . The study showed 80% power to detect a 50% reduction in the dengue incidence of 10 cases/10 000 inhabitants . To prepare for implementing the trial , two workshops were held with the health area medical team , vector control team and formal and informal community leaders to explain the set-up and objective of the trial . Separate meetings were held with the population of the selected clusters , as described below . In all arms , routine control activities were continued . In the control arm , this was the only strategy . This study was approved by the ethical committee of the Institute of Tropical Medicine ‘Pedro Kourí’ , national health authorities , Institutional Review Board of the Institute of Tropical Medicine and University of Antwerp ( Belgian registration no . B300201111923 ) . During a meeting in each study cluster with community members after randomization , but before the start of the intervention , community approval was obtained . Written informed consent was obtained from the head of every household ( 18 years or older ) included in the ITC arm . Residual spraying , as described in the study setting section , is occasionally implemented as a routine activity by the control programme . Therefore , with approval of the ethics committees , the heads of households ( 18 years or older ) in the RIT arm signed at each application the ACP worker’s activity report form as proof of consent , instead of a separate form . The RIT and ITC used/applied insecticides approved by the WHOPES . The trial was registered at the Current Controlled Trials register ( no . ISRCTN27037293 ) . The RIT insecticide and ITC were purchased from the study budget and freely applied/distributed to the population . The study was conducted in accordance with the Helsinki Declaration of 1964 and subsequent revisions .
The 63 clusters ( 21 per study arm ) comprised 16 790 houses ( S1 Map ) . All clusters completed the study protocol from April 2011 up to the end of October 2012 , and all were included in the analysis ( Fig 1 ) . The pre-intervention entomological infestation levels and epidemiological characteristics were comparable in the three study arms ( Table 1 ) . The local pre-intervention ( 2010 ) strain of Aedes mosquitoes showed 100% mortality after 1 h of exposure to 0 . 05% deltamethrin ( 400 mosquitoes exposed ) . The RIT coverage over the five treatment cycles was , on average , 5 033 out of 5 180 premises ( 97 . 2%; range: 5 016–5 063 per round ) [42] . A total of 12 937 ITCs were distributed , with an average of 2 . 3 ITC/house . Of the 5 617 households in the ITC clusters , 94 . 4% chose to receive and hang the ITC in their house , and less than 5% removed them 6 months after distribution . No important harms or unintended effects were reported in the intervention arms . The bioavailability of deltamethrin was better when sprayed on cement walls than on unpainted wooden materials , and it remained high ( 85 . 0% ) for up to 3 months after application ( Fig 2 ) , on metal it declined quickly up to below 50% after 3 months . The bioavailability of the insecticide in the ITC after 6 months of use was good: the mortality of mosquitoes exposed to ITC that was never washed , washed one time and washed two times was 96 . 7% , 90 . 4% and 87 . 5% , respectively . The HPLC gave a range of 29–120 mg of active ingredient/m2 , which is above the minimum effective concentration of 25 mg/m2 [43] . Before the intervention , the HI ( March 2011 ) was 1 . 21% , 1 . 21% and 0 . 86% for the RIT , ITC and control arms , respectively ( Table 1 ) , and the dengue incidence rate ( January 2010 to March 2011 ) was 3 . 8 , 4 . 0 and 5 . 5/10 000 persons/month , respectively . Fig 3 shows that the dengue case incidence rate in the study arms was higher than in the Santiago municipality as a whole , as was expected owing to the selection of study sites with among the highest levels of entomological infestation in the city . After the interventions were started , the usual seasonal increase in dengue case incidence can be observed in all study arms over the entire intervention period . Both RIT and ITC did not show a significant protective effect on the entomological infestation level ( Fig 4 ) or dengue case incidence rate ( Table 2 ) ; on the contrary , slight negative effects were observed for ITC on HI ( rate ratio: 1 . 2 , 95% CI ( 1 . 0–1 . 5 ) ) and for RIT on dengue incidence ( rate ratio: 1 . 4 , 95% CI ( 1 . 1–1 . 9 ) ) . As RIT is known to have a temporary effect , we evaluated the effect of RIT by the number of months post-application over two of the four rounds ( Table 3 ) . In the first month after application , the PI reduced by 46% and the HI , by 37% . The effect faded out over time . No significant temporal effect was observed on the dengue incidence rate: average IR was 10 . 1 ( 95% CI 4 . 6–15 . 6 ) , 13 . 7 ( 95% CI 9 . 5–18 . 0 ) , 10 . 5 ( 95% CI 7 . 6–13 . 4 ) and 7 . 8 ( 95% CI 4 . 4–11 . 2 ) per 10 . 000 inhabitants , in the month of RIT application and one , two and three months after application respectively . When searching for the best-fitting model and adjusting for potential residual confounders , the negative crude effect of intervention disappeared . In the model ( Table 4 ) with the lowest AIC , the dengue case incidence rate by cluster and month was best explained by the epidemic period ( IRR = 5 . 50 ) , incidence rate in bordering house-blocks ( IRR = 1 . 03 ) and incidence rate before the intervention ( IRR = 1 . 02 ) ; the study arm was not withheld in this model as an independent determinant .
Residual insecticidal treatment significantly reduced the Stegomyia indices up to 1 month after application , but this effect was not sustained and , in our setting , it did not result in a reduction of clinical case incidence . Adding ITC to the routine Aedes control programme in Cuba had no impact on the incidence of clinical dengue cases , did not substantially reduce entomological infestation levels and did not mitigate seasonal dengue outbreaks . The design of this study is its major strength: randomized controlled trial testing with standardized interventions in 21 clusters of around 250 houses in each intervention arm ( sufficient power to detect a 50% reduction in HI and dengue incidence ) with a follow-up period of 17 months and the combination of entomological and epidemiological indicators to evaluate the intervention effects . According to a recent systematic review , the latter is hardly encountered in published studies evaluating the impact of dengue vector control interventions [12] . A relative weakness is that the outcome measures were extracted from routine surveillance data . This could have resulted in non-differential underreporting [44] and overall dilution of effects . Underreporting is expected to be limited to entomology data , as training for onsite inspection and for immature Aedes data collection were conducted for the vector campaign workers before the study; furthermore , the provincial quality control team reinforced the quality control system and intensified monitoring over the entire study period . For collecting data on dengue cases , the routine Cuban surveillance system combines a passive approach with active case finding from the moment the first dengue case is confirmed [31] . Underreporting of symptomatic cases is thus unlikely to be substantial . The secondary outcomes selected were Aedes immature indices because monitoring adult indices is faced with low reproducibility [45] . At any rate , for both proxies the relationship with transmission remains unclear [46] , hence we relied more on the evaluation of the impact on dengue disease incidence . On the other hand , the availability of information from the entire study clusters ( and beyond ) and with fine-grained monthly repeated measurements for entomology enabled the analysis to be adjusted for probable confounding factors and small-scale temporal variability . The internal validity of the study design seems secured . However , it is necessary to consider whether the observed low or short-lived entomological effect could be explained by other factors . The susceptibility of the local Aedes strain to deltamethrin was 100% at baseline . Until the start of this study , this insecticide was not used in the Santiago municipality , and no resistance would have developed . Resistance to deltamethrin can emerge after six months of intensive use , as seen in an outbreak in Brazil [47] . However , in our trial , deltamethrin was not used in the entire city . Instead , it was used in only two trial arms consisting of 42 clusters , and if the observed overall lack of effect is due to the development of resistance , we expect to observe an effect in at least the first 6 months of study that then wanes over time , which was not the case . To ensure optimal implementation , which is especially important for RIT , standard operating procedures were adhered to and quality control procedures were established . In order to produce a long-term preventive effect regularity of treatment cycles was respected in line with the residual efficacy reported by the manufacturer . The bioavailability of insecticide after RIT was similar to findings reported in a study in Peru [21] , with mosquito mortalities of 83% versus 76% on brick/cement and of 60% versus 64% on unpainted wood 3 months after application , in our and Peru study , respectively . The transient effect on entomological indicators after RIT application is in line with the results in the Peru study [21] , although , in contrast , we could demonstrate a significant effect on immature indices in the first month after application . The bioavailability of deltamethrin on ITC after 6 months was satisfactory , but slightly lower than observed in Thailand [48] . Both RIT and ITC implementation had excellent coverage and remained at much higher levels over time than was observed in other settings [49] . RIT was well accepted , as it did not differ much from the vector control measures routinely applied during outbreaks . The ITC , a new tool for the inhabitants of the Santiago municipality , was introduced after consultation and information rounds and with the active involvement of household members . As the clusters did not border each other , we do not expect a ‘no difference’ finding to be due to a spill-over effect , as reported in an Aedes control study using ITC [32] . To control for possible epidemiological pressure from surrounding areas on dengue incidence within the treatment clusters , we included the dengue incidence in house-blocks bordering the study clusters as a variable in the multivariable analysis . However , this did not lead to changes in our effect estimates . The dissociation between entomological and epidemiological impacts could be due to the settings , with already low infestation levels or not attaining the ( unknown ) level to which entomological infestation has to be ( sustainably ? ) reduced to stop or reduce transmission . Because dengue is transmitted by mosquito bites during the day , daytime human mobility is an important factor affecting dengue transmission dynamics [50] . Differential or non-differential mobility between study arms could have diluted the intervention effect . There are no arguments for the former . As for the latter , if local effects were indeed greatly diluted , then schools , workplaces , markets and mobility hubs outside the study clusters must also be covered in addition to the living and working places within the clusters . The lack of demonstration of an intervention effect could also be explained by the existence of an already intensive routine programme with the differential intensity of control actions over time and place in the function of the epidemiological situation and events . In addition to the inter-sector and community-based actions , the programme is heavily insecticide based with frequent indoor and outdoor fogging . In another Cuban municipality , ITC implementation also did not result in a reduction in entomological infestation [25] , whereas community-based environmental management without an increase in insecticide use above routine programme levels showed a clear effect [31 , 51] . This contrasts with the coverage-dependent ITC effect observed in Venezuela and Central Thailand [23 , 24] and the demonstration of a significant protective effect of window screens on Aedes infestation [52] in settings with higher infestation levels and less-intensive routine control programmes . In contrast to the findings of this trial , IRS , an approach similar to RIT used in this trial , was estimated to reduce dengue incidence by 64% ( OR: 0 . 36 , 95% CI ( 0 . 14–0 . 88 ) ) in an observational study [19] . However , Ko et al . [20] obtained similar results as in our study and could not demonstrate the effect of IRS on dengue incidence ( OR: 1 . 15 , 95% CI ( 0 . 63–2 . 10 ) ) . In this setting , multiple insecticides were used at the same place at the same time ( routine programme and study interventions ) , and this could have counteracting effects , as was already suggested when no benefit resulted from the combination of insecticide treated bed-nets with IRS or with a durable impregnated lining for malaria control [53 , 54] . Studies on the interactions between different insecticide-based vector control efforts are urgently needed to clarify the possible mechanisms involved . As a dengue vaccine is now available and is being rolled-out , but given its incomplete effectiveness , it is necessary to combine it with the best vector control options available to produce a synergistic effect on dengue incidence [55] . This contrasts to the control of urban yellow fever outbreaks by immunization [56] , because the vaccine for the same showed much higher efficacy [57] . Furthermore , many countries currently have to address the burden of three circulating arboviruses , namely , dengue , chikungunya , and Zika , and they cannot base their control efforts on the dengue vaccine alone . The lack of an observed effect of RIT and ITC on dengue incidence in the present cluster randomized controlled trial should perhaps not be very surprising . In a recent meta-analysis , none of the included randomized studies evaluating dengue adult vector control tools demonstrated a significant impact on dengue incidence , in contrast to other studies with less robust study designs that demonstrated an impact on dengue transmission with the intervention of house screening , IRS and environmental management [12] . In view of the high cost of RIT and ITC implementation [42 , 58] and the poor effect on dengue transmission in a low incidence setting shown in this study , the effectiveness should be evaluated in an array of contexts with higher case load and a different set-up of a routine control program . While RIT must be urgently evaluated further with robust study designs to provide more evidence on its potential effectiveness for reducing dengue transmission , a paradigm shift may also be needed . This includes evaluating/monitoring daytime human mobility in future studies [19 , 50] or covering premises where most household members work , go to school or spend a substantial part of their day in geographical units of intervention , which makes a cluster randomized trial difficult to design and implement . The complexity of the dengue transmission dynamics forces us to also think out of the box . Furthermore , for routine control programmes , instead of aiming to cover entire municipalities with undifferentiated vector control efforts or focusing on the reactive implementation of insecticide-based interventions in response to clinically apparent disease manifestations , it is necessary to shift to a new strategy by using risk stratification to concentrate proactive , sustained efforts in areas at high risk for transmission [59] . Additionally , several questions remain to be addressed concerning the timing of residual adulticide application for dengue control and outcome assessment: Would application be more effective if performed proactively when an increase in caseload or an epidemic of arboviral disease is forecasted ? Could it delay the start of an epidemic wave , so that time is given to prepare more efficient and comprehensive vector control measures and health system responses ? Because an effective routine control program ( with frequent indoor and outdoor fogging ) and a limited amount of dengue ( possibly due to this control program ) existed in the study area , it is possible that RIT and ITC added little if any additional benefit . Therefore , before making recommendations about RIT and/or ITC during epidemics and in endemic settings with more intense arbovirus transmission and/or with less intensive routine control measures in place , the impact of these control measures on dengue transmission in these settings should be demonstrated .
|
This is the first cluster randomized controlled trial providing evidence on the effect of applying residual insecticide treatment and deploying insecticide treated curtains on clinical dengue case incidence . In a context of intensive routine control activities , low Aedes infestation and moderate transmission , we did not observe an entomological nor epidemiological effect of these interventions . Besides the epidemiological context , we hypothesized two factors to explain our results: the counteracting effect of multiple insecticides applied in the same place at the same time and the importance of human mobility in dengue transmission dynamics . The specific impact of residual insecticide treatment and insecticide treated curtains on dengue transmission should be unambiguously demonstrated in settings with more intense arbovirus transmission and/or with less intensive routine control measures in place , before making recommendations for their implementation in local vector control programs in such contexts .
|
[
"Abstract",
"Introduction",
"Methodology",
"Results",
"Discussion"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"geographical",
"locations",
"randomized",
"controlled",
"trials",
"animals",
"north",
"america",
"clinical",
"medicine",
"pharmacology",
"infectious",
"disease",
"control",
"caribbean",
"insect",
"vectors",
"zoology",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"agrochemicals",
"epidemiology",
"disease",
"vectors",
"insects",
"agriculture",
"arthropoda",
"insecticides",
"people",
"and",
"places",
"cuba",
"mosquitoes",
"eukaryota",
"drug",
"research",
"and",
"development",
"entomology",
"clinical",
"trials",
"disease",
"surveillance",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"organisms"
] |
2017
|
The additional benefit of residual spraying and insecticide-treated curtains for dengue control over current best practice in Cuba: Evaluation of disease incidence in a cluster randomized trial in a low burden setting with intensive routine control
|
From gastrulation to late organogenesis animal development involves many genetic and bio-mechanical interactions between epithelial and mesenchymal tissues . Ectodermal organs , such as hairs , feathers and teeth are well studied examples of organs whose development is based on epithelial-mesenchymal interactions . These develop from a similar primordium through an epithelial folding and its interaction with the mesenchyme . Despite extensive knowledge on the molecular pathways involved , little is known about the role of bio-mechanical processes in the morphogenesis of these organs . We propose a simple computational model for the biomechanics of one such organ , the tooth , and contrast its predictions against cell-tracking experiments , mechanical relaxation experiments and the observed tooth shape changes over developmental time . We found that two biomechanical processes , differential tissue growth and differential cell adhesion , were enough , in the model , for the development of the 3D morphology of the early tooth germ . This was largely determined by the length and direction of growth of the cervical loops , lateral folds of the enamel epithelium . The formation of these cervical loops was found to require accelerated epithelial growth relative to other tissues and their direction of growth depended on specific differential adhesion between the three tooth tissues . These two processes and geometrical constraints in early tooth bud also explained the shape asymmetry between the lateral cervical loops and those forming in the anterior and posterior of the tooth . By performing mechanical perturbations ex vivo and in silico we inferred the distribution and direction of tensile stresses in the mesenchyme that restricted cervical loop lateral growth and forced them to grow downwards . Overall our study suggests detailed quantitative explanations for how bio-mechanical processes lead to specific morphological 3D changes over developmental time .
From gastrulation to late organogenesis , animal development involves many examples of reciprocal genetic and biomechanical interactions between epithelial and mesenchymal tissues [1] . Ectodermal organs , such as hairs , feathers , teeth and mammary glands are easily accessible examples of organs whose development is based on epithelial-mesenchymal interactions . Despite the diversity in their mature form and function , ectodermal organs share several common features during the early steps of development [2–4] . Organogenesis begins with the appearance of local epithelial thickenings , placodes , followed by condensation of the underlying mesenchymal cells . Later , the placode develops into a bud that grows into or out of the mesenchyme ensued by further growth and morphogenesis specific to each organ [2] . The genetic regulation of this latter process is relatively well understood [2 , 3] , whereas the contributing cellular and bio-mechanical mechanisms have remained largely unexplored . In this study we explore this latter question for tooth development as an example ectodermal organ . Teeth develop through a complex process which combines cell signalling , extensive cell movements and tissue deformation [5–8] . At the late bud stage ( Fig 1A ) , a signalling centre appears in the epithelium , the primary enamel knot , and a mesenchymal condensate forms beneath the epithelial bud . This is followed by the emergence of two epithelial folds , the buccal and lingual cervical loops , on the respective sides of the tooth germ . These initially grow laterally but progressively change their orientation , or angle of growth , downwards ( cap stage , Fig 1B ) . The angle of growth of the cervical loops is defined here as the angle between the tips of the cervical loops and the primary enamel knot in a frontal plane ( Fig 1C ) . The cervical loops largely delineate the boundaries of the tooth germ . The epithelium between the cervical loops is called the inner enamel epithelium and its growth and folding largely determine the overall shape of the tooth crown . The rest of the epithelial sheet , called the outer enamel epithelium , will not form part of the tooth crown . Similarly , the mesenchyme enclosed by the cervical loops , called the dental mesenchyme , will become part of the tooth crown , whereas the mesenchyme in contact with the outer enamel epithelium , called the follicular mesenchyme , will not . The tissue located on the apical side of the enamel epithelium is called the suprabasal layer and it is already present at the earliest stages of tooth development ( Fig 1A ) . At the late cap stage , secondary enamel knots form in the inner enamel epithelium and a cusp will arise under each one of them [9] . The tooth germ progressively elongates in the anterior-posterior axis and cervical loops form in the anterior and the posterior of the tooth germ as a continuation of the buccal and lingual cervical loops . The angle of growth of the cervical loops has a general effect on the sharpness of the tooth , with smaller angles leading to sharper teeth [10] , and in some species on the positioning of secondary enamel knots [11] . At later stages , enamel secreting ameloblasts and dentin secreting odontoblasts differentiate at the interface between the inner enamel epithelium and mesenchyme to form the mineralized tooth . In this study we build a new mathematical model of early tooth development . Our aim is to understand how the tooth germ changes in shape from the bud to the cap stage . This implies understanding how the cervical loops form , what determines the orientation of their growth in the different parts of the tooth germ and how that affects the shape of the tooth germ . Theoretical studies in development and tissue mechanics predict that epithelial folding can be the result of differential growth between adjacent tissues [12–14] . For example , differential growth between the epithelium and the suprabasal layer modelled in 2D has been used to produce buckling reminiscent of the bud to cap stage transition [14] . Although not considering biomechanical properties such as differential adhesion , the previous studies underscore the requirement of differential growth in the progression of epithelial development beyond the bud stage . Here we hypothesize that the growth and orientation of the cervical loops depends on both the differential growth and differential adhesion between the epithelium , the suprabasal layer , and the mesenchyme . Variation in differential growth and adhesion should then lead to variation in the orientation of the cervical loops . We implement this hypothesis in a new 3D cell-based model of tooth morphogenesis using the recent EmbryoMaker modelling framework of animal development [15] ( see S1 Appendix ) . EmbryoMaker is essentially a modelling tool that uses a mathematical implementation of the basic cellular and molecular processes known in animal development . We use this modelling framework to build a tooth specific model by specifying to EmbryoMaker the distribution of cell types at the bud stage , gene expression dynamics and signalling , and the cell behaviours involved in early tooth development . This study , however , does not restrict itself to study the wild-type mouse molar . The aim is also to understand how the length and orientation of the cervical loops in 3D changes with differential growth and adhesion . There are several previous mathematical models of tooth development [16 , 17] . These models implement only the inner enamel epithelium and cannot be used to model the cervical loops and the dynamics that define the boundary between inner and outer enamel epithelium . Furthermore , these earlier models do not explicitly consider the three main tooth germ tissues , epithelium , mesenchyme , and suprabasal layer and , thus , cannot study , in an integrated way , how their differential adhesion and growth affect early tooth development . In addition , EmbryoMaker implements much more refined cell and tissue biomechanics than previous models , thereby allowing us to decompose the different roles of tissue growth and adhesion dynamics .
In order to assess which hypotheses were capable of reaching cap morphology , we performed a parameter screening on the tissue-specific growth parameters . We ran the tooth-specific model under different combinations of tissue growth rates ( sepi , ssup and smes parameters ) and under the three different hypotheses ( Fig 3 , S3 Fig ) . For all hypotheses , we found that the cervical loops appeared at the buccal and lingual sides of the tooth germ when the proliferation rate in the epithelium was high relative to the suprabasal and mesenchymal proliferation rates ( Fig 3A , 3B and 3E ) . The length of these loops increased with the epithelial to suprabasal proliferation ratio ( S3A–S3C Fig ) . High suprabasal growth rates relative to the epithelium growth rates led to enlarged buds lacking cervical loops ( Fig 3C and 3F ) . However , only in hypotheses II and III ( Fig 3B and 3E ) did the cervical loops grow downwards similarly to the wild-type ( small growth angle ) . In hypothesis I , the cervical loops grew with an angle close to 180° ( Fig 3A ) , and the follicular mesenchyme was roughly as thick as the dental mesenchyme , unlike wild-type tooth germs ( Fig 1B ) . Thus , we rejected hypothesis I and concluded that the formation of the cervical loops requires the proliferation rate of the follicular mesenchyme to be lower than that of the dental mesenchyme . In hypothesis II , low mesenchymal growth ( smes<0 . 20 ) led to an abnormal shape of the dental mesenchyme , with additional epithelial folds between the cervical loops ( Fig 3D ) . In hypothesis III , the height of the dental mesenchyme ( the difference in height between the tip of the cervical loops and the base of the signalling centre ) increased with the proliferation rate of the mesenchyme ( Fig 3G ) . Thus , in hypotheses II and III , a sufficiently high proliferation of the dental mesenchyme is necessary for normal cap morphology to arise . Next we wanted to test whether the model was able to create a realistic cap morphology with tissue growth rates estimated from experimental observations . We measured the length of the dental epithelium and the surface area of the suprabasal layer from the time-lapse sequence of a developing mouse molar recently published by Morita and collaborators [6] at different time points , thus creating a growth curve for both the epithelium and the suprabasal layer . The time-lapse in Morita et al . was performed on a molar thick frontal section , thus the data was essentially 2D . In order to better compare that data to our model , we created a 2D version of the tooth model ( Fig 4A and 4B ) , based on the same principles as the original model , and performed a large parameter screening of tissue growth and adhesion parameters . For each simulation we calculated the growth curves of the epithelium and the suprabasal layer , and then compared them to the empirical ones by calculating the standard error between the theoretical and empirical curves ( S4 Fig , see S1 Appendix ) . We performed a sensitivity analysis of the standard error measurement against each of the parameters chosen for the screening ( S5 Fig ) . In both hypothesis II and III , the model showed the largest sensitivity on the epithelial growth parameter ( sepi ) ( S5A and S5E Fig ) , whereas it showed the lowest sensitivity to the mesenchymal growth parameter ( smes ) and the five adhesion parameters ( S5C , S5D , S5G and S5H Fig ) . In both hypotheses II and III , a relatively large number of model runs ( approximately 500 in each case ) had a good fit ( standard error < 10 ) with the empirical growth curves . Visual inspection of the morphology in these better fitting subsets showed that the great majority of model runs achieved a proper cap morphology in the 2D model ( e . g . Fig 4C–4F ) . Even if the model predicted correctly the overall morphology of the tooth germ , it could be that this was accomplished by different patterns of cell movement compared to the real tooth . We proceeded to compare cell movement in the 2D model ( see S1 Appendix ) with the empirical cell trajectories during tooth development [6] . In their study , Morita et al . recorded multiple cell trajectories on the epithelium and suprabasal layer during their time-lapse ( Fig 5A and 5D , see also [6] ) . In our model , suprabasal cells showed a consistent movement towards the tips of the cervical loops in both hypotheses II and III ( Fig 5B and 5C ) . We then decided to focus on the behaviour of epithelial cells . In Morita’s experiments , they tracked a small group of epithelial cells at the tip of both cervical loops , and they observed that by the end of the tracking period they still remained at the tip , even though the tissue had grown significantly ( Fig 5D ) . When looking at the movement of epithelial cells in our simulations , differences were observed between hypothesis II and III . In hypothesis II , cells that were at the tip of the cervical loops at the start of the tracking period had moved outwards by the end of the tracking , and cells that were medially located at the start moved to the tip of the cervical loops ( Fig 5E ) . Lateral movement of epithelial cells in II was caused by the fact that cells only proliferated close to the enamel knot and slid sideways as the epithelium grew ( Fig 5E ) . In hypothesis III , cells that started at the tip of the cervical loops remained in them by the end of the tracking period ( Fig 5F ) . Thus , it seems unlikely that epithelial cells proliferate only near to the enamel knot , since the cell movements observed ex vivo should be different than the ones presented by Morita and collaborators . Thus , we decided to reject hypothesis II and , for the rest of the study we only show the computational analyses corresponding to hypothesis III . Note , however , that the analyses described in the following sections were also performed for hypothesis II and led to the same conclusions than III , but we chose to exclude them for the sake of simplicity . We performed a parameter screening in the 3D model for the five adhesion parameters: epithelium homotypic ( bEE ) , epithelium-suprabasal ( bES ) , epithelium-mesenchyme ( bEM ) , suprabasal homotypic ( bSS ) , and mesenchyme homotypic adhesion ( bMM ) . The resulting morphologies displayed differences in the length and orientation of the cervical loops ( Fig 6 , S6 Fig ) . By looking at the frontal sections , we observed that high mesenchyme homotypic adhesion led to short cervical loops oriented downwards ( small growth angle ) , whereas low values led to long cervical loops oriented bucco-lingually ( large growth angle ) ( Fig 6A ) . The latter effect was more marked when the suprabasal homotypic adhesion was high ( Fig 6A , top row ) . The cervical loops were also oriented downwards when the epithelium-mesenchyme adhesion was high , especially when the mesenchyme homotypic adhesion was also high ( Fig 6B , top row ) . This seemed to occur because the epithelium deformed to maximize its contact area with the dental mesenchyme , even when the mesenchymal homotypic adhesion was null ( e . g . Fig 6B left column ) . In contrast , with null epithelial-mesenchymal adhesion , the cervical loops failed to surround the dental mesenchyme , and the growth angle was high ( Fig 6B bottom row ) . In order to understand how differential adhesion affects the shaping of the tooth germ , we plotted the spatial patterns of mechanical stress at the level of cell-cell contacts . We observed that there was a strong association between the orientation of the cervical loops and the spatial patterns of stress in the tooth germ . In simulations where high mesenchymal homotypic adhesion led to downwards oriented cervical loops , there was tension along the surface of the follicular mesenchyme surrounding the tooth germ ( Fig 7A and 7B , left column ) . When that adhesion was null and the cervical loops grew in the bucco-lingual direction , there was no such tension in the mesenchyme ( Fig 7C , left column ) . We also observed tension in the suprabasal layer in simulations where high suprabasal homotypic adhesion led to bucco-lingually oriented cervical loops ( Fig 7A and 7C , right column ) . We did not observe high tension or compression at the interface between the epithelium and the suprabasal layer , nor between the epithelium and the mesenchyme when the epithelial-suprabasal and epithelial-mesenchymal heterotypic adhesions were high ( S7 Fig ) . Mouse molars are longer in the antero-posterior axis than in the bucco-lingual axis . This has been associated with cervical loops in the buccal and lingual sides forming earlier and growing more downwards than the cervical loops in the anterior and posterior sides [7 , 9] . In the model , we observed that the cervical loops form in the same fashion in most cases ( Fig 8A–8D ) , that is , also with an anterior-posterior versus bucco-lingual asymmetry . In other words , the overall shape of the epithelium in 3D is similar between mouse and model . For this to happen , however , the tooth germ in the initial condition needs to be at least slightly longer in anterior-posterior axis than wider in the bucco-lingual axis ( as it is the case in mouse tooth germs from very early on , [7 , 9] ) . If we choose initial conditions with radial symmetry , cervical loops grow equally in all directions , a mode of development that would produce a single fang or canine shaped tooth ( Fig 8E–8H ) . Results from the model like those shown in Fig 6B suggest that strong adhesion between epithelium and mesenchyme results in downward oriented cervical loops . A prediction arising from the adhesion dynamics is that an experimental separation of the epithelium and the mesenchyme should lead to a fast increase of the cervical loop growth angle . To test the model prediction , we performed an enzymatic separation of epithelium and mesenchyme on dissected E14 . 5 mouse first molars sliced in thick frontal sections ( roughly 400 μm ) . Molar sections were submerged in dispase laying flat on the bottom of a glass dish so that one of the sliced surfaces was facing upwards . Dispase digests extracellular matrix , including the basement membrane and , given enough time , the epithelium and mesenchyme will separate from one another ( S1 Video , also see [6] ) . The results show that the follicular mesenchyme on one side of the tooth germ peeled off from the epithelium , and eventually the whole mesenchyme recoiled towards the other side of the tooth germ ( n = 6 ) . At the same time , the cervical loops detached from the mesenchyme and rapidly changed their orientation , with the growth angle increasing to reach roughly 180° ( Fig 9A ) . This reorientation of the cervical loops was unlikely to be due to growth since it occurred in less than 1 hour . No recoil in the mesenchyme was observed when the same experiment was performed on E13 . 5 tooth germs ( bud stage , S8 Fig ) . We then proceeded to reproduce the separation assay in silico . Since the tissue deformations essentially took place within a two-dimensional plane , we used the 2D version of the model . We first simulated tooth development until cap stage ( Fig 9B ) , then we set the epithelial-mesenchymal adhesion to 0 . We resumed the simulation until the system was at equilibrium again ( i . e . no cell movement ) ( Fig 9C ) . We repeated the same procedure with different combinations of adhesion parameters ( S9 Fig ) . We observed in all cases that the mesenchyme buccal and lingual to the tooth germ retracted towards the mid line of the tooth germ and in most cases the growth angle increased ( Fig 9B and 9C , S9 Fig ) , in accordance with the experimental observations ( Fig 9A ) . In order to quantify the model and experimental deformations , we tracked tissue deformation over time using specific morphological landmarks in the real tooth germs and in the model ( Fig 9A–9C ) , and performed a principal components analysis on shape data extracted from these biological landmarks ( real tooth germs n = 3 , model tooth germs n = 1 , Fig 9D , see Methods ) . The first principal component ( PC ) explained 91 . 3% of the total variation as an increase of the growth angle of the cervical loops ( Fig 9D ) . Furthermore , time evolution both ex vivo and in silico consistently correlated with an increase in PC1 ( Fig 9D , arrows ) . By plotting the mechanical stresses over the tooth germ during the separation simulations we were able to obtain insights about the origins of the observed tissue deformations . The follicular mesenchyme retraction coincided in time , in the model , with a decrease in the tension between its cells suggesting that this tension may be also responsible for the recoil in the follicular mesenchyme in the ex vivo experiments ( Fig 9B and 9C , S10 Fig ) . The reorientation of the epithelium and suprabasal layers coincided with an increase in tension in the suprabasal cells and a relaxation of the compression of the epithelial cells ( Fig 9B and 9C , S10 Fig ) . This reorientation coincided with the arising in the suprabasal layer of an arch of tension along the bucco-lingual axis ( Fig 9C , S10 Fig ) . In other words , a large part of the tension is distributed along a line of contiguous cells going from the buccal to lingual side of the suprabasal layer . By following the evolution of the mechanical stresses during the in silico separation , it can be seen that this arch was a result of the expansion of the epithelium that , by being mechanically attached to the suprabasal layer , pulled the suprabasal layer along the length of the cervical loops . The separation of the epithelium and the mesenchyme allows the suprabasal layer to relax the tension in this arch . As a result , the curvature and tension in this arch decrease and the cervical loops reorient in the bucco-lingual direction ( S10 Fig ) .
It has been argued that the compression of an epithelium ( along its plane ) will lead to its buckling ( out of plane folding ) ( reviewed in [13] ) . The formation of small folds , or villi , in the small intestine , for example , has been proposed to occur by this mechanism [12] . In our model , the expansion of the growing epithelium is restricted by both the mesenchyme and the suprabasal layer . Suprabasal growth tends to push the epithelium outwards against the mesenchyme , but excessive growth will lead to a globular shape , reminiscent of an enlarged tooth bud ( e . g . S11C Fig , left most column ) . On the contrary , as long as the suprabasal layer grows relatively slower than the epithelium , the latter will need to fold in order to keep contact with the former due to their heterotypic adhesion ( e . g . S11C Fig , left most column ) . In other words , a buckled ( cap ) shape requires a higher surface ( i . e . epithelium ) to volume ( i . e . suprabasal layer ) ratio than a globular ( bud ) shape . The mesenchyme acts as a barrier to the outwards expansion of the epithelium , both due to the accumulated tension on the external layers and the pressure of the growing mesenchyme underneath the enamel knot . Even when high suprabasal growth is forcing the tooth germ into a globular shape , high mesenchymal growth can contribute to the epithelial folding by pushing the epithelium from underneath ( e . g . S11C Fig , top row ) . For the same reason , high mesenchymal growth prevents epithelial buckling in the central part of the tooth germ and forces the epithelium to fold around the growing mesenchyme ( Fig 3D ) , a process which also depends on the epithelial-mesenchymal heterotypic adhesion . Cervical loops , therefore , form only on the sides ( buccal , lingual , anterior and posterior ) of the dental mesenchyme . Our model predicts that three different factors regulate the orientation of the cervical loops . This asymmetry can be understood from the geometry of the initial conditions ( see S1 Appendix for more details ) . The early tooth bud in vivo and in the initial conditions of the model are both longer in the antero-posterior ( AP ) axis than in the bucco-lingual ( BL ) axis . That asymmetry was not observed when we simulated teeth whose initial conditions had radial symmetry ( i . e . initial conditions are as long in the AP than in the BL axis ) ( Fig 8E–8H ) . Seen from below ( i . e . from the dental mesenchyme ) , the initial epithelium looks like an ellipsoid and , as such , it has a higher curvature in the AP sides than in the BL sides ( Fig 8A–8D ) , i . e . a higher ratio between surface and enclosed volume ( S12A Fig ) . Because of that , the same amount of growth will lead to less elongation of the cervical loops in the AP than in the BL sides . We have devised a simplified geometrical argument showing that the elongation rate of the BL loops would follow an exponential growth function with exponent proportional to t ( time ) , whereas the in the AP loops the exponent would be proportional to t/2 ( S12B Fig , see details of the argument in S1 Appendix ) . The fact that this asymmetry was not so apparent when the mesenchymal homotypic adhesion or the epithelial-mesenchymal adhesion were very low is simply due to the fact that in these cases the cervical loops did not grow downwards anyway . Our model , thus , makes apparent that even when the tissues grow at the same rate everywhere , the AP vs . BL asymmetry of the initial conditions would inevitably lead to deeper lateral cervical loops than anterior and posterior loops . The large recoil observed in the follicular mesenchyme after the experimental separation suggests that there is a line of tension running from buccal to lingual side of the tooth germ . A similar deformation was observed in the follicular mesenchyme in the in silico separation , although with less intensity . The most visible deformation in the epithelium during the separation experiment is the reorientation of the cervical loops towards the bucco-lingual axis . The same kind of reorientation is observed in the in silico experiment . In the model , the reorientation of the cervical loops is due to the combination of two processes . One is the expansion of the epithelium that was under compression before the separation , and the other is the relaxation of the tension in the suprabasal layer . This latter tension is the result of the pull on the suprabasal layer by the expanding epithelium ( as explained above ) . This latter tension is roughly distributed as an arch that will tend to flatten after the separation , thus reorienting the cervical loops in the bucco-lingual direction ( S10 Fig ) . The combination of compression in the epithelium and tension in the suprabasal layer may also account for the epithelial deformation observed in the ex vivo separation . Our experiments , however , were not able to discern between stresses in the epithelium and the suprabasal layer . In a recent study [22] , a different kind of mechanical perturbation was performed at the placode stage of tooth ( E12 . 5 ) , i . e . before the tooth bud forms . Using thick frontal slices of tooth germs , Panouspopoulou and Green performed a cut in the oral epithelium at one side of the placode and observed that the epithelium by the side of the placode recoiled towards the mid line . They also interpreted this recoil as a consequence of a bucco-lingually oriented tension , in this case located in the suprabasal layer of the tooth placode . Along these lines we have observed that , when the separation experiment is performed on bud stage tooth germs ( E13 . 5 ) the mesenchyme did not recoil ( S8 Fig ) , indicating that the tension in the mesenchyme may build up between the bud and cap stages , as our model predicts ( S13 Fig ) . In this study we have explored , theoretically and experimentally , the role differential growth and adhesion on the transition from the bud stage , common to several ectodermal organs , to the specific cap shape of the early tooth germ . Our model simulates the formation of the tooth germ by combining the aforementioned processes , and accounting for cell and tissue mechanical interactions that result in tooth specific shape transformations . Even though our model accounts only for early tooth morphogenesis , it does so by implementing a set of cell and mechanical processes common to the ensemble of ectodermal organs . Thus , the dynamics produced by this model leading to early tooth-specific shapes ( or failing to do so ) , may shed light on how other ectodermal organs undergo their specific transformations after bud stage . For instance , some of the model dynamics may also apply to the formation of epithelial folds surrounding a mesenchymal condensate in the hair follicle after its bud stage [23] . In a homologous fashion , the formation of these folds may be a result of increased epithelial growth relative to suprabasal growth , whereas a high epithelial-mesenchymal adhesion may account for the surrounding of the mesenchyme by these folds . In another type of ectodermal organs , such as mammary , salivary glands and lungs , the epithelium folds in order to form branched structures , but never surrounds the mesenchyme [24] . Ex vivo and in silico studies of mouse lung epithelium suggest that mechanical buckling of the epithelium due to intrinsic growth plus mechanical interactions with the surrounding extracellular matrix is sufficient to account for branch formation [25] . In addition , it has been shown that the lung mesenchyme also contributes to epithelial folding by mechanically constricting the lung epithelium during branch formation [26] . Future studies should address whether differential tissue growth and mechanical interactions between epithelium and mesenchyme observed in the development of different ectodermal organs are regulated by a conserved developmental mechanism . Our model does not consider active cell migration or cell contraction ( although EmbryoMaker can implement these cell processes ) . Active cell migration over an extracellular matrix substrate and cell intercalation have been shown to generate tissue-scale mechanical stresses [27 , 28] . It has been shown that mesenchymal cells at the early bud stage actively migrate towards the source of an FGF8 gradient in the bud epithelium [29] . It has also been shown at the tooth placode stage that perturbation of the Shh pathway altered the width and depth of the placode , suggesting the presence of cell intercalation in the suprabasal layer [22] . Morita and collaborators also argued that active migration mediated by high F-actin turnover and the LIMK-cofilin pathway is present in the growing regions of the tooth germ epithelium [6] . Even though we acknowledge that active cell migration and cell intercalation may have a role in tooth development and , perhaps , would improve our model if included , we show that several features of tooth development related to tissue growth , cell movement , and tissue mechanics can already be explained by considering only passive cell movement resulting from cell adhesion and proliferation . Our new mathematical model of tooth development provides detailed quantitative explanations on how biomechanical processes may drive tooth germ morphology to change the specific way it does in 3D space and over developmental time . This included explanations on how morphology will change in specific ways when these biomechanical processes are altered and , thus , understanding not only the wild-type but also its variational properties . To our knowledge no such explanations have been provided for any ectodermal organ , although they are well studied in other aspects of their development . In spite of the increase in complexity of tooth germ morphology during development , two biomechanical processes seem enough to explain it . This result highlights how the combination of experimental results with computational models of biomechanical processes can help providing relatively simple explanations for seemingly complex processes such as the development of morphology and its variation .
All animal work was conducted accordingly to the guidelines required by the Finnish authorities ( ESAVI-2984-04 . 10 . 07–2014 , KEK13-020 ) . Mouse specimens were sacrificed by anaesthetising with CO2 first followed by cervical dislocation . The following section describes the basics of EmbryoMaker and how we build the tooth-specific model based on it ( see [15] for a more extensive description ) . Mesenchymal and suprabasal cells are made of spherical bodies , that we call nodes , whereas epithelial cells are made of cylindrical bodies consisting of two nodes ( one basal and one apical bound by an elastic link ) ( S1A and S1B Fig ) . The movement of nodes follows an overdamped Langevin equation of motion , ∂r→i∂t=∑j=1j=nvfAiju^ij ( 1 ) where ri is the position vector of node i , nv is the number of nodes in the neighborhood of node i , t is time , fAij is the modulus of the force acting between node i and j and uij is the unit vector connecting i and j ( see S1A–S1D Fig ) . The modulus and sign of the force is dependent on the distance between the two nodes , {fAij= ( piREC+pjREC ) ( dij− ( piEQD+pjEQD ) ) ifdij< ( piEQD+pjEQD ) fAij=kijADH ( dij− ( piEQD+pjEQD ) ) if ( piEQD+pjEQD ) ≤dij≤ ( piADD+pjADD ) fAij=0if ( piADD+pjADD ) >dij ( 2 ) When the distance between nodes i and j ( dij ) is shorter than the sum of their radii at equilibrium ( pEQD ) , there is a repulsive force proportional to the sum of the pREC of each node ( this coefficient determines their incompressibility ) . When this distance is longer than the equilibrium distance but shorter than the sum of the maximum radii of i and j ( pADD ) , there is an attractive force between nodes i and j . This force is proportional to kijADH , kijADH=gimgjnbmn ( 3 ) where gim is the amount of adhesion molecule m expressed in node i and bmn is the adhesive affinity between adhesion molecules m and n . The direction of force vectors differ between mesenchymal-mesenchymal , epithelial-epithelial and the epithelial-mesenchymal node interactions , since vectors need to be normal to the contact interface between nodes and nodes have different shapes in epithelial cells and mesenchymal cells ( see [15] for a detailed explanation ) . The apical and basal nodes of epithelial cells are connected by an elastic spring that opposes any departure from an equilibrium distance between the apical and basal nodes of each cylinder [15] . The force generated by the spring is calculated as follows , f→Sij=kijHOO ( dij−pijEQS ) s^ik ( 4 ) where kijHOO = piHOO + pjHOO is the elastic coefficient of the spring ( which is determined by the sum of the mechanical parameter pHOO in both nodes ) , dij is the distance between node i and j , pijEQS is the equilibrium length of the spring between node i and j and sik is the unit vector connecting the two epithelial nodes . Two additional force components are required in epithelial cells in order for them to organise as one layered sheets [15] . A radial force acts along the apical-basal axis of the cell and tends to restore displacements in that axis in respect to neighbouring cells in the epithelium , whereas a rotational force acts tangential to surface of the epithelium and tends to orient the apical-basal axis of cells normal to the epithelial plane . These forces are calculated as follows , fESTij→=kijESTmijkl→·cij→|mijkl→|mijkl^ ( 5 ) fERPij→=piERPsik→·cij→|sik→|cij^ ( 6 ) where fijEST is the radial bending force and fijERP is the rotational bending force . We define cij as the vector connecting neighboring node i and j , sik and sjl as the vectors that connect each apical node to their basal counterparts and mijkl as the sum of sik and sjl which defines the vector normal to the apical or basal surface between i and j . The radial bending force always acts on the direction of mijkl , and is proportional to the deviation of the angle formed by mijkl and cij from 90° . kijEST is the sum of the mechanical parameter pEST of nodes i and j ( see [15] for a detailed explanation ) . The rotational bending force is proportional to the deviation of the angle formed by sik and cij from 90° , but in this case the direction of the force is parallel to cij , thus promoting a tilting of the epithelial cylinder that reaches an equilibrium ( that is the force modulus becomes 0 ) when the apical-basal axis of the epithelial cylinder is normal to the apical/basal cell surface . kijERP is the sum of the mechanical parameter pERP of nodes i and j ( see [15] for a detailed explanation ) . In summary , thus , the forces acting on an epithelial node are: ∂r→i∂t=f→Sik+∑j=1nd ( fAiju^ij+f→ESTij+f→ERPij ) ( 7 ) where k is the node in the same cylinder than i and the sum is made over all the neighboring nodes except for k . Cell division is implemented by placing a new cell in a random position close to the old one . EmbryoMaker implements many other cell behaviors ( such as apoptosis , cell contraction , extracellular secretion , cell growth , cell polarization ) as rules acting on these nodes but we do not explain them because they are not used in the tooth-specific model . Different gene products can be specified as transcription factors , adhesion molecules , extracellular diffusible signals , or receptors ( see S1 Appendix ) . They can also be specified to regulate different cell behaviours , such as cell division . Each cell has a variable , PPHA , indicating the progression of the cell cycle ( from 0 to 1 ) . When PPHA reaches 1 , the cell divides . Molar germs were dissected from NMRI mice and sliced into thick frontal slices using fine tungsten needles . Molar sections were submerged on a glass Petri dish containing a dispase II solution , laying flat on the bottom of the dish . A small amount of DNAseI was added to the enzymatic solution in order to reduce the viscosity of the medium and facilitate tissue separation . A time-lapse sequence of the tooth sections in the solution was recorded for about 1 hour and a half in a Olympus SZX9 microscope . The tissues usually started to separate from each other about 25 minutes after being submerged in the solution , and it usually took about 40 minutes for the whole deformation to take place . For each tooth germ ( n = 3 ) and the model ( n = 1 ) , 120 pictures that correspond to every time frame from the total duration of the separation experiment were taken ( total = 480 pictures ) at the same magnification . For each picture , a suite of 4 landmarks was collected in two dimensions using the freely available software tpsDig version 2 . 17 [30] to precisely characterize the architecture of tooth germs ( Fig 9A ) and was digitized twice to assess measurement error due to digitizing ( total = 960 landmark configurations ) . Landmarks were located at the maximum of curvature at the junction between outer enamel epithelium and oral epithelium ( landmarks 1 and 4 ) and at the maximum of curvature of the cervical loops ( landmarks 2 and 3 ) . From these 960 landmark configurations , shape was extracted by removing extraneous information of size , position and orientation via a Generalized Procrustes Analysis ( GPA; e . g . [31] ) . A principal components analysis ( PCA ) was performed on these shape data to detect and localize very fine morphological variation in relation to the geometry of the tooth germs ( Fig 9D ) . A one-way Procrustes analysis of variance ( ANOVA ) and a one-way ANOVA was used with centroid size ( i . e . , square root of the sum of the squared distances of all landmarks from their centroid ( e . g . [32] ) and decompose the total shape variation according to the main effect of ‘tooth germ’ ( i . e . variation among tooth germs ) and measurement error due to digitising in shape and size ( e . g . [33] ) . The ANOVAs for both size and shape reveal that the main effect of ‘tooth germ’ is highly significant ( P < 0 . 0001 ) compared to digitizing error , meaning that digitizing error in placing landmarks on tooth germs is highly negligible ( S1 Table ) . Consequently , averages of both digitizing sessions were calculated to obtain a single configuration of landmarks per picture and these 480 mean shapes were used in all subsequent analyses . All analyses were carried out using the freely available software MorphoJ [34] .
|
The genes and signalling pathways involved in ectodermal organ development ( teeth , hair , mammary glands , etc . ) are relatively well studied . However , the bio-mechanical processes by which these organs grow and change shape from an early primordium ( an epithelial bud similar among different organs ) is far less understood , especially in mammalian development . This study combines simple experiments and a multi-scale , cell-based computational model to understand these processes for one ectodermal organ: the tooth . Our model implements the different mechanical properties of the different cell types involved in morphogenesis ( i . e . epithelial and mesencymal ) and their interactions . By exploring model behaviour and contrasting it with experimental data on tissue growth rates and mechanics we found that , in spite of their relative complexity , the shape changes occurring in early tooth development , the overall sharpness of the tooth and its variation can largely be explained by the two most simple bio-mechanical processes: differential growth and differential adhesion between tooth tissues . These results suggest that simple mechanical interactions between cells and tissues may underlie the complex tissue deformations observed in the morphogenesis of several ectodermal organs .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"classical",
"mechanics",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"mechanical",
"stress",
"epithelial",
"cells",
"damage",
"mechanics",
"digestive",
"system",
"dentition",
"animal",
"cells",
"deformation",
"biological",
"tissue",
"head",
"physics",
"teeth",
"anatomy",
"cell",
"biology",
"physiology",
"jaw",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"digestive",
"physiology",
"molars"
] |
2018
|
Differential tissue growth and cell adhesion alone drive early tooth morphogenesis: An ex vivo and in silico study
|
Gene duplications within the conserved Hox cluster are rare in animal evolution , but in Lepidoptera an array of divergent Hox-related genes ( Shx genes ) has been reported between pb and zen . Here , we use genome sequencing of five lepidopteran species ( Polygonia c-album , Pararge aegeria , Callimorpha dominula , Cameraria ohridella , Hepialus sylvina ) plus a caddisfly outgroup ( Glyphotaelius pellucidus ) to trace the evolution of the lepidopteran Shx genes . We demonstrate that Shx genes originated by tandem duplication of zen early in the evolution of large clade Ditrysia; Shx are not found in a caddisfly and a member of the basally diverging Hepialidae ( swift moths ) . Four distinct Shx genes were generated early in ditrysian evolution , and were stably retained in all descendent Lepidoptera except the silkmoth which has additional duplications . Despite extensive sequence divergence , molecular modelling indicates that all four Shx genes have the potential to encode stable homeodomains . The four Shx genes have distinct spatiotemporal expression patterns in early development of the Speckled Wood butterfly ( Pararge aegeria ) , with ShxC demarcating the future sites of extraembryonic tissue formation via strikingly localised maternal RNA in the oocyte . All four genes are also expressed in presumptive serosal cells , prior to the onset of zen expression . Lepidopteran Shx genes represent an unusual example of Hox cluster expansion and integration of novel genes into ancient developmental regulatory networks .
The characterization of Hox genes in the 1980s awakened the idea that there may be similar processes controlling body patterning in divergent animals and gave the first opportunity to compare the control of developmental processes between taxa at a molecular level . In animals as evolutionarily divergent as insects , annelids and vertebrates , Hox genes encode transcription factors deployed in early development , most notably to control spatial identity along the anteroposterior axis of the developing embryo [1] . Conservation of Hox gene function is reflected in their constrained evolution . First , there is high conservation of encoded protein sequence , particularly within the 60-amino acid homeodomain motif ( encoded by the homeobox ) containing three alpha helices . Second , Hox genes are often arranged in a genomic cluster , which was generated by tandem gene duplication early in animal evolution [2] , [3] . Gene order is generally constrained , partly through shared and long-range regulatory elements [1] , [4] , [5] . Third , after expansion of the Hox cluster in early animal evolution there has been relatively little variation in gene number . The ancestor of all Ecdysozoa , Lophotrochozoa and Deuterostomia possessed 7 to 10 Hox genes [3] , and most bilaterian animals still have approximately this number despite hundreds of millions of years of subsequent evolution . The lack of expansion of the Hox gene cluster within Bilateria is intriguing and is in contrast to the pattern of evolution seen for many other sets of genes [6] , [7] . Exceptions are Hox cluster expansion to 15 genes in amphioxus [8] , [9] and duplication of the entire gene cluster in vertebrates [2] , [5] , [10] . There are few recorded cases of tandem duplication within the Hox gene cluster . The best characterised example relates to the Hox paralogy group 3 ( PG3 ) gene of insects , called zerknullt ( zen ) , which has duplicated in a beetle ( Tribolium castaneum ) to yield zen and zen2 [11] , and in cyclorrhaphan flies to generate zen and the highly derived bicoid ( bcd ) [12] . A further duplication specific to the genus Drosophila generated zen2 [13] . Furthermore , early in insect evolution the zen/PG3 gene lost its ancestral function of providing positional identity along the anteroposterior axis , and acquired a novel role in extra-embryonic tissue formation [14] , [15] , [16] . There are indications that the Hox gene cluster also expanded in Lepidoptera . Analysis of the Domesticated Silkmoth Bombyx mori genome revealed a large array of divergent homeobox genes , named Shx ( Special homeobox ) genes , between pb and zen [17] . With 12 Shx loci described , in addition to zen , the canonical Hox genes and another divergent gene ftz , the Silkmoth has the largest Hox gene cluster described [17] . The Silkmoth Shx sequences are highly divergent; some loci have internal duplications manifest as two or three homeobox sequences per gene , and some have disruptive mutations and are probably pseudogenes . The Hox gene cluster has also been characterised in the nymphalid butterflies Heliconius melpomene and Danaus plexippus ( Monarch ) where four homeobox genes were found between pb and zen [18] , [19] . To date , the timing of the gene duplications , the ancestral condition for the Lepidoptera , variation in Shx gene number and gene expression have not been addressed . Here we investigate the origin and evolution of Shx genes through sequencing and assembly of genomes from six species representing successively diverging lepidopteran lineages as well as an outgroup from Trichoptera ( caddisflies ) . We find that four distinct Shx genes arose from the zen gene in the ancestor of the Ditrysia , the clade encompassing most Lepidoptera , and that this complement , not the expanded number found in Bombyx , is the norm across lepidopteran evolution . By modelling tertiary structure , we show that Shx protein sequence is compatible with folding into helix-loop-helix-turn-helix homeodomains . Finally , we determine the expression of Shx genes in early developmental stages of the Speckled Wood butterfly Pararge aegeria . These data suggest that Shx genes encode homeodomain proteins with probable roles in extra-embryonic tissue specification and formation . The lepidopteran zen gene may play a more downstream role in extraembryonic membrane function following serosal closure .
We generated low coverage genome sequences for six species chosen for their phylogenetic positions ( Figure 1B ) . Shx sequence data were also extracted from genome projects of the Silkmoth [20] , the Diamondback moth Plutella xylostella [21] , and the butterflies H . melpomene [18] and the Monarch D . plexippus [19] . The last two species are members of the Nymphalidae , the largest butterfly family , which we elected to sample further using the Comma and Speckled Wood butterflies ( Polygonia c-album and Pararge aegeria ) . To deduce the ancestral condition for the major ditrysian clade encompassing all butterflies and the majority of moths [22] , [23] , [24] , we also selected the Scarlet Tiger moth Callimorpha dominula ( family Arctiidae ) . To examine deeper in the evolutionary history of Lepidoptera , we chose the Horse Chestnut Leafminer moth Cameraria ohridella ( family Gracillariidae ) which , along with the Diamondback moth ( Yponomeutoidea ) represents one of the earliest evolutionary lineages of Ditrysia [21] , [22] , [23] , [24] . As an outgroup to Ditrysia we selected the Orange Swift moth Hepialus sylvina ( synonym Trioda sylvina , family Hepialidae ) , and for an outgroup to the Lepidoptera we used a caddisfly Glyphotaelius pellucidus ( order Trichoptera ) . The Trichoptera and Lepidoptera together form the sister clade to the Diptera ( flies ) . Genomic DNA was sequenced using Illumina HiSeq technology , and multiple assemblies constructed using a range of k-mer sizes . For each species , we sequenced between 31 . 6 and 83 . 1 million paired-end reads granting coverage ranging from 6× to 17× as determined using a k-mer spectrum approach . We generated draft genome assemblies from 337 Mb to 1 . 4 Gb using de Bruijn approaches , yielding N50 values up to 5 . 3 kb . These datasets also provide the first estimates of genome size for these species ( Table 1 ) . Since our goal was gene and homeobox sequence hunting , rather than large-scale synteny analysis , relatively low N50 sizes are sufficient . To determine if the coverage generated was suitable , we searched the assemblies for the canonical Hox genes ( lab , pb , Dfd , Scr , Antp , Ubx , abdA , AbdB ) and ftz . All Hox genes were identified for all species , apart from the homeobox of Orange Swift Ubx , affording confidence in our sequencing approach to identify novel Hox genes in non-model lepidopteran species . In order to confirm that we did not lose genes during assembly of the raw read data , we also applied an alternative assembly strategy that maximally includes all sequence reads . This did not reveal any additional homeobox sequences . We were able to reconstruct genomic scaffolds around the Shx , zen , pb and Dfd genes by manually inspecting and aligning contigs from multiple assemblies , enabling the definition of gene models spanning multiple exons , as well as confirmation of linkage between adjacent genes in several species ( Figure 1 , Table S1 ) . To examine the gene duplication events that generated Shx genes , we used molecular phylogenetic analysis and comparison of gene content between different species . Homeodomain phylogenetic trees demonstrate that the Shx genes form a monophyletic group ( BP 86 , PP 0 . 99 ) and are more closely related to zen than to any other Hox gene ( Figure 1A , Figure S1 ) . This suggests that Shx genes originated by tandem duplication from an ancestral zen gene , consistent with their genomic location between pb and zen ( Figure 1B ) . Sequence alignments incorporating conserved domains outside the homeodomain confirmed this result ( Figure S2 ) . In phylogenetic analyses , Shx genes divide into four distinct orthology groups each present in Speckled Wood , Comma , Scarlet Tiger moth , Horse Chestnut Leafminer and the Diamondback moth . The ShxA , ShxB , ShxC and ShxD groups identified in the butterflies H . melpomene and Monarch therefore originated in the clade Ditrysia , which radiated 100 to 140 Myr ago and encompasses the vast diversity of lepidopteran species [25] , [26] . The identity of putative ShxC genes of the Diamondback moth and Horse Chestnut Leafminer is not clear when only the homeodomain is used , but the existence of conserved motifs outside the homeodomain strongly argues for orthology with ShxC , as does overall protein sequence similarity , gene linkage and phylogenetic analysis with an extended alignment ( Figure 1B , Figures S2 , S3 ) . Our re-analysis of the Silkmoth genome identifies the previously reported Shx1 to Shx11 [17] , plus four additional homeodomain-containing open reading frames which fall within the ShxA and B clades and lie between pb and zen , here named Shx13-16 ( Figure 1 , Figure S4 ) . This observation contrasts with the stability of Shx genes through most of ditrysian evolution . We also investigated the Hox complement in the Orange Swift Moth , an outgroup to Ditrysia but within Lepidoptera , and the caddisfly ( order Trichoptera ) , the sister order to Lepidoptera . We find the Orange Swift moth has no bona fide Shx genes , but several copies of zen gene that do not branch within established Shx groups in our phylogenetic analysis ( Figure 1 , Figure S1 and S3 ) . Three ( zen2 , zen3 and zen4 ) cluster with lepidopteran zen genes while zen1 has a more ambiguous affinity ( Figure 1 , Figures S1 and S3 ) . Presence of diagnostic motifs C-terminal to the homeodomain suggests all are duplications of zen ( Figure S2 and Figure S5G ) . It is less probable that they share a common origin with Shx , with extensive divergence causing ambiguity of orthology assignment . Exons coding for the homeodomains plus a single probable 5′ exon of a zen gene are located on separate scaffolds that could not be linked . The absence of zen duplication before lepidopteran radiation was confirmed by recovery of only a single zen gene in the caddisfly genome . Duplication and divergence of zen is therefore independent in Lepidoptera and Diptera . Shx homeodomains have undergone faster sequence change than homeodomains encoded by zen or the canonical Hox genes . Homeodomain sequence of lab , pb , Dfd , Scr , Antp , Ubx , ftz , abdA and AbdB have 97% to 100% invariant sites across the four ditrysian Lepidoptera genomes sequenced in this study , canonical zen has 98% invariant sites and ShxA , ShxB , ShxC and ShxD have only 83% , 55% , 38% and 38% invariant sites respectively . Although lepidopteran zen and Shx genes are paralogues , and both descend from an ancestral zen , we retain the name Shx established in Bombyx [17] to reflect the more extreme sequence divergence in their homeodomains and to avoid confusion with earlier work . A number of conserved sites within the homeodomain are retained in Shx and zen , and S10 has been identified as unique to Hox3 orthologues ( Figure S5I , red boxes ) [15]; however , outside the homeodomain Shx proteins are radically different from each other and from zen ( Figure S2 , Figure S5C–F ) . All lineages of ditryisian Lepidoptera ( except Bombyx ) have maintained a consistent complement of four different Shx genes , in addition to canonical zen , suggesting the genes have distinct functions . We examined whether gene-specific functions might be reflected in distinct protein motifs . Shx proteins have several short conserved motifs C-terminal to the homeodomain; these are different between the four proteins suggesting they may interact with different co-factors ( Figure S2 , Figure S5C–G ) . Lepidopteran zen shows more extensive protein conservation between species; these motifs are non-overlapping with those of the dipteran zen . Furthermore , analysis of caddisfly shows that motifs shared between basal Diptera and caddisfly have been lost in the Lepidoptera ( Figure S2 , Figure S5G , H ) . Rapid sequence evolution between closely related insect orders is consistent with a previous observation that outside the homeodomain there are no well conserved sequence motifs in zen genes of insects [27] . To investigate the dynamics underpinning diversification of Shx genes , we tested for signatures of selection by comparing synonymous ( dS ) and non-synonymous ( dN ) rates of substitutions in the homeobox region of Shx , zen and Hox genes in a maximum-likelihood framework . These analyses confirmed that there is strong purifying selection acting on the zen homeodomain in Lepidoptera ( dN/dS or ω = 0 . 002 ) comparable to that inferred for canonical Hox genes ( ω ratio of 0 . 001 ) . However , the Shx genes show a marked increase in coding substitutions with a dN/dS ratio of 0 . 06; ShxB ( ω = 0 . 1 ) , ShxD ( ω = 0 . 09 ) and ShxC ( ω = 0 . 05 ) show more coding divergence than ShxA ( ω 0 . 02 ) . Accordingly , an excess of non-synonymous substitution is detected on the branch leading to the ShxB , ShxC and ShxD clade with an inferred ω ratio greater than 1 suggesting an episode of positive selection ( Figure S6 ) . We compared substitution ratios among codons within Shx proteins to determine whether some amino acids show evidence of positive selection . Using a site-model applied to Shx homeodomains only , we found an increased ω ratio at some sites but no statistical support ( Table S2 ) . However , taking the zen outgroup into account , the branch-site model found significant support ( 2Δℓ = 4 . 94 , p<0 . 05 ) for positive selection at five sites ( BEB pp>0 . 95 ) . These sites are located between alpha helices and not known to be functionally involved in protein-DNA interaction ( Table S2 ) . As the Shx homeodomains have diverged extensively from the ancestral zen sequence , we asked whether they had undergone disabling mutations that might prevent them forming stable tertiary folds compatible with binding DNA . We deployed homology modelling based on a well-resolved experimentally-determined tertiary structure of a related Hox protein: that of the Drosophila Antp homeodomain bound to a 13-mer DNA sequence . Using the Comma and Speckled Wood butterfly sequences of ShxA , ShxB , ShxC , ShxD and zen , we first computed the native energy of the deduced structures modelled on the known Antp protein structure . Each yielded a stable predicted helix-loop-helix-turn-helix structure typical of a homeodomain ( Figure 2 ) , although stability was lower when modelled in complex with the specified 13-mer DNA sequence ( Note S1 ) . This suggests that the DNA sequence used was not optimal for these homeodomains . To find more suitable DNA sequences , we used an in silico evolution approach and applied this to protein sequences of Comma , Speckled Wood and Horse Chestnut Leafminer , plus Drosophila Antp as a control . Starting with homopolymeric runs of either A , C , G or T , we ran 1000 cycles of ‘mutation’ and ‘selection’ to find the most energetically stable complexes , and generated consensus DNA sequences representing predicted optimal DNA binding sites for each homeodomain ( Figure 2; Note S1 ) . The evolved consensus sequence generated for Drosophila Antp was an approximation of the known DNA motif including the core ATTA which contacts with helix 3 of the homeodomain , plus a G residue immediately 5′ . The evolved preferred DNA sequences for ShxA , ShxB and ShxC proteins included core ATTA or ATCA motifs , while the ShxD homeodomain showed more variation between the species preferring GTTA , ATTA or TTTA ( Figure 2; Note S1 ) . The zen proteins are somewhat different , tolerating a T in position 4 of the core . These results indicate that Shx and zen proteins have potential to fold into stable helix-loop-helix-turn-helix motifs compatible with sequence-specific DNA-binding . These analyses may not predict the exact in vivo binding sites [28] , [29] . During insect oogenesis , localisation of RNA derived from maternal gene expression establishes the future positions of embryonic and extra-embryonic regions within the oocyte , as well as its body axes ( for an overview of lepidopteran embryology , see Kobayashi et al . [30] ) . Maternal transcripts of zen and ShxC ( and weakly ShxD ) were detected by RT-PCR in ovarioles dissected from Speckled Wood female imagos ( Figure 3A ) . Consistent with this , we also identified these transcripts in a maternal transcriptome dataset [31] ( ShxC:PaContig23051 , GB:GAIX01013843 . 1 , GI:509161192; ShxD:PaContig8659 , GB:GAIX01015570 . 1 , GI:509158266 ) . After egg-laying ( AEL ) each Shx gene has a distinct temporal expression profile ( Figure 3A ) . Our observations and comparison with other lepidopteran species [32] , [33] suggests the onset of blastoderm cellularization and major zygotic transcription commences around 8 h AEL; expression of all four Shx genes plus zen is clearly detected between 8 and 12 h AEL . In situ hybridisation to dissected ovarioles revealed that the spatial distribution of maternal ShxC and ShxD RNA is quite different to that of transcripts from their progenitor gene , zen ( Figure 4 ) . Pre-fertilisation transcripts from ShxC are detected in the nurse cells connected to the oocyte and are concentrated in a novel and striking asymmetrical ‘hourglass’ pattern which excludes the region later fated to become embryonic tissue , and corresponds to the presumptive serosal membranes ( Figures 3B and 4C , Figure S7A–C ) . In contrast , transcripts of ShxD are faintly distributed throughout the developing oocyte without clear subcellular localisation ( Figure 4D ) and zen transcripts are specifically detected in the follicle cells surrounding the oocyte ( Figure 4E ) . In the embryo at 10 h AEL , transcripts of ShxA , ShxB , ShxC and ShxD are each detected in clear hourglass patterns in the cellularised blastoderm matching the earlier maternal ShxC RNA location in the oocyte ( Figure 4F–I; Figure S7E–I ) . The location of Shx transcripts thus marks a clear distinction between the future embryonic regions ( ‘germ anlage’: small cells lacking Shx expression ) and extraembryonic regions ( larger cells expressing Shx genes ) . Within this latter domain , transcripts of ShxD are detected most strongly in the extraembryonic cells bordering the germ anlage ( Figure 4I; Figure S7E–F , H–I ) . At the anterior pole of the egg near the micropyle , a cluster of cells with an increased concentration of ShxD transcripts correspond to a small region that previously lacked maternal ShxC transcripts ( Figure S7D–F ) . In comparison , zen transcripts at 10 h AEL are very weakly detected throughout the blastoderm ( Figure 4J ) . Between 10 and 12 h AEL , the extraembryonic region expands over the germ anlage forming a protective serosal cell layer between the germ anlage and the vitelline membrane ( Figure 3B ) . During this cell movement , ShxC and ShxD transcript levels , already lowered in the anterior ( Figure S7E , F and I ) , reduce dramatically throughout the serosal layer ( Figure 4M and N ) . However transcripts of ShxA and ShxB , which are only of zygotic origin , continue to be detected predominantly in the serosal layer , even after it has enveloped the germ anlage ( Figure 4K , L ) . Transcripts of zen are detected in the serosa for the first time at this stage ( Figure 4O ) showing that expression patterns of zen and the Shx diverge dramatically in both time and space during butterfly embryogenesis . Significant zygotic transcription of the ShxA and ShxD genes was also detected in the large yolk cells beneath the blastoderm at 10–12 h AEL where transcripts were restricted to the nuclei suggesting either incipient transcription or RNA degradation in cytoplasm ( Figure 4F , I; Figure S7H–J ) .
The common ancestor of living arthropods most likely had 10 Hox genes arranged in a single genomic cluster: lab , pb , zen , Dfd , Scr , ftz , Antp , Ubx , abdA and AbdB [3] . The primary roles of Hox genes in bilaterian animals , including arthropods , are to encode positional information and to instruct position-specific cell fate along the anterior posterior axis of the embryo . Two clear exceptions are ftz , which evolved a role in parasegment formation in insects , and zen . The evolutionary history of insect zen has been well studied . In chelicerates and a crustacean the orthologous gene has a typical Hox gene expression pattern [34] , [35] , while during insect evolution the gene diverged in sequence and acquired a different expression pattern and developmental role [14] . In addition to loss of Hox-like function , the zen gene of insects has undergone independent tandem duplications in the Flour Beetle ( to yield zen and zen2 ) and the cyclorrhaphan flies ( to yield zen and bcd ) [12] , [14] . In the Drosophila clade , within the Cyclorrhapha , zen has duplicated again to yield zen and zen2 [36] , [37] . Zen expression has been studied for a range of pterygote insects , including the Desert Locust Schistocerca gregaria , the Milkweed Bug Oncopeltus fasciatus [27] , the Flour Beetle [38] , and the flies [39] . Expression of the Hox3/zen precursor has also been analysed in an outgroup to the Pterygota , the apterygote Firebrat Thermobia domestica [40] . To some extent , inference of ancestral states within the insects is complicated by interspecific variation in the structure and function of the extraembryonic membranes and progression of embryogenesis [27] . In all pterygote insects studied however , zen expression is confined to the extraembryonic tissues with a dominant expression domain associated with early zygotic specification of the serosa , which in some species is accompanied by later , weaker expression in the amnion [14] , [27] , [38] , [41] . Where zen duplication has occurred , both sub- and neofunctionalisation has occurred . Whereas zygotically expressed zen functions in extraembryonic membrane specification in Drosophila , maternally expressed bcd has radically diverged in sequence , and functions as an anterior determinant in the oocyte [12] , [39] . A subsequent Drosophila zen duplication resulted in a putatively dispensable zen2 paralog [36] , unlike in the Flour Beetle where early-acting zen-1 mainly specifies the serosal membranes and late-acting zen-2 coordinates the fusion of amnion and serosa , initiating dorsal closure [38] . In the present study , we demonstrate that the zen gene duplicated during evolution of the Lepidoptera , independently of its duplication in Diptera and Coleoptera . In the Ditrysia , a clade encompassing most of lepidopteran diversity , these duplications generated four distinct Shx genes located next to the ancestral zen gene . Lepidopteran zen and Shx genes are co-orthologues of the ancestral zen gene , hence ShxA to ShxD could logically be called zen2 to zen5 . We retain the term Shx to avoid contradiction with earlier work , and to reflect their extensive sequence divergence and their shared ‘hourglass’ expression pattern in the blastoderm suggesting common functional roles . Additional Shx duplications occurred in the silkmoth lineage , but we find these are not typical of Lepidoptera . In the Orange Swift moth ( Hepialidae ) , which diverged from a more basal node in lepidopteran phylogeny , Shx genes are not present but there is evidence of independent zen gene duplication . These data indicate that the generation of four recognisable Shx genes from an ancestral zen gene occurred after the Ditrysia had diverged; the common ancestor of Ditrysia and Hepialidae may have had multiple copies of zen but none had acquired sequence characters of Shx genes . The common ancestor of Lepidoptera and Trichoptera had just a single zen gene . The Shx genes are therefore an evolutionary novelty of ditrysian lepidopterans . It is striking that all these examples of tandem gene duplication within insect Hox clusters can be traced to the same progenitor gene , zen . Indeed , we find no evidence of duplication of any other Hox gene within the Lepidoptera , and no such event has been reported in another insect . Why should the zen gene be prone to tandem gene duplication ? The answer is likely to lie in the transition from an embryonic to extraembryonic function in the insects . If genomic clustering is important to Hox gene function , through shared enhancers or long-range chromatin effects , then tandem duplication of a canonical Hox gene would most likely disrupt regulation and generate a dominant effect mutation . Conversely , the expression of zen in extra-embryonic structures probably relies on a distinct regulatory mechanism less integrated with that of neighbouring genes; the immediate effect of duplication may therefore simply be increase of transcript dosage . The functional redundancy that is generated then offers potential for subsequent mutations to modify expression of either , or both , daughter genes . After origin of the Shx genes , in an ancestor of the Ditrysia clade , the genes diverged radically in sequence , both within and outside the homeodomain . Within the Lepidoptera , the Shx genes also show an accumulation of coding substitutions , compared to other Hox genes , which likely reflects episodes of positive selection on some sites . In particular , we detect evidence of positive selection after the initial Shx gene duplicated to give ShxA and a progenitor of ShxB , ShxC and ShxD . We also find no Shx pseudogenes ( except in the atypical Bombyx ) , but instead retention of the core set of these genes . Together these observations argue for functional constraints on Shx proteins and the acquisition of new essential roles for these genes in the biology of ditrysian lepidopterans . Sequence divergence in the homeodomain raised the question of whether Shx proteins are still capable of functioning as DNA-binding proteins , potentially regulating the expression of other genes . Evidence that this biochemical role has most likely been retained comes from molecular modelling . We show that despite the extensive accumulation of amino acid substitutions in Shx homeodomains , they still have potential to fold into stable helix-loop-helix-turn-helix motifs with appropriate interaction surfaces for binding to DNA . An in silico evolution approach revealed that the Shx and zen proteins may have subtly different DNA sequence binding preferences , though these are not likely to be grossly dissimilar from target sequences recognised by canonical Hox proteins . We stress that these in silico approaches do not reveal definitive binding sites [28] , [29]; however , they give confidence in the assertion that Shx proteins in Lepidoptera are likely to act as DNA-binding proteins . What roles might Shx genes play in lepidopteran biology ? Embryonic development is similar in the Silkmoth [42] and the Small White butterfly Pieris rapae suggesting conservation across the Ditrysia [30] . Following egg-laying the fertilised egg ( zygote ) undergoes continuous mitotic divisions and in the Silkmoth two regions can be distinguished in the cellular blastoderm based on cell density: the germ anlage which will become the embryo , and the remaining cells which will form the extraembryonic tissues notably the serosa [30] , [42] . As observed for the Speckled Wood butterfly in the current study , in the Small White and Silkmoth , the presumptive serosa has a distinctive hourglass-shape [30] . At 10 h AEL in the Speckled Wood extraembryonic cells become polyploid , large and flat , and by 12 h this sheet of presumptive serosal cells moves over a region where more compact embryonic cells begin to sink into the yolk in the interior of the egg [32] . Serosal closure completes around 12 h AEL in the Speckled Wood butterfly ( summarised in Figure 3B , Figure S8 ) . As the embryonic germ anlage grows , cells at the edge of the anlage differentiate into a second extraembryonic membrane , the amnion , which extends around the ventral surface [30] , [42] . The expression pattern of lepidopteran zen is intriguing because it differs from other insects . In Pterygota , except the Milkweed Bug , zen functions in early embryogenesis in the early specification of the extraembryonic membranes [14] , [16] , including in those species with a zen gene duplication . In the Lepidoptera , we find zen has largely lost this association and is instead expressed in follicle cells and then in the serosa following closure . Lepidopteran zen is therefore likely to have derived roles in the downstream functions of the serosal membrane . For example , we note that as the Speckled Wood zen expression intensifies , the maturing serosa takes on a glossy appearance indicative of cuticle secretion [43] . It has been suggested that the serosa plays roles in the innate immune system , processing environmental toxins , yolk catabolism , cuticle formation and desiccation resistance [44] , [45] . The contrast between zen and Shx gene expression is striking . Our data reveal that Shx genes have a close association with development of the extraembryonic tissues of the Speckled Wood butterfly , but the zen gene does not . Indeed , all four Shx genes are expressed in the presumptive serosa well before zen expression is observed . We suggest that following zen gene duplication in Lepidoptera , the divergent Shx genes retained an ancestral association with extraembryonic membrane specification , while zen gene function diverged radically . It would be a mistake , however , to consider all four lepidopteran Shx genes equivalent , as they have diverged from each other in sequence and in spatiotemporal expression patterns . Most strikingly , in the Speckled Wood there is maternal expression of ShxC and ShxD , but not ShxA and ShxB . It is notable that zen is maternally expressed in Locust and some basal fly species [39] , [41] , whilst in other pterygote insects zen transcripts are zygotically-derived . Maternal expression of ShxC and ShxD suggests that maternal expression may be an ancestral property of the zen gene [41] . However , in the flies and Locust zen transcripts are diffusely distributed within the oocyte , whereas in the Speckled Wood maternally-derived ShxC transcripts are tightly localised in a very distinctive hourglass shape , clearly prefiguring the region where extraembryonic tissues will later emerge after cellularisation . This hourglass pattern of ShxC transcripts within the single cell represents one of the most complex examples of RNA localisation ever reported in any species , and suggests that the Shx genes specify the future serosal tissue domain within the unfertilised oocyte . Differences between Shx gene expression domains are also seen in the embryonic stages: expression of ShxC and ShxD in serosal tissue is joined by expression of ShxA and ShxB , before these two genes become the dominant expressed Shx genes after serosal cell movements around the embryo . The evolution of Shx genes provides some parallels to the evolution of bcd in Diptera . In both cases , the zen gene has undergone tandem duplication , daughter genes have diverged in sequence and there has been recruitment to patterning roles in the unfertilized oocyte .
DNA was extracted from individual adult specimens of the Comma butterfly ( Polygonia c-album ) , the Speckled Wood butterfly ( Pararge aegeria ) , the Scarlet Tiger moth ( Callimorpha dominula ) , the Orange Swift moth ( Hepialus sylvina ) and a caddisfly ( Glyphotaelius pellucidus ) , and from 75 pooled specimens of the Horse Chestnut Leafminer moth ( Cameraria ohridella ) using a phenol-chloroform method [46] . Sources of specimens are given in Table S3 . Paired-end libraries were constructed and sequenced by Oxford Genomics Centre ( http://www . well . ox . ac . uk ) using standard Illumina procedures ( http://www . illumina . com ) . Between 32 million and 83 million 101 bp paired-end reads were collected for each species ( Table 1 ) using HiSeq2000 methodology . Low quality scoring bases were trimmed using sickle ( https://github . com/najoshi/sickle . git ) . We assembled the reads using de Bruijn-based packages Velvet [47] and ABySS [48] with k-mers ranging from 31 to 61 . Table 1 reports assemblies with the best combination of N50 and assembly length; these are available from the Oxford University Research Data Archive ( DOI: 10 . 5287/bodleiandury . 3 ) . Alternative assemblies were also examined to assist with scaffolding around particular genes . As an additional method to identify homeodomain sequence contained in the reads , we also performed assembly using Fermi that implements an overlap-layout consensus approach using a FM-index and is designed to preserve all information in the raw reads [49] . Raw sequencing data are deposited in the NCBI BioProject database under accession number PRJNA241175 . Genome size was determined using the k-mer spectrum approach: the frequency of all possible k-mers of a given length were calculated and plotted to reveal a peak representing the k-mer coverage ( Ck ) , while low and high k-mer coverages correspond to sequencing errors and repeated regions respectively ( Figure S9 ) . K-mer coverage was converted to actual base coverage ( C ) using Ck = C× ( L−k+1 ) /L where L is the read length and k the k-mer size . K-mers were counted and distributions calculated using Jellyfish [50] for a k-mer size of 17 ( Table 1 ) . Analysis of the previously sequenced genomes of Bombyx mori , Heliconius melpomene and Plutella xylostella used data from Silkdb ( http://silkworm . genomics . org . cn/ ) , Butterflygenome ( www . butterflygenome . org ) , KONAGAbase ( http://dbm . dna . affrc . go . jp/px/ ) and the NCBI genome database . Scaffolds corresponding to the region pb-Dfd were downloaded and annotated according to conserved amino acid translations , sequence alignments and , where available , species-specific EST traces . Genome assemblies generated in this study were searched using the Hox homeodomains of H . melpomene , B . mori and P . xylostella using tBLASTn , scaffolds corresponding to significant hits ( 1e-6 ) were extracted and redundant scaffolds dismissed . Gene identification used a combination of phylogenetic analysis and amino acid signatures inside and outside the homeodomain ( see below ) . Contigs containing homeoboxes were manually extended by generating a scaffold tilepath from assemblies obtained at multiple k-mer sizes . Conserved amino acid domains were also used to search for new contigs when scaffolds could not be extended . Gene annotation was carried out manually . Operations were carried out using Geneious V6 ( Biomatters Ltd ) . Gene models were submitted to NCBI with accession numbers listed in Table S1 . For phylogenetic analysis , translated sequences were cropped to either the homeodomain only or the homeodomain plus C-terminal sequence until the deduced stop codon ( ‘extended’ sequences ) ( for deduced translations see Figure S5 ) . The extended sequences were aligned using Cobalt [51] , and edited to exclude sites with a>50% missing data . Maximum-likelihood trees were built using RAxML [52] with an LG+Γ model and 500 bootstrap replicates . Bayesian analysis was performed using Phylobayes-MPI with a C20 pre-defined mixture of profile and a gamma distribution of among-site rate variation [53] . To evaluate the selective processes at play through the evolution of Shx genes , the dN/dS ratio ( or ω ratio ) of the synonymous and non-synonymous rates of substitution was estimated in a maximum likelihood framework using the codeml program of the PAML package [54] . The ‘branch’ model was employed to evaluate the selective effects along the branches leading to distinct groups of Shx genes ( topology as in Figure 1 ) by assigning 2 , 3 or 6 independent ω ratios . Alternatively , site models and branch-site models were employed to assess positive selection at the codon level and the significance of selective effect was assessed using a likelihood ratio test . The probability of sites being under positive selective was evaluated using Bayes Empirical Bayes criteria ( posterior probability>0 . 95 ) . To search for additional motifs outside the homeodomain , deduced translations of genes from Diptera , Trichoptera and Lepidoptera were aligned using Cobalt . Dipteran analysis used the Mothfly ( Clogmia albipunctata ) , Horsefly ( Haematopota pluvialis ) , Dancefly ( Empis livida ) , Scuttlefly ( Megaselia abdita ) , Fruit fly ( Drosophila melanogaster ) and Marmalade Hoverfly ( Episyrphus balteatus ) . Lepidopteran analysis used the five genomes sequenced in the current study plus H . melpomene and the Diamondback moth Plutella xylostella . Caddisfly ( Trichoptera ) sequences were from the present study , and were compared to the Diptera/Lepidoptera alignments ( Figure S5 ) . Conserved motifs were defined as three or more consecutive amino acids present in at least half the species examined and where each residue is shared between divergent lineages ( for Lepidoptera one of Hepialus/Cameraria/Plutella vs . one of Heliconius/Polygonia/Pararge ) . We examined the spatial and temporal expression patterns of the 4 Shx genes and zen in the Speckled Wood ( Pararge aegeria ) using RT-PCR and whole mount in situ hybridization ( WMISH ) . Since zen is involved in extra-embryonic tissue formation in other winged ( pterygote ) insects , we paid particular attention to oogenesis and early embryonic development . For RT-PCR analysis , RNA was extracted from eggs and ovaries obtained from mated 4-day old females taken from a large outbred laboratory stock [31] . To examine zygotic expression , fifty embryos were pooled for time points 0 , 1 , 2 , 3 , 6 , 8 , 10 , 12 , 15 , 20 , 25 , 30 and 48 hours after egg laying ( AEL ) in triplicate . In Lepidoptera , egg laying is nearly synchronous with fertilization , and time after egg laying ( AEL ) can therefore be taken as a proxy for time of development . To examine maternal expression , two mated 4-day old females were sacrificed , the abdomens removed and ovaries dissected in ice-cold PBS . Previtellogenic and vitellogenic regions were separated before RNA extraction [31] . RNA was purified using RNeasy Mini Kit ( Qiagen ) and cDNA synthesized using BioScript ( Bioline ) . The expression of zen and Shx genes was assessed by 35 cycles of RT-PCR using GoTaq polymerase ( Promega ) and primer annealing temperatures of 55°C . All primers are given in Table S4 and their respective position and orientation in Figure S10 . Riboprobes were synthesized using a T7/SP6 DIG RNA labeling Kit ( Roche Applied Science , Penzberg , Germany ) either from linearized plasmids ( ShxA , B , zen ) or PCR amplified templates from Speckled Wood cDNA ( ShxC , D ) purified using QIAquick ( Qiagen , Hilden , Germany ) . In the latter method , initial amplifications using gene-specific primers were followed by a second PCR implementing a modified reverse or forward primer with a T7 5′ tail ( 5′-TAATACGACTCACTATAGGG+Fw/Rev-3′ ) resulting in an antisense or sense template . Regions of genes targeted for RT-PCR and for WMISH are shown in Figure S10 . In-situ hybridisation was carried out on 10 and 12 h AEL eggs which had been dechorionated prior to fixation using 4% sodium hypochlorite . Ovarioles and embryos were fixed in a 1∶1 mixture of heptane and 5 . 5% formaldehyde in PBS in glass vials ( 30 min at 25°C , then 4°C overnight , with gentle rotation ) before gradual dehydration in methanol and storage at −20°C . Samples were hybridized with the riboprobes at 55°C and processed as detailed in Note S2 , developed from Brakefield et al . ( 2009 ) [55] . After WMISH , samples were counter stained with SYTOX Green ( Invitrogen; 450–490 nm ) and imaged using a MZ FL III Stereo-Fluorescence Microscope ( Leica , Wetzlar , Germany ) using a ProgResC3 sensor ( Jenoptik , Jena , Germany ) . The program MODELLER-9v7 [56] was used to model deduced Shx and zen homeodomain sequences onto a previously published crystal structure of the Drosophila Antp homeodomain bound to the DNA sequence AGAAAGCCATTAGAG ( pdb code 9ant; [57] ) . Energetic stability of sequences was initially assessed using the sum of pairwise atomic interactions , estimated solvent interaction and overall combined energy ( see Note S1 ) . To assess stability of binding to the 9ant DNA sequence , the ROSETTA program was deployed ( see Note S1 ) [58] . To identify energetically preferred DNA target sequences for each homeodomain , an in silico evolution approach was applied to Shx and zen homeodomains of Speckled Wood , Comma and Horse Chestnut Leafminer . Proteins were modelled in complex with homopolymer sequences using ROSETTA and then random changes introduced over the 11-core positions with elevated sampling in the inner 9 positions . After each round of mutation protein-DNA complexes were remodelled , side-chain and base-pair packing energies recalculated , and the lowest energy structure , as assessed using ROSETTA and dFIRE3 , used as the next template for mutation [59] . After 1000 rounds of mutation , starting from each homopolymer run , the DNA sequences associated with the 50 lowest energy structures were used to build consensus sequences . Finally , substitution without in silico evolution was used to test for bias introduced from starting with homopolymer runs ( see Note S1 ) . Structures were displayed using CHIMERA [60] and consensus DNA sequences with WebLogo ( http://weblogo . berkeley . edu ) .
|
We have examined gene duplication in a set of ancient genes used in patterning of animal embryos: the Hox genes . These genes code for proteins that bind DNA and switch on or off other genes , and they are very similar between distantly related animal species . Butterflies and moths , however , have additional Hox genes whose origin and role has been unclear . We have sequenced the genomes of five species of butterfly and moth , and of a closely related caddisfly , to examine these issues . We found that one of the Hox genes , called zen , duplicated to generate four new genes in the evolution of the largest group of butterflies and moths . Further mutations greatly modified the DNA sequence of the new genes , although maintaining potential to encode stable protein folds . Gene expression also changed so that the new Hox-derived genes are deployed in egg and early embryonic stages marking the tissues that will later envelop , nourish and protect the embryo .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"arthropoda",
"invertebrates",
"developmental",
"biology",
"genomics",
"genome",
"evolution",
"moths",
"and",
"butterflies",
"genetics",
"biology",
"and",
"life",
"sciences",
"comparative",
"genomics",
"molecular",
"evolution",
"computational",
"biology",
"evolutionary",
"biology",
"gene",
"duplication",
"animals",
"insects",
"organisms",
"evolutionary",
"developmental",
"biology"
] |
2014
|
Ancient Expansion of the Hox Cluster in Lepidoptera Generated Four Homeobox Genes Implicated in Extra-Embryonic Tissue Formation
|
PKR-like endoplasmic reticulum ( ER ) kinase ( PERK ) is an ER-associated stress sensor protein which phosphorylates eukaryotic initiation factor 2α ( eIF2α ) to induce translation attenuation in response to ER stress . PERK is also a regulator of lipogenesis during adipocyte differentiation through activation of the cleavage of sterol regulatory element binding protein 1 ( SREBP1 ) , resulting in the upregulation of lipogenic enzymes . Our recent studies have shown that human cytomegalovirus ( HCMV ) infection in human fibroblasts ( HF ) induces adipocyte-like lipogenesis through the activation of SREBP1 . Here , we report that PERK expression is highly increased in HCMV-infected cells and is necessary for HCMV growth . Depletion of PERK , using short hairpin RNA ( shRNA ) , resulted in attenuation of HCMV growth , inhibition of lipid synthesis and reduction of lipogenic gene expression . Examination of the cleavage of SREBP proteins showed PERK depletion inhibited the cleavage of SREBP1 , but not SREBP2 , in HCMV-infected cells , suggesting different cleavage regulatory mechanisms for SREBP1 and 2 . Further studies showed that the depletion of SREBP1 , but not SREBP2 , reduced lipid synthesis in HCMV infection , suggesting that activation of SREBP1 is sufficient to induce lipogenesis in HCMV infection . The reduction of lipid synthesis by PERK depletion can be partially restored by expressing a Flag-tagged nuclear form of SREBP1a . Our studies also suggest that the induction of PERK in HCMV-infected cells stimulates SREBP1 cleavage by reducing levels of Insig1 ( Insulin inducible gene 1 ) protein; this occurs independent of the phosphorylation of eIF2α . Introduction of an exogenous Insig1-Myc into HCMV infected cells significantly reduced HCMV growth and lipid synthesis . Our data demonstrate that the induction of PERK during HCMV infection is necessary for full activation of lipogenesis; this effect appears to be mediated by limiting the levels of Insig1 thus freeing SREBP1-SCAP complexes for SREBP1 processing .
Viruses rely on the host cells to make viral proteins , replicate viral genomes and produce infectious virions . All the building blocks and energy required for these biosynthetic activities are derived from intermediary metabolism in the host cells . It is important to understand how viral infection manipulates host cell metabolism since it may reveal new targets for antiviral therapy . Studies in the last few years have shown that infection of HCMV can cause dramatic alterations of glucose and glutamine metabolism in the host cells [1]–[3] . Induction of the adipocyte specific glucose transporter 4 ( GLUT4 ) , to replace the less efficient GLUT1 , allows HCMV infected cells to significantly increase glucose uptake [4] . Coupled with increased glucose uptake , glycolysis is greatly upregulated [1] . However , instead of producing energy in the tricarboxylic acid ( TCA ) cycle , a large amount of the glucose-derived carbon exits the mitochondria in the form of citrate to be converted to cytosolic acetyl-CoA to support fatty acid synthesis , which is necessary to support the viral infection [2] . Our recent studies have shown that HCMV infection is able to induce adipocyte-like lipogenesis through the activation of SREBP1 [5] . SREBP proteins , the principle regulators of expression of genes involved in lipogenesis , belong to the basic helix-loop-helix leucine zipper ( bHLH-Zip ) family of transcriptional factors [6] . Unlike other members in this family , SREBPs are made as inactive precursors anchored in the membrane of the ER in complex with SCAP ( SREBP cleavage activation protein ) ( Fig . 1 ) . All SREBP precursors consist of a transcriptional domain at the N-terminus of approximately 500 amino acids containing the bHLH-Zip region , two hydrophobic transmembrane segments interrupted by a short segment ( approximately 30 amino acids ) , and a regulatory domain at the C-terminus of approximately 590 amino acids [7] . In sterol-overloaded cells , SCAP binds to cholesterol which promotes binding to the ER membrane protein Insig1 ( insulin inducible gene 1 ) ; under these conditions the SREBP-SCAP complex is retained in the ER by Insig1 and remains inactive ( Fig . 1 ) . In sterol-depleted cells , the interaction with Insig1 is weakened allowing COPII proteins to recruit the SREBP-SCAP complexes into COPII vehicles and export the complex to the Golgi apparatus where SREBPs are sequentially cleaved by S1P and S2P proteases to release the transcriptional domain ( also called the mature SREBPs ) that are translocated into the nucleus [8] , [9] ( Fig . 1 ) . The mature SREBPs then bind to sterol regulatory elements in the promoters of lipogenic genes to activate their transcription and increase lipid synthesis [6] . Studies in our lab and others have shown that HCMV-infection can induce the cleavage of SREBP1 [5] and SREBP2 [10] . Our data also showed that the HCMV-induced cleavage of SREBPs , and the activation of the host cell lipogenic program , required the SCAP protein [5] , thus the virus appears to be using the SCAP-mediated mechanism for SREBP transport and cleavage . However , the virus has altered this mechanism . In normal cells , the cleavage of SREBP1 is controlled by cellular sterol levels [6]; however , during HCMV infection the normal sterol feedback control is overridden and the cleavage of SREBP1 is constitutive regardless of high levels of sterols [5] . The ER responds to the accumulation of unfolded and/or misfolded proteins in its lumen by activating the UPR , which relieves the ER stress by reducing protein translation , degrading ER-localized mRNAs and misfolded proteins , as well as assisting protein folding . This is primarily accomplished through the activation of three ER-membrane associated sensor proteins: PKR-like ER kinase ( PERK ) , inositol-requiring enzyme-1 ( IRE1 ) and activating transcription factor-6 ( ATF6 ) [11] . Under normal conditions , binding of the ER chaperone protein glucose regulated protein 78 ( GRP78 , also called BiP ) to the ER luminal domains of PERK , IRE1 and ATF6 keeps these three proteins inactive . Following ER stress , BiP dissociates from the ER sensor proteins and preferentially binds to unfolded/misfolded proteins to assist in their folding . This leads to activation of PERK , IRE1 and ATF6 to induce the UPR . The activation of these three proteins affects broad aspects of cell fate and the metabolism of proteins , amino acids and lipids . However , it has been shown that HCMV modulates the effects of all branches of the UPR in order to maintain positive growth conditions for the virus [12] , [13] . For example , our previous studies have shown that HCMV infection activates PERK , an eIF2α kinase , yet its role in eIF2α phosphorylation and translational attenuation is minimized in infected cells [12] . However , other functions of PERK have been reported , specifically , PERK is an important regulator of lipogenesis during adipocyte differentiation through activation of SREBP1 cleavage and upregulation of key lipogenic enzymes [14] . Thus we postulated that activation of PERK during HCMV infection leads to increased SREBP cleavage and lipogenic activation . In the following study we show that PERK expression is significantly increased during HCMV infection and is essential for HCMV growth . Depletion of PERK in HCMV-infected cells resulted in inhibition of viral growth , lipid synthesis , expression of key lipogenic genes , and the cleavage of SREBP1 .
Upon induction of the UPR , PERK activates itself by homodimerization and autophosphorylation [15] , [16] . Our previous studies have shown that HCMV infection induces phosphorylation of PERK [12] . To determine whether PERK expression is elevated during HCMV infection , we examined PERK protein levels by Western analysis . HF cells were infected with HCMV and cell lysates were harvested at the indicated hours post infection ( hpi ) . Figure 2A shows PERK protein levels were greatly increased at 24 hpi and remained at high levels throughout the infection time course to 96 hpi . In order to assess the importance of PERK induction during HCMV infection , we introduced a lentiviral vector expressing short hairpin RNAs ( shRNA ) against PERK mRNA ( shPERK ) to knockdown PERK expression , or a control shRNA against GFP mRNA ( shGFP ) . Figure 2B shows that PERK expression can be eliminated efficiently by two independent shPERKs ( #1 and #2 ) in HF cells after three days treatment . To test viral growth when PERK is depleted , HF cells were treated with shPERK for three days , followed by one day of serum starvation and then infection with HCMV at a multiplicity of infection ( MOI ) of 3 . Viral samples were harvested at the indicated time points ( Fig . 2C ) and viral titers were determined using the 50% tissue culture infective dose ( TCID50 ) method . In Figure 2C , the solid line indicates a normal HCMV growth curve in HF cells treated with the control shRNA , shGFP . The two dashed lines show the severe inhibition of HCMV growth upon depletion of PERK by shPERK #1 and #2 ( Fig . 2C ) . The data indicate that PERK plays a critical role in HCMV lytic infection . We further tested viral protein levels in cells depleted of PERK . Whole-cell extracts were prepared from HCMV-infected cultures treated with shGFP or shPERK #1 at 72 hpi and evaluated by Western analysis . Figure 2D shows that the levels of an immediate-early protein ( IE86 ) , and early protein ( pp52 ) , as well as two late proteins ( pp65 and pp28 ) were not altered in PERK-depleted cells , indicating that PERK depletion does not impact viral gene expression . These data show that the effects of PERK depletion on HCMV growth are not due to disruption of the temporal expression and accumulation of viral proteins , suggesting that the problems arise at the level of virion assembly , where lipid synthesis is required for membrane formation . We and others have reported that inhibition of lipid synthesis can inhibit HCMV growth [2] , [5] . Beside its role in ER stress , PERK also serves as a critical regulator of lipid metabolism [14] . Given these data and the results in Figure 2 , it is predicted that PERK might be essential for HCMV-induced lipogenesis . We examined cellular lipid levels using a fluorescent lipophilic dye BODIPY 493/503 to visualize lipid droplets [17] . Figure 3A shows that mock-infected HF cells expressing the control shRNA ( shLuc ) have low levels of lipid droplets and these are further reduced in PERK depleted cells . In agreement with our previous data [5] , [18] , the level of lipid droplets is greatly increased after HCMV infection; however , this is dramatically reduced in PERK-depleted cells ( Fig . 3A ) . To further confirm the inhibition of lipid production by PERK depletion during HCMV infection , we also assayed total lipid synthesis by measuring incorporation of 14C-labeled acetate into lipids . Two independent assays showed that total lipid synthesis in HCMV-infected cells was about 3 times higher than that in mock-infected cells at 48 hpi ( Fig . 3B and 3C ) . Consistent with lipid droplet staining , depletion of PERK resulted in a 50 to 65% reduction in total lipid synthesis in mock-infected cells . In HCMV-infected cells , total lipid synthesis was reduced 60–70% by PERK depletion ( Fig . 3B and 3C ) . Our previous data showed HCMV infection induced expression of key lipogenic genes [5] . We used quantitative RT-PCR to determine the levels of mRNAs encoding key lipogenic enzymes [acetyl-CoA carboxylase 1 ( ACC1 ) , ATP-citrate lyase ( ACL ) , fatty acid synthetase ( FAS ) and HMG-CoA reductase ( HMGCR ) ] in normal and PERK-depleted cells . Figure 4 shows that HCMV infection induces expression of ACC1 , ACL , FAS , and HMGCR in cells treated with shGFP , confirming previous results [5] . However , the induction of these mRNAs by HCMV infection was substantially inhibited in PERK depleted cells ( Fig . 4 ) . The data in Figure 3 and 4 indicate that PERK is necessary for the full induction of lipogenesis in HCMV-infected cells . The expression of lipogenic genes is regulated by the SREBPs . Mammalian cells have two genes encoding three SREBP isoforms: SREBP1a , SREBP1c and SREBP2 [19] , [20] . SREBP1a and 1c are encoded by a single gene , SREBF-1 , located on human chromosome 17p11 . 2 [21] , [22] . The SREBP1a and 1c transcripts are produced through the use of alternative transcription start sites and only differ in their first exon resulting in a loss of 30 amino acids at the N-terminus of SREBP1c; SREBP1a and 1c cannot be differentiated immunologically . In cultured human fibroblasts , the SREBP1a transcript predominates [23] . The three SREBP proteins control the expression of more than thirty genes involved in lipid metabolism [7] . SREBP2 predominantly upregulates genes involved in cholesterol biosynthesis and SREBP1c activates genes for fatty acid biosynthesis , while SREBP1a regulates gene expression for both fatty acid and cholesterol synthesis [24] . Previously , we have reported that SREBP1 is cleaved and activated in HCMV infection [5] . SREBP2 is also cleaved in HCMV infection , as reported by others [10] . Given the importance of PERK in lipid synthesis and induction of lipogenic genes during HCMV infection , we predicted that PERK depletion may inhibit activation of SREBPs . Cleavage of either SREBP1 or SREBP2 will produce mature forms of SREBPs with a size of approximately 60 kilodalton ( kDa ) . Figure 5A shows a Western analysis of cell lysates from an HCMV infection time course probed with an anti-SREBP2 antibody . The antibody detects the 125 kDa SREBP2 precursor ( P ) as previously reported , the level of the precursor declines as the HCMV-infection proceeded [10] . Concurrently a protein band of approximately 58 kDa increases during HCMV-infection ( Fig . 5A ) , and appears to be the mature form of SREBP2 . However , a longer exposure of the same blot reveals an additional band migrating at about 60 kDa , slightly above the 58 kDa protein ( Fig . 5B , arrow ) . To identify which band represents the real mature form of SREBP2 , we examined the cleavage of SREBP2 in HCMV infected cells depleted of SREBP2 using an shRNA , shSREBP2 . In mock- and HCMV-infected cells , shSREBP2 efficiently reduced the precursor ( P ) levels of SREBP2 compared to mock- or HCMV-infected cells treated with the shGFP control ( Fig . 5C ) . Examination of the two proteins migrating at 58 and 60 kDa in infected cells showed that depletion of SREBP2 resulted in the loss of the less intense 60 kDa protein ( arrow ) , while the 58 kDa protein band remained unchanged ( Fig . 5C ) . Similar results were obtained in cells treated with two other shSREBP2s ( Fig . S1A ) which target different regions of the SREBP2 mRNA . Additionally , the 60 kDa but not the 58 kDa protein was not detected in cells depleted of SCAP protein ( Fig . S1B ) which is required for the cleavage of SREBP2 [6] . All these results show that the less intense 60 kDa band represents the mature form of SREBP2 and that the 58 kDa band represents a virally induced , cross reacting protein . We next assessed the cleavage of both SREBP1 and SREBP2 in mock- and HCMV-infected cells treated with shGFP or shPERK . Normal precursor levels of SREBP1 and SREBP2 were detected in shGFP control treated cells; after HCMV infection , both SREBP1 and SREBP2 had decreased precursor levels but the mature forms accumulated as shown in Figure 5D . In mock-infected control cells treated with shGFP , a small amount of mature SREBP1 was detected and this was lost by depletion of PERK . In infected cells PERK depletion resulted in a severe reduction of the mature forms of SREBP1 . Interestingly , neither the precursor nor the mature form of SREBP2 were affected by PERK depletion in mock- and HCMV-infected cells ( Fig . 5D ) , indicating that SREBP1 and SREBP2 are cleaved by different mechanisms in HCMV infection , and PERK is necessary for the cleavage of SREBP1 only . The fact that SREBP1 and 2 are cleaved by different mechanisms during HCMV infection is supported by the data in Figure 6 . We previously reported that SREBP1 can be cleaved successfully even in the presence of high level of sterols in HCMV infected cells , indicating that the normal sterol feedback control of the SREBP1 maturation is overridden by HCMV infection [5] . We examined the cleavage of SREBP2 in the same experiment . Figure 6 shows that SREBP1 is cleaved either in the presence or absence of supplemental sterols in HCMV-infected cells . However , under the same infection conditions the cleavage of SREBP2 was inhibited by the supplement of sterols . The difference in the maturation of SREBP1 and 2 may underlie their different roles in HCMV infection . Since SREBP1 and SREBP2 have different sets of target genes , they may play different roles in HCMV-induced lipogenesis . Thus we determined how lipid synthesis and HCMV growth were affected in cells that are depleted for either SREBP1 or SREBP2 . We first determined how the depletion of one SREBP affected the levels of the other . Figure 7A shows that an shRNA against SREBP1 ( shSREBP1 ) caused a dramatic loss of the precursors and mature forms of SREBP1 . However , this loss of SREBP1 had no effect on the precursor level of SREBP2 in uninfected HF cells and there was a slight increase of the mature form of SREBP2 in HCMV-infected cells ( Fig . 7A ) ; this agrees with the results of SREBP1 knockout experiments [25] . A similar result was seen in SREBP2 depleted cells ( Fig . 7B ) ; depletion of SREBP2 reduced the precursor and mature form of SREBP2 in mock and infected cells , but had little effect on the levels of the SREBP1 precursors or mature forms . These data show that the depletion of one SREBP protein has no effect on the levels or cleavage of the other . We next tested total lipid synthesis in SREBP1 and SREBP2 individually depleted cells . Figure 8A showed that depletion of SREBP1 inhibited lipid synthesis in both mock and HCMV infected cells . Unexpectedly , the depletion of SREBP2 resulted in no reduction of lipid synthesis in HCMV-infected cells . In mock-infected cells , lipid synthesis was actually increased by the depletion of SREBP2 , due to a compensatory effect of the loss of SREBP2 activity [26] . To further evaluate the importance of SREBP proteins in HCMV infection , we tested HCMV growth in SREBP1 or SREBP2 depleted cells . The viral growth data in Figure 8B show that either SREBP1 or SREBP2 depletion slowed HCMV growth in HF cells . However , SREBP1 depletion clearly had more severe effects than SREBP2 depletion . This further confirms that SREBP1 is playing a more central role in HCMV-induced lipogenesis than SREBP2 . As shown above , PERK depletion significantly slows HCMV growth and inhibits HCMV-induced lipogenesis; these correlate with reduction of the levels of the mature forms of SREBP1 in PERK-depleted , HCMV-infected cells . Thus we asked if lipid synthesis can be restored , at least partially , by expressing the nuclear ( mature ) form of SREBP1 in PERK-depleted , HCMV-infected cells . Since SREBP1a is the predominant transcript of the SREBF-1 gene in fibroblasts [23] , we used a retroviral vector to transduce a Flag-tagged SREBP1a nuclear form , 2×FLAG-SREBP1a ( N ) , into HF cells . HF cells , 30–50% confluent , in 60 mm dishes were infected twice by retroviruses expressing 2×FLAG-SREBP1a ( N ) or GFP ( control ) . After establishing these cells , they were tested for expression . Figure 9A shows good expression of 2×FLAG-SREBP1a ( N ) in HF cells after retroviral transduction . The 2×FLAG-SREBP1a ( N ) and GFP expression cells were treated with lentiviral vectors expressing shLuc or shPERK for three days . The cells were then serum-starved for 24 hours and then infected with HCMV ( MOI = 3 ) . Lipid synthesis was assayed at 48 hpi . Figure 9B shows that in the control cells , transduced by the GFP expressing retrovirus and treated by shLuc , HCMV infection increased cellular lipid synthetic levels by greater than three fold compared to mock . PERK depletion greatly reduced the normal lipid synthetic rate in the mock infected cells and eliminated the activation by HCMV . These data agree with those shown in Figure 3B and 3C . In the cells transduced with the SREBP1a ( N ) expressing retrovirus and treated with shLuc , lipid synthesis was significantly increased compared to the GFP expressing controls . The lipid synthetic level was only modestly increased by HCMV infection in cells expressing SREBP1a ( N ) and shLuc; this is likely due to fact that the active SREBP1a is already in the nucleus of viral infected cells , maximally activating transcription of lipogenic genes . Interestingly , lipid synthesis was reduced about 60% by PERK depletion in mock-infected cells expressing SREBP1a ( N ) . However , this level remains three times higher than that in mock-infected cells expressing GFP , demonstrating that SREBP1a ( N ) expression can , in part , compensate for the effects of PERK depletion . In PERK-depleted HCMV-infected cells , lipid synthesis was doubled by the expression of SREBP1a ( N ) compare to the expression of GFP , again showing that SREBP1a ( N ) expression can , in part , compensate for the effects of PERK depletion . However , the observation that SREBP1a ( N ) cannot fully compensate for PERK depletion , especially in mock-infected cells , suggests that PERK has additional effects on SREBP1 function or other aspects of lipid synthesis that remain to be discovered . As shown in Figure 1 , the SREBP1-SCAP complex is retained in the ER membrane by Insig1 when sterol levels are adequate . Thus an appropriate level of Insig1 must be maintained under these conditions to keep the SREBP1 in the uncleaved , inactive state . Importantly , Insig1 is a very unstable protein with a half-life shorter than 30 min [27]; thus the control of the Insig1 levels may be an important factor in the control of SREBP1 activation . In this regard , it is known that SREBP1 is cleaved more efficiently , and more of the active transcription factor accumulates in the nucleus , in Insig1 deficient cells [28]; and the cleavage of SREBP1 is also enhanced via direct depletion of Insig1 using siRNA [29] . Due to the lack of appropriate antibody to detect endogenous Insig1 in human cells , we introduced Myc-tagged Insig1 to examine if Insig1 protein levels could be altered due to the increase of PERK expression in HCMV infection . Plasmid pCMV-Insig1-Myc , a cDNA of Myc-tagged human Insig1 cloned in pCDNA3 vector [30] , was electroporated into HF cells and stably transfected cells were selected using the drug G418 . Pooled G418 resistant cells were treated with shGFP or shPERK for three days , followed by one day of serum starvation , then the cells were either mock- or HCMV-infected . Whole cell extracts were prepared for Western analysis at 48 hpi . In shGFP-treated , mock-infected cells the level of Insig1-Myc was nearly undetectable ( a short , S , and a long , L , exposure of the Indig1-Myc Western are shown ) . In contrast , HCMV infection of these cells showed increased Insig1-Myc levels in the shGFP expressing cells , presumably due to HCMV-mediated transcriptional activation of the major immediate early promoter used in this expression vector ( Fig . 10A ) . In shPERK-treated , mock-infected cells the depletion of PERK caused a significant increase in Insig1-Myc and this increase was even greater in the HCMV-infected cells . These results are consistent with a model where low levels of PERK result in increased Insig1 levels and high levels of PERK , as occurs in HCMV-infected cells , result in low levels of Insig1 . Recent data have shown that induction of the UPR by thapsigargin treatment induces a rapid loss of Insig1 in a PERK- and phospho-eIF2α-dependent manner [14] . This suggests that PERK's function as an eIF2α kinase may be involved . Thus we examined the phosphorylation status of eIF2α in the shGFP and shPERK treated cells . Figure 10A shows that levels of phospho ( P ) - and total ( T ) -eIF2α were increased during HCMV infection as we have previously reported [12] . However , there is no change in either total- or phospho-eIF2α levels comparing HCMV-infected cells expressing shGFP or shPERK . These data suggest that during HCMV infection phosphorylation of eIF2α is not mediated by PERK , in agreement with a recent report that the phosphorylation of eIF2α induced by HCMV infection is mediated , nearly exclusively , by the RNA activated protein kinase ( PKR ) [31] . Thus PERK's function to activate SREBP1 is independent of its eIF2α kinase activity during HCMV infection . The data in Figure 10A are consistent with the HCMV-mediated induction of PERK causing a reduction in endogenous Insig1 levels resulting in the increase of SREBP1 cleavage and lipid synthesis . Thus we predicted that increasing Insig1 levels through the introduction of an exogenous Insig1 would reduce HCMV growth and lipid synthesis in viral infected cells . To test this , we examined viral growth and total lipid synthesis in cells expressing exogenous Insig1-Myc . To maximize Insig1-Myc production we transiently transfected HF cells , via electroporation , with the plasmid expressing Insig1-Myc or its vector control , pCDNA3 . Subsequent to electroporation the cells were infected with lentiviruses expressing shPERK or the control shGFP . In other experiments ( Fig . 2D , 5D and 10A ) the efficient depletion of PERK required three days of shPERK treatment and followed by one day of serum starvation prior to HCMV infection . Since the level of transiently transfected plasmid DNA decreases rapidly after transfection , we reduced the shPERK treatment . Hence , cells were cultured overnight at 37C after electroporation , and then treated with shPERK or shGFP for two days; this was followed by two hours of serum-starvation and then HCMV infection . Viruses were harvested at 72 hpi for titration and lipid synthesis was assayed at 48 hpi . The inset in Figure 10B shows that due to the shortened shPERK treatment the PERK protein level was reduced only about 60% , not as thoroughly as in other experiments . Nonetheless , this level of PERK depletion resulted in significant reduction in viral production in cells electroporated with the control vector plasmid ( Fig . 10B ) . In shGFP treated control cells , the introduction of Insig1-Myc resulted in a 17-fold decrease in viral production . This is a substantial reduction considering that the Insig1-Myc was introduced by electroporation . Figure S2A shows a typical image of HF cells electroporated with a plasmid expressing red fluorescent protein ( RFP ) and three independent electroporations show that only 50 to 60% of the cells are transfected ( Fig . S2B ) . The data suggest that increasing the levels of Insig1 in the infected cells can overcome the effects of induced PERK during infection; the further depletion of PERK in the Insig1-Myc transfected cells had little additional effect . The total lipid synthesis assay results in Figure 10C show that in Mock-infected vector control cells , PERK depletion reduced lipid synthesis by approximately 40% , the introduction of Insig1-Myc into these cells only slightly reduced lipid synthesis and inhibition of lipid synthesis resulting from the combination of PERK depletion and by Insig1-Myc expression was additive . In vector-electroporated and HCMV-infected cells , lipid synthesis was reduced 40% by PERK depletion , recall that PERK was not totally depleted under the conditions used ( Inset Fig . 10B ) . The introduction of Insig1-Myc into the shGFP treated , HCMV-infected cells significantly reduced total lipid synthesis , once more showing that increased production of Insig1 can reverse , at least partially , the effects of PERK induction in infected cells . Again it should be noted that the effect level of Insig1-Myc must be considered in the light that this is a transfections experiment where only a subset of the cells are overexpressing Insig1-Myc . The combination of Insig1-Myc expression and PERK depletion resulted in the greatest reduction of total lipid synthesis in infected cells . All of the above data are consistent with a mechanism where the increased amounts of PERK in HCMV-infected cells results in lowering the endogenous levels of Insig1 which would promote the transport of the SREBP1-SCAP complex , cleavage of SREBP1 and the subsequent activation of lipogenic genes .
The activation of ER stress affects broad aspects of cell fate and metabolism , including lipid metabolism . Our data show that expression of PERK , an ER sensor protein , was highly elevated by HCMV infection and that this is critical for HCMV growth . Depletion of PERK resulted in decreased viral growth , lipid synthesis and expression of key lipogenic genes . Previous data have shown that induction of the UPR by thapsigargin treatment induces a rapid loss of Insig1 in a PERK- and phospho-eIF2α dependent manner [14] . Our data also suggest that a PERK mediated mechanism for controlling Insig1 levels is used by HCMV . Specifically , the increase in PERK production may result in the loss of endogenous Insig1 to further activate the cleavage of SREBP1 in HCMV-infected cells . In HCMV infected cells , this process is independent of PERK-mediated eIF2α phosphorylation since eIF2α is mainly phosphorylated by PKR in HCMV infection [31] . While our study shows that PERK depletion affects the SREBP1 cleavage pathway via Insig1 , it also suggests that PERK may have other effects on SREBP expression and lipid synthesis . For example , we note that coincident with the reduction of the mature SREBP1 by PERK depletion , the level of the SREBP1 precursors is also decreased in HCMV-infected cells ( Fig . 5D ) . We cannot rule out this result from PERK-mediated stabilization of the precursor and/or mature SREBP1 . However , a transcriptional mechanism may be more likely since SREBF-1 is a target gene of SREBP proteins , both nuclear forms of SREBP1 and SREBP2 can transcriptionally activate SREBF-1 [7] . Thus the inhibition of SREBP1 maturation caused by PERK depletion would be expected to reduce SREBF-1 transcription resulting in reduced levels of SREBP1 precursors . That PERK depletion affects SREBF-1 transcription is supported by quantitative RT-PCR data ( Fig . S3 ) which show RNA level of SREBP1 was reduced by PERK depletion in HCMV-infected cells . PERK is well known to have transcriptional effects via the transcription factor Nrf2 ( NF-E2-related factor-2 ) , a prosurvival transcription factor [32] which moves from the cytoplasm to the nucleus upon phosphorylated by PERK [32] , [33] . However , previous studies from this lab have shown that Nrf2 is maintained in the cytoplasm during the course of HCMV infection despite PERK activation [34] . PERK also has an intrinsic lipid kinase activity , one function of which is to activate the multifunctional protein kinase Akt [35] . However , Akt is only temporally activated during the immediate-early/early phase of HCMV infection [36] , [37] and the total protein levels of Akt decrease in late HCMV infection [37] . Thus it is questionable whether Akt affects lipid synthesis during infection; however , PERK's lipid kinase activity may have other functions related to the activation of lipid synthesis that are exploited by HCMV . Previous studies have suggested that SREBP1 is more important for lipogenesis and adipocyte differentiation than SREBP2 [24] . Studies using transgenic and knockout mice showed that SREBP1a regulates gene expression related to both fatty acid and cholesterol synthesis , SREBP1c activates only genes related to fatty acid synthesis , and SREBP2 is more specific for genes related to cholesterogenesis [7] . Our data agree with these findings , in HCMV infected cells depletion of SREBP1 ( both 1a and 1c ) reduced total lipid synthesis while depletion of SREBP2 did not have the same effect . One explanation for this is that the transcriptional capacity of SREBP2 may be partially replaced by SREBP1a . Still , SREBP2 is not irrelevant during HCMV infection; however , clearly the virus grows better when SREBP1 is present . In normal cells the three SREBP isoforms are matured via the same mechanism ( Fig . 1 ) [24] . In HCMV-infected cells , however , the depletion of PERK reduced the level of the mature SREBP1 but had no effect on the level of the mature SREBP2 . Thus SREBP1 and 2 are processed differentially in HCMV-infected cells . A difference in the processing mechanisms is also indicated by the observation that in infected cells excess sterols do not inhibit SREBP1 processing but do inhibit SREBP2 processing . Our data suggest that PERK induction in infected cells reduces the levels of Insig1 , possibly the PERK effect is targeted to Insig1 in association with SREBP1-SCAP complexes and not SREBP2-SCAP complexes . Our data suggest that neither the depletion of SREBP 1 nor SREBP2 can completely inhibit lipid synthesis; this may be due to incomplete depletion , partial overlap of functions or the involvement of other lipogenic transcription factors . In this regard , most lipogenic genes can also be activated by the carbohydrate responsive element-binding protein ( ChREBP ) [38] . ChREBP associates with Max-like protein X ( Mlx ) to serve as a glucose responsive transcriptional factor . It can work synergistically with SREBP1 to upregulate transcription of lipogenic genes . In mammalian cells , high level of nutrients , such as high glucose , can induce lipogenesis independent of SREBP1 [39] , [40] . It has been recently reported that genetically engineered mice overexpressing GLUT4 can induce lipogenesis through upregulation of ChREBP in adipose tissues [41] . Our recent studies have shown that HCMV infection greatly increases GLUT4 expression [4] , suggesting that ChREBP may be activated during HCMV infection , in addition to the SREBPs . In conclusion , our data demonstrate that the induction of PERK during HCMV infection is necessary for full induction of lipogenesis . Depletion of PERK leads to the inhibition of the maturation of SREBP1 , lipid synthesis and HCMV growth . Our data suggest that PERK mediates this effect by limiting the levels of Insig1 , thus freeing SREBP1-SCAP complexes for SREBP1 processing . This function of PERK appears to be independent of its role as an eIF2α kinase .
Primary and life-extended human foreskin fibroblasts ( HFs ) [42] were propagated and maintained in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% fetal calf serum , 100 U/ml penicillin , 100 µg/ml streptomycin , and 2 mM GlutaMAX ( all reagents were obtained from Invitrogen ) . Cholesterol ( C3045 ) and 25-hydroxycholesterol ( H1015 ) were purchase from Sigma and used at concentrations described in Figure 6 . The following plasmids were used in this study: pCMV-Insig1-Myc [30] , pCDNA3 . 1-2×FLAG-SREBP1a ( N ) ( Addgene plasmid 26801 ) , pDsRed1-C1 ( Clontech ) ; pBabePuro-GFP was made by subcloning EGFP cDNA into BamHI and EcoRI sites of retroviral vector pBabe-Puro [43] . Lentiviral expression plasmids encoding for control shRNA shGFP [44] , shLuc [4] were previously described . Lentiviral expression plasmids encoding for shPERK ( TRCN0000001401 , TRCN0000001399 ) , shSREBP1 ( TRCN0000020604 ) and shSREBP2 ( TRCN0000020665 , TRCN0000020667 , TRCN0000020668 ) were purchased from OpenBiosystems . Plasmid electroporation into HF cells was performed using U-023 program for the Amaxa Nucleofector and electroporation kit for primary fibroblasts ( Lonza ) as described in the kit instruction . HCMV ( Towne strain ) stocks were prepared and purified as previously described [36] . All HCMV experiments were performed in serum-starved HF cells by infection with a Towne strain of HCMV ( MOI of 3 ) derived from a bacterial artificial chromosome clone modified to express green fluorescent protein ( GFP ) under the control of the simian virus 40 early promoter [45] . For lipid droplet staining using BODIPY 493/503 , the wild type Towne strain of HCMV without the cassette containing the gene that encodes GFP was used . For viral growth assays , cells in 60 mm dishes were washed once with serum-free DMEM after 2 hours of viral incubation at 37C and then refed with 2 ml of serum-free DMEM . Viruses were harvested at indicated times and viral titers were determined using the TCID50 method . The experiments were set up in duplicate . Lipid droplets were stained with BODIPY 493/503 [4 , 4-difluoro-1 , 3 , 5 , 7 , 8-pentamethyl-4-bora-3a , 4a-diaza-s-indacene] ( Molecular Probes; Invitrogen ) as described [17] . Briefly , confluent HF cells in 35 mm dishes were serum-starved for one day , then mock or HCMV infected in serum-free DMEM . At 48 hpi , cells were rinsed with ice-cold PBS and fixed with 4% paraformaldehyde for 30 min at room temperature , followed by three washes with PBS . Cover the cells with 1 ml of 1 . 0 µg/ml BODIPY 493/503 staining solution and incubate 10 min at room temperature , protected from ambient light . Wash the cells three times with PBS and mounted using Vectashield containing DAPI . The images were captured at the same microscopy exposure setting . Total lipid synthesis assay was performed as described previously [5] . Retroviral vector pBabePuro-2×FLAG-SREBP1a ( N ) was made by subcloning cDNA of Flag-tagged SREBP1a nuclear form from pCDNA3 . 1-2×FLAG-SREBP1a ( N ) into BamHI and EcoRI sites in pBabe-Puro vector . Retroviruses , Retro-GFP and Retro-SREBP1a ( N ) , were made as described [43] . Briefly , retroviral vector pBabePuro-GFP or pBabePuro-2×FLAG-SREBP1a ( N ) was cotransfected with amphotropic packaging plasmid QPSI ( a gift from Dr . J . Alan Diehl , University of Pennsylvania ) at a ratio of 1∶1 into 293T cells . Conditioned culture medium containing infectious retroviral particles was harvested at 24 , 48 and 72 hours post-transfection , and used to infect exponentially growing HF cells in the presence of 8 µg/ml polybrene . Total RNA was isolated using the RNeasy Mini Kit from Qiagene according to the manufacturer's protocol . Two µg of total RNA was used for first-strand cDNA synthesis by using SuperScript first-strand synthesis system ( Invitrogen ) . For quantitative PCR , input cDNA was analyzed in triplate . All reactions were performed by using Taqman Universal PCR Master Mix kit in ABI 7900 PCR system ( Applied Biosystems ) . The following primer sets ( Applied Biosystems ) were used: ACC1 ( assay identification [ID] Hs01046047_m1 ) , ACL ( assay ID Hs00153764_m1 ) , β-actin ( assay ID Hs99999903-m1 ) , FAS ( assay ID Hs00188012_m1 ) , HMGCR ( assay ID Hs00168352_m1 ) , and SREBP1 ( assay ID Hs01088691_m1 ) . Gene expression data were normalized to β-actin mRNA levels . Lentiviral vectors expressing shRNA were made as described [46] . Subconfluent HF cells were infected with lentiviral vectors in the presence of 8 µg/ml polybrene for 3–4 h , followed by the addition of fresh medium . For gene silence experiments , cells need to be treated with shRNA at least for 3 days and followed by one day of serum starvation , then infected with HCMV ( MOI of 3 ) in serum-free DMEM for designed assays . Western blotting was performed by the procedures described previously [47] . The following antibodies were used in this study: anti-actin ( MAB1501; Chemicon ) , anti-eIF2α ( ab50733 , Abcam ) , anti-ex2/3 [48] , anti-Flag ( MA1-91878 , Thermo Scientific ) , anti-Myc ( sc-789 , Santa Cruz ) , anti-PERK ( Ab65142 , Abcam ) , anti-phospho-eIF2α ( 44-728G , Invitrogen ) , anti-pp28 ( sc-56975 , Santa Cruz ) , anti-pp52 ( sc-69744 , Santa Cruz ) , anti-pp65 ( sc-52401 , Santa Cruz ) , anti-SREBP1 ( 557036; BD Biosciences ) , anti-SREBP2 ( Ab30682 , Abcam ) . GenBank accession and protein ID numbers of all genes/proteins mentioned in this study are listed in the following: ACC1 ( NM_198834 ) , ACL ( NM_001096 ) , Actin ( NP_001091 ) , eIF2α ( NM_004094 ) , FAS ( NM_004104 ) , HMGCR ( NM_000859 ) , IE86 ( HCMV UL122 , AAR31449 . 1 ) , Insig1 ( AY112745 ) , PERK ( NM_004836 ) , pp28 ( HCMV UL99 , ACM48076 . 1 ) , pp52 ( HCMV UL44 , ACM48032 . 1 ) , pp65 ( HCMV UL83 , ACM48061 . 1 ) , SREBP1a ( NM_001005291 ) , SREBP1c ( NM_004176 ) , and SREBP2 ( NM_004599 ) .
|
HCMV , a β-herpesvirus , is a significant pathogen which infects most of the human population by puberty . Primary HCMV infection can be unnoticed in healthy people , but can be life threatening for the immunocompromised and it is the most important cause of congenital infection in the developed world , frequently leading to deafness , mental retardation and developmental disability . HCMV infection alters cellular signaling and metabolism in order to establish and maintain an optimal cellular environment that can accommodate the increased demands for nutrients , energy , and macromolecular synthesis that accompany viral infection . On the other hand , increased demands for nutrients , energy and increased protein loading to the ER can induce ER stress , particularly the unfolded protein response ( UPR ) . HCMV induces the UPR in infected cells but also highly regulates its effects . Our recent studies showed HCMV infection can also induce adipocyte-like lipogenesis by activation of the transcription factor SREBP1 . We now provide evidence that the induction of the UPR is connected to lipogenic activation during HCMV infection . We show that the ER stress sensor protein , PERK , is critical for lipogenic activation induced during HCMV infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology",
"biology",
"microbiology"
] |
2013
|
PKR-Like Endoplasmic Reticulum Kinase Is Necessary for Lipogenic Activation during HCMV Infection
|
Genetic robustness , or fragility , is defined as the ability , or lack thereof , of a biological entity to maintain function in the face of mutations . Viruses that replicate via RNA intermediates exhibit high mutation rates , and robustness should be particularly advantageous to them . The capsid ( CA ) domain of the HIV-1 Gag protein is under strong pressure to conserve functional roles in viral assembly , maturation , uncoating , and nuclear import . However , CA is also under strong immunological pressure to diversify . Therefore , it would be particularly advantageous for CA to evolve genetic robustness . To measure the genetic robustness of HIV-1 CA , we generated a library of single amino acid substitution mutants , encompassing almost half the residues in CA . Strikingly , we found HIV-1 CA to be the most genetically fragile protein that has been analyzed using such an approach , with 70% of mutations yielding replication-defective viruses . Although CA participates in several steps in HIV-1 replication , analysis of conditionally ( temperature sensitive ) and constitutively non-viable mutants revealed that the biological basis for its genetic fragility was primarily the need to coordinate the accurate and efficient assembly of mature virions . All mutations that exist in naturally occurring HIV-1 subtype B populations at a frequency >3% , and were also present in the mutant library , had fitness levels that were >40% of WT . However , a substantial fraction of mutations with high fitness did not occur in natural populations , suggesting another form of selection pressure limiting variation in vivo . Additionally , known protective CTL epitopes occurred preferentially in domains of the HIV-1 CA that were even more genetically fragile than HIV-1 CA as a whole . The extreme genetic fragility of HIV-1 CA may be one reason why cell-mediated immune responses to Gag correlate with better prognosis in HIV-1 infection , and suggests that CA is a good target for therapy and vaccination strategies .
Genetic robustness is defined as the ability of a biological entity ( e . g . a protein or organism ) to maintain function in the face of mutations [1] , [2] . More robust proteins or organisms tolerate higher mutation rates while less robust ( more ‘brittle’ or ‘fragile’ ) proteins or organisms are intolerant of mutation and are more likely to lose function or be driven to extinction by high mutation rates . Viruses that replicate via RNA intermediates using non-proofreading polymerases exhibit high mutation rates , suggesting that robustness should be particularly advantageous to them [3] . Indeed , under some conditions , viral populations that exhibit high robustness at the expense of high fitness might be favored over those that have high fitness but low robustness [4]–[6] . In other words , robustness/fragility might be a more potent selective force than fitness under some circumstances . While the robustness or fragility of RNA viruses has been investigated in several studies [4]–[9] , most reports characterize robustness by treating an entire viral genome as a single biological entity . However , viruses encode a variety of proteins that execute a range of functions to enable replication , and the genetic robustness of individual proteins is expected to vary within a given virus . Even within a single protein , individual domains may exhibit variation in robustness/fragility . Proteins that perform complex or multiple functions that are highly dependent on accurate structure ( e . g . enzymes ) , should tend to be more genetically fragile , i . e less tolerant of mutation , and exhibit greater sequence conservation than those that do not ( e . g . proteins that simply provide peptide binding sites to recruit other proteins ) . However , in their natural setting , i . e . a susceptible host , animal viruses often replicate in a hostile and changing environment , shaped partly by adaptive immune responses . Such immune responses can take the form of antibodies that target proteins displayed on the surface of the virion , or cytotoxic lymphocytes that can target epitopes in nearly any viral protein . Hence , at least some viral proteins , even those that should exhibit genetic fragility , are placed under strong evolutionary pressure to diversify in sequence , because they are targeted by adaptive immune responses . Under such conditions , otherwise fragile proteins might be expected to evolve higher robustness . However , the potential trade-off between robustness and fitness might impinge on each property , and it is unclear whether simple fitness , or the acquisition of robustness , would constitute the dominant selective force on a given viral protein in a natural setting . One such protein in which the competing needs to preserve function yet diversify sequence might conflict in a particularly acute manner , is human immunodeficiency virus type I ( HIV-1 ) capsid protein ( CA ) . As a critically important viral protein , it is placed under strong selective pressure to maintain its structure and perform several functions . Conversely , as a highly expressed , immunologically visible protein , CA also experiences a competing pressure to diversify its sequence . CA exists both as a domain within the Gag precursor polyprotein during particle assembly , and as an autonomous protein in the mature virion . Both Gag and CA are multifunctional proteins [10]–[13] . Specifically , during virion morphogenesis , Gag molecules assemble at the plasma membrane and drive the formation of roughly spherical immature virions , containing radially arrayed Gag molecules , that bud through the plasma membrane [14] , [15] . However , during and after viral budding , the viral protease ( PR ) is activated and catalyses the cleavage of Gag at five positions , causing profound morphological transformations [13] . In particular , the liberated CA protein forms a conical capsid that encapsulates the nucleocapsid-genomic RNA complex [16] , [17] . Like all retroviral CA proteins , a single HIV-1 CA molecule is composed of two domains: the N-terminal domain ( NTD ) comprised of 146 amino acids , and C-terminal domain ( CTD ) comprised of 85 amino acids . The NTD structure consists of an N-terminal β hairpin and 7 succeeding α helices , while the CTD has 4 α helices , and a C-terminal unstructured region of 11 residues [18]–[20] . The NTD and CTD are joined by an interdomain linker region ( residues 146–150 ) . Other noteworthy features of CA include the major homology region ( MHR residues 153–172 ) of the CTD , which is a highly conserved 20 amino-acid region found in all retroviruses [21] and the likely binding site of ABCE1 [22] , and a loop ( NTD residues 85–93 ) which binds the cellular protein cylophilin A [23] . The mature HIV-1 capsid consists of ∼1100 CA monomers assembled into a hexameric lattice with 12 pentameric declinations [16] , [24] , that are distributed in such a way that the viral capsid takes the form of a fullerene cone . Within the lattice , the NTD is primarily responsible for intra-hexamer contacts , while the CTDs form dimers that link adjacent hexamers [25] , [26] . However , interactions between the NTD and CTD also contribute to proper capsid formation [26]–[29] . In addition to its major structural role , CA is a key determinant of several other biological properties of HIV-1 . For example , HIV-1 capsid is the key determinant that enables HIV-1 to infect non-dividing cells [30] , [31] . Related work has identified genetic or physical interactions with host proteins karyopherin β transportin-3 ( TNPO3 ) , nucleoporin 153 ( NUP 153 ) , nucleoporin 358 ( NUP358 ) /RanBP2 and cleavage and polyadenylation factor 6 ( CPSF6 ) [32]–[36] . Moreover , interactions between HIV-1 capsid and cyclophilin A influence nuclear import and subsequent integration site selectivity , as well as replication efficiency , in a cell type dependent manner [37]–[40] . CA is a key target of intrinsic , innate and adaptive immune defenses . Specifically , CA is targeted by TRIM5α [41] , and may be detected by undefined sensors in dendritic cells [42] . HIV-1 CA also adapts under immune selection in vivo [43]–[46] . In particular , host CD8+ cytotoxic T lymphocyte ( CTL ) responses to HIV-1 infection are critical determinants of viral control [47] , [48] , and there is an association between Gag- and capsid-specific CD8+ T-cell responses and in vivo viral burden[49] . The emergence of viral ‘escape’ mutations in CA can result in higher viremia [50] , [51] . However , escape mutations often incur a significant fitness cost , which may be subsequently compensated for by secondary mutations in CA , or drive reversion when immune pressure is lifted [45] , [51] , [52] . Previous studies in which CA was mutated have aimed to elucidate the importance of particular domains , regions , and residues in CA functions . Those targeted mutagenesis studies largely relied upon insertion [53] , deletion [54] , or alanine or proline scanning [55] , [56] . Here , we took a different approach to investigate the genetic robustness of HIV-1 CA . Specifically , we describe the generation of a large randomly mutagenized library of CA sequences to simulate the natural process of random mutation that occurs during HIV-1 replication . Strikingly , we find that CA is extremely intolerant of nonsynonymous substitutions , with ∼70% of random single nucleotide substitutions leading to a >50-fold reduction in replicative fitness . We also determined the biological basis for this extreme genetic fragility and found that requirements imposed by the need to accurately and efficiently assemble a mature virion are largely responsible . Indeed , a subset of mutants were temperature sensitive ( ts ) , and the conditional non-viability of these mutants was always manifested during the formation of virions . Interestingly , fewer than half of the CA mutations that might be expected to occur in vivo , based on their high replicative fitness , were actually observed in natural populations . This finding suggests that CA sequence is constrained in vivo , not only by the need to maintain replicative fitness , but also by other unknown selective pressures . The mutational fragility of HIV-1 CA demonstrated herein is consistent with the relatively high overall degree of amino acid sequence conservation observed in natural populations , and potentially explains the apparently limited capacity of HIV-1 to evade immune responses directed against epitopes in CA .
To construct a library of random mutants of HIV-1 CA , a low fidelity PCR approach was used , as illustrated in Figure 1A . In this scheme , CA encoding sequences were amplified by error-prone PCR and cloned using a TOPO TA cloning Kit to generate a library with an estimated complexity of 1×104 clones . Plasmid DNA extracted from the pooled library was then digested , and CA sequences inserted into a replication competent proviral clone encoding EGFP in place of Nef ( pNHGcapNM accession:JQ686832 , referred to as WT or parental virus hereafter ) using unique NotI and MluI restriction sites , introduced by silent mutagenesis into sequences flanking those encoding CA ( Figure 1A ) . Thereafter , nearly 1000 individual colonies were picked and proviral plasmid DNA was extracted from individual mini-cultures . The mutagenized CA sequence was then sequenced for each clone , and after failed sequencing reactions or chromatograms indicating the presence of more than one template were removed , the resulting library consisted of 680 arrayed proviral plasmids . The distribution of all nucleotide changes ( Figure 1B ) , missense mutations ( Figure 1C ) , and nonsense and other mutations ( Figure 1D ) within the library was then determined . The PCR-mutagenesis conditions were optimized in pilot experiments so as to reduce the representation of non-mutated and heavily mutated CA sequences in the library . Consequently , the most frequent mutation in the library was a single nucleotide or amino acid substitution ( Figure 1B , C , D ) . To measure the genetic robustness of CA , we analyzed the ability of the 680 mutants to replicate . Specifically , replicative fitness was measured by means of a spreading infection assay , in which MT-4 cells were challenged with a low volume of virus-containing supernatant derived from transfected 293T cells ( equivalent to an MOI of ∼0 . 01 for the WT virus ) . Multiple rounds of the virus replication cycle were allowed to occur before halting the experiment when the WT virus had infected >50% of cells ( 72 to 80 hours later ) . Replicative fitness for each CA mutant was expressed as number of cells that were infected ( EGFP-positive ) , as a percentage of the number of cells that were infected by the WT virus ( Figure 2A , B ) . Additionally , to allow for the isolation of temperature sensitive mutants that could facilitate the subsequent characterization of the replication defects associated with CA mutations ( see below ) , this viability screen was carried out at two temperatures , 35°C ( Figure 2A ) and 39 . 5°C ( Figure 2B ) . Virtually all of the viral clones containing WT CA sequence , or only silent mutations , replicated at levels close to that of the starting construct . This indicated that the frequency with which defective viruses were generated by cloning artifacts or silent mutations was extremely low ( Figure 2A , B ) . Moreover , this analysis revealed CA to be extremely fragile , or intolerant of randomly introduced amino acid substitutions ( Figure 2A , B ) . If the cut-off for being viable is arbitrarily set at 2% of the parental virus replicative fitness , only 35% and 28% of the 135 clones that encoded single amino acid substitutions were viable at 35°C and 39 . 5°C respectively . These figures could overestimate the proportion of mutants that might be expected to be fit in vivo , as requirements for replication in an infected individual might be more stringent than in highly permissive MT-4 cells , and 2% of WT virus fitness in MT-4 cells is a rather generous cut-off figure for a designation of ‘viable’ . Increases in the number of amino acid substitutions further decreased the frequency of viable mutants; only 6% and 3% of the 125 clones containing two random amino-acid substitutions were viable at 35°C and 39 . 5°C respectively . Only 1 of 73 clones containing 3 random amino acid substitutions was viable , and none of the 45 library clones with four or five random substitutions could replicate at all . In total , the subset of mutants that had a unique single amino acid substitution within CA ( 135 mutants ) covered 102 ( 44% ) of the 231 CA residues ( for 33 CA residues there was more than one unique mutation at each position ) . To refine our estimate of CA robustness , two more assays were conducted at a natural temperature of 37°C , using only the panel of CA mutants that had unique single amino acid substitutions . First , in the MT-4 spreading assay , 40 mutant viruses ( 30% ) exhibited at least 2% of parental virus fitness , and are listed in Table 1 . Conversely , Table 2 contains the longer list of 95 mutants that had less than 2% of WT fitness in the spreading infectivity assay , and are considered non-viable . Single-cycle infectivity was also measured by transfecting 293T cells with proviral plasmids and measuring infectivity using a larger dose of virus ( equivalent to an MOI ∼1 for the parental clone ) and treating the MT-4 cells with dextran sulfate 16 h post-infection to limit replication to a single-cycle . Replicative fitness in spreading versus single-cycle assays was well correlated ( Table 1 ) , although most mutant viruses fared better in the single-cycle assay than in the spreading assay . This phenomenon possibly reflects the effect of transfection in the single-cycle assay , whereby overexpression of the viral proteins may suppress defects , as opposed to the spreading infection assay where more natural levels of viral gene expression are attained . Alternatively , the multiple rounds of replication in the spreading assay might amplify the effect of a defect in the single-cycle assay . Most of the viable CA mutants displayed attenuated infectivity , with 89% of single amino acid substitution causing a greater than 2-fold reduction in infectious virion yield . This suggests that it is difficult to improve upon the fitness of the parental CA , and that perhaps the parental CA sequence is close to a fitness peak . In previous analyses of panels of single amino acid mutants of RNA viruses , the distributions of mutational fitness effect ( DMFE ) values were bimodal in nature , with substitutions causing either lethality , or only minor decreases in fitness [4]–[9] . This phenomenon was observed to some extent for HIV-1 CA ( Figure 2A–D ) . However , the proportion of viable mutants was significantly smaller than has been observed in previous studies [4]–[9] , and the selective occurrence of mutants with only minor fitness defects was not prominent ( Figure 2A–D ) . A comparison of the Tables 1 and 2 reveals that different mutations at certain positions in CA , such as I2 , can result in markedly different fitness effects , suggesting that the type of residue change may , in some cases , be of some importance in determining fitness . However , inspection Tables 1 and 2 indicates that this was not typically the case , and the position within CA rather than the type of residue change appeared generally more important in determining the effect of mutations on fitness . Of the 231 amino acids that comprise CA , 56 . 3% lie within CA's 11 helices , 5 . 6% are in the N-terminal β-strand , 3 . 9% are in the cyclophilin A binding loop , and the remaining 34 . 2% are in other inter-helical ‘loops’ . These numbers can be used to provide context to the summary of mutant fitness values to indicate regions of particular genetic fragility in CA ( Table 3 , Figure 3A ) . Notably , the N-terminal β-strand , the cyclophilin binding loop , and interdomain linker appeared comparatively robust , with 70–100% of mutants in these regions retaining some viability . In contrast , helices were less tolerant of mutation , with a mere 16% of mutants in these domains retaining some viability , while the interhelical loops , where 39% of mutants retained viability , had intermediate robustness . Helices 2 , 5 , 6 , and 7 were particularly fragile , with all 25 mutations in these sequences causing non-viability . A graphical representation of these results , where the fitness of each mutant is plotted against its position in the linear CA sequence ( Figure 3A ) , confirms that regions of mutational fragility are non-uniformly distributed in CA , with mutations in NTD helices being extremely likely to cause inactivation ( Figure 3A ) . When displayed in the context of a capsid hexamer structure ( PDB: 3GV2 , Figures 3B , C ) , amino acids corresponding to viable mutants ( in green , Figure 3B ) appeared to occur preferentially in surface exposed residues . Conversely , mutations in the interior of the CA structure almost always generated non-viable mutants ( displayed in red , Figure 3C ) . This finding is particularly evident when the CA hexamer is viewed from the point of view of interior of the assembled capsid ( Figure 3B , C , rightmost diagrams ) , and reinforces the impression of an inner CA structure , or ‘core , ’ composed of helices that are particularly sensitive to mutation . Analysis of the solvent accessible surface area for individual mutated residues confirmed these findings . Specifically , mutated residues with solvent accessible surface areas of less than 50 Å2 ( an appropriate cutoff for surface exposure based on previous mutagenesis analyses [57] ) caused significantly greater fitness defects ( p = 0 . 003 ) . Viruses with such mutations and exhibited a mean fitness value of 2 . 8% of WT . Conversely , viruses harboring mutations of more surface exposed amino acids ( solvent accessible areas of greater than 50 Å2 ) exhibited a mean fitness value of 18% of WT . As outlined in the introduction , CA performs a number of functions in HIV-1 replication , including mediating the assembly of immature virions ( in the context of the Gag precursor ) and formation of a mature conical capsid . Moreover , a capsid of optimal stability is thought to be important during the uncoating step of the HIV-1 cycle in newly infected cells , and CA is also key for the import of the viral genome into the nucleus of newly infected cells . To determine which of these functions , if any , contributed the genetic fragility of HIV-1 CA , we performed a number of assays to elucidate the nature of the replication defects in constitutively or conditionally non-viable CA mutants . To facilitate the elucidation of the nature of the replication defects in CA mutants , we first focused on the subset of conditionally non-viable , ts mutants . At outlined in Figure 2 , initial determinations of mutant fitness were carried out at 35°C and at 39 . 5°C . While most CA mutants were approximately equally fit or unfit at both temperatures , a few exhibited a ts phenotype that was quite large ( Figure S1 ) . Some of the ts mutants ( circled in Figure S1 and listed in the materials and methods ) included double mutants . We therefore generated CA mutants that encoded each single amino acid substitution in isolation . Ultimately , this yielded eleven single amino-acid CA mutants with substantial ts replication phenotypes . Because the initial screen measured replicative fitness over multiple rounds of replication , it could not determine what specific step in the viral life cycle was impaired in the constitutively or conditionally non-viable mutants . We therefore determined the single-cycle infectivity of the ts CA mutants using virions generated in 293T cells at 35 , 37 , or 39 . 5°C , which were then used to infect MT-4 cells at 35 , 37 , or 39 . 5°C ( Figure 4A , B ) . Importantly , the WT control , pNHGcapNM , maintained a consistent viral titer regardless of production and infection temperatures ( Figure 4 ) . Most of the ts CA mutants had similar , or modestly reduced , infectivity as compared to the parental virus when virion production and infection was done at the permissive temperature ( 35°C ) . Strikingly however , single-cycle replication of all the ts mutants was inhibited at least 50 to 1000-fold when restrictive temperatures were applied during virion production ( Figure 4A , B ) . Conversely , all of the ts mutants were completely unaffected when restrictive temperatures were applied only during infection and not during production ( Figure 4A , B ) . This result indicated that the defects associated with conditionally non-viable ts CA mutants occurred exclusively and irreversibly , during or shortly after , particle production , and not during early steps of the viral life cycle ( e . g . uncoating or nuclear import ) . Analysis of Gag expression and processing by the WT virus and the ts CA mutants revealed that similar levels of cell-associated Gag precursor Pr55 and capsid p24 were present regardless of the production temperature ( Figure 5A ) . However , unlike the WT control , each of the ts CA mutant viruses generated 57% to 88% less extracellular virion-associated p24 at the restrictive temperature ( Figure 5A ) . Because it was possible that the mutations could have affected the CA protein stability or recognition by the p24 antibody , we also analyzed Gag expression , processing and particle generation by the ts mutants using an anti-p17 matrix ( MA ) antibody ( Figure 5B ) . This analysis yielded similar results . Specifically , there were approximately uniform levels of p17 in cell lysates for each of the ts CA mutants regardless of temperature . However , reductions in extracellular particulate p17 protein of 2 . 4–25-fold indicated that particle production was decreased by 59–96% for the ts mutants at the nonpermissive temperature . Although the 2 . 3 to 25-fold reduction in particle production observed for the ts CA mutants is smaller than the corresponding 50 to 1000-fold reductions in infectious virion yield , it is important to note that the majority of this reduction in infectivity ( specifically 57% to 96% ) was due to the inability of these mutants to efficiently generate particles . However , because the block in the generation of extracellular particles was not absolute , and there was residual generation of poorly infectious particles , it appeared that particle formation by the ts mutants was attenuated to varying degrees at the non-permissive temperature rather being than completely defective . The inability of the ts mutants to efficiently generate virions at the restrictive temperature is entirely consistent with the finding that it was the temperature during virion production , not inoculation , that determined the phenotype of all of the ts mutants we identified , and that all these mutations conferred defects that are manifested during virion morphogenesis . Careful inspection of both anti-CA and anti-MA blots of the ts mutants at the nonpermissive temperature revealed a slight abnormality in Gag processing . The parental virus appeared to generate a single ∼41 kDa band that reacted with anti-CA and anti-MA at both permissive and restrictive temperatures ( presumptively the p41 MA-CA-p2 processing intermediate , Figures 5A and 5B ) . Conversely , one or sometimes two additional protein species , of similar but not identical mobility to p41 were observed for each of the ts mutants , specifically at the restrictive temperature . Thus , for each of the ts CA mutants , attenuated particle formation was accompanied by perturbation of Gag processing , which may have contributed to the overall defects in particle yield and infectiousness . Notably , eliminating viral protease activity eliminated the temperature-induced reduction in particle yield for all eleven ts CA mutants ( Figure 5C ) . Thus , capsid mutations that caused temperature-dependent attenuation of particle formation do not do so prior to protease activation . This being so , and because inoculation temperature did not affect infectivity , it appeared that all eleven conditionally non-viable CA mutants have defects that are manifested during , and not before or after , virion morphogenesis . The above analysis of the eleven conditionally non-viable CA mutants suggested that requirements imposed during particle production , and not during any other phase of the viral life cycle , are responsible for the mutational fragility of HIV-1 CA . However , it was possible that the selection of conditionally rather than constitutively non-viable mutants could have biased this conclusion . Thus , we examined the ability of the larger set of 81 constitutively non-viable mutants to generate extracellular particles , using western blot assays ( Figure 6A ) . Most ( 74% ) of the constitutively non-viable mutants exhibited a greater than 5-fold reduction in the yield of extracellular particles . A minority ( 20% ) of these lethal mutations did not affect the magnitude of particle production ( less than 2-fold changes in particle yield ) , but instead resulted in the efficient generation of non-infectious particles ( Figure 6C ) . The remaining 6% of mutants gave intermediate phenotype with 2- to 5-fold reductions in particle yield . In a few instances , for example at residues K30 , M39 , and I115 , a change to one amino acid resulted in efficient generation of extracellular particles that were non-infectious , while a change to another amino acid resulted in attenuated particle production . Nonetheless , there did not appear to be any general trend in changes of amino acid properties that might cause mutants to efficiently generate non-infectious virions as opposed to being attenuated for particle generation . Nearly all of the mutants that yielded WT levels of non-infectious particles mapped to the inner core of the hexamer , in a manner that was similar to the general distribution of inactivating CA mutations ( compare Figures 6B and 3C ) . Of note , many constitutively non-viable mutants displayed evidence of perturbed processing , with the generation of additional ∼40 kDa Gag protein species , similar to observations with the conditionally non-viable mutants at the restrictive temperature ( Figure 5 A , B ) . Overall , these data reinforce the conclusion that the primary underlying cause of mutational fragility in HIV-1 CA is the need to mediate accurate and efficient particle assembly . To further examine the nature of the defects in conditionally and constitutively non-viable CA mutants , we employed electron microscopy ( EM ) to quantify and visualize particle morphogenesis . To examine the ts CA mutant mutants , 293T cells ( at 39 . 5°C ) were transfected with the parental or one of six randomly selected ts CA mutant proviral plasmids , plus a plamid expressing a modified Vpu-resistant human tetherin ( delGI , T45I ) [58] to retain virions at the plasma membrane and facilitate their visualization . Inspection of 150 cells for each ts CA mutant clearly showed these mutants generate far fewer total particles at the restrictive temperature ( Figure 7A shaded bars ) , with reductions ranging from 4- to 221-fold ( 75–99 . 5% fewer particles ) . Some , but not all particles generated by the ts CA mutants appeared morphologically normal , but there appeared to be a higher fraction of abnormal particles and budding structures as compared to the parental virus ( Figure 7B ) . However , the rarity with which the ts CA mutants generated discernable viral structures made it difficult to reach definitive conclusions about the morphological accuracy ( or otherwise ) of the particles that were formed . We also performed thin-section EM analyses of cells transfected with three randomly selected constitutively non-viable mutants ( K25I , L52F , and N195S ) that exhibited reduced extracellular particle yield ( Figure 7C , D ) . Again , this was done in the presence of tetherin ( delGI , T45I ) to facilitate particle imaging . This analysis revealed reductions in particle formation that were similar in character to those observed for the conditional ts CA mutants . Specifically , particle formation for these mutants was reduced ∼6 to 28-fold ( 82–96% fewer particles ) as compared to the parental virus . The small number of particles that were generated by the K25I and L52F mutants appeared morphologically similar to the parental virus , while the rare budding structures generated by the N195S mutant were obviously irregular ( Figure 7D ) . Interestingly , defects similar to that exhibited by N195S have been previously reported for CA mutations made at a nearby residue ( D197 ) [56] . As an additional control for these experiments , particle formation by a constitutively non-viable mutant ( G60W ) that gave high levels of non-infectious particulate p24 in the western blot assay was quantified by EM analysis . In fact , the G60W mutant generated particles that were as , or even more , numerous than the parental virus . Overall , the data acquired via EM analysis of particle formation broadly corroborated the estimates of particle formation efficiency obtained using western blot assays ( Figures 5A , 5B and 6A ) , and reinforce the notion that most randomly introduced CA mutations induced substantial attenuation , but not complete abolition , of particle formation . Because a fraction ( 20% ) of the constitutively non-viable CA mutants generated normal levels of extracellular particles that were non-infectious , we randomly selected three such mutants , ( R18G , G60W , and M215V ) for examination by thin-section EM ( Figure 7E ) . Each of these mutants exhibited morphologies that were indistinguishable from the WT virus . Additionally , cell free virions produced by these mutants , plus one additional non-viable mutant ( K30N ) that also exhibited efficient particle formation , were also examined by cryo-electron microscopy ( Figure 7F ) . This analysis again revealed particle morphologies that were similar to WT , with the possible exception of G60W , whose cores were more difficult to discern . The above data demonstrates that a large fraction of randomly introduced mutations impair the ability of HIV-1 CA to support efficient and accurate virion morphogenesis . To determine whether a similarly large fraction of CA mutants affected another critical function of CA , namely infection of non-dividing cells [31] , we screened all CA mutants that exhibited easily measurable infectivity in single-cycle infection assays for their ability to infect cells in the presence or absence of aphidicolin ( which arrests the cell cycle in S-phase ) . Only one of the 58 mutant viruses tested , N57S , exhibited substantially higher infectivity in the absence than in the presence of 2 µg/ml aphidicolin ( Figure 8 ) . A mutation at this position , N57A , has previously been reported to impose cell cycle dependence on HIV-1 infection [30] . Because only one of the 58 randomly introduced mutations tested affected the cell cycle independence of HIV-1 infection , these data suggest that CA sequence is constrained to a large extent by the need to accurately and efficiently assemble into functional viral particles , but to only a small extent by another key function , i . e . the need to infect non-dividing cells . Our measurements of in vitro fitness of HIV-1 CA mutants suggested that fitness should place tight constraints on the sequences of CA that are found in natural populations . Therefore , to compare the fitness of single amino acid HIV-1 CA mutants with their occurrence in vivo , we obtained a set of 1 , 000 CA sequences isolated from HIV-1 subtype B infected individuals , and determined the frequency with which mutations in our random mutant library were found in these naturally occurring sequences . A plot of the frequency with which CA mutations occur in natural populations against their measured in vitro fitness ( Figure 9A ) revealed that only mutants that exhibited a fitness of at least 40% of the WT CA sequence in our in vitro assays occurred at a frequency greater than 1% in natural populations . Curiously , however , the nineteen random mutants that exhibited >40% of parental virus fitness had bimodal distribution of occurrence in vivo . Some occurred relatively frequently ( i . e . in >3% of natural sequences Figure 9B ) , while others occurred extremely rarely ( <0 . 3% of natural sequences Figure 9C ) . That is to say that a subset of mutations that incur little or no apparent fitness cost appear to be largely absent from natural populations of subtype B HIV-1 . Consistent with their lack of effect on fitness , virtually all of these apparently innocuous , but rare , mutations occurred on the outer surfaces of the hexamer , and avoided the inner hexamer core ( Figure 9D ) . The existence of a subset of mutations that gave high fitness in MT-4 cells in vitro , yet were virtually absent in natural populations , suggested that some mutations may be selected against in vivo in a manner that was not revealed by our fitness assays . Previous studies have demonstrated that CA mutations can have cell-type dependent effects on HIV-1 infectiousness [38] , [39] , [59] , [60] , and it was therefore possible that these mutants might exhibit fitness defects that are manifested in natural target cells , but not in MT-4 cells . Thus , we performed replication assays in primary cell types using 8 apparently fit viruses containing mutations that occurred rarely in natural populations ( Figure 9C ) and 5 randomly chosen viruses containing mutations that occurred frequently ( Figure 9B ) . In fact , there was no significant difference between the ability of the rarely occurring or frequently occurring CA-mutant viruses to infect PBMC , primary CD4+ T cells or macrophages in short term ( quasi single-cycle ) infection assays ( Figure 10A , B , C ) . Furthermore , there was no difference in the capacity of the rarely and frequently occurring mutants to replicate in a spreading infection assay in PBMC ( Figure 10C ) . These results suggest that the rarity of apparently fit CA mutant viruses in natural HIV-1 subtype B populations is not due to differences in their capacity to replicate in primary cells . Rather , this finding suggests the presence of some unknown selective pressure that allows some intrinsically fit CA variants , but not others , to persist in vivo . To further examine the relationship between the impact of CA mutation on in vitro fitness and occurrence in natural HIV-1 subtype B populations , we next compared naturally occurring variability across the CA sequence ( as measured by Shannon entropy value ) with the propensity of CA domains to tolerate mutation . The expectation was that regions that are more robust ( Figure 3A ) should exhibit higher variability in natural populations . Like a previous analysis of the Shannon entropy of the N-terminal of subtype C HIV-1 [43] , this analysis revealed regions of high entropy in CA ( >0 . 4 ) while other regions were more conserved ( Figure 11A ) . There was a degree of accord between the ability of CA domains to tolerate mutation and Shannon entropy values . For example , there were shared regions of comparatively high robustness and high entropy , such as the cyclophilin binding loop and regions encompassing residues ∼5 to ∼35 , and ∼175 to ∼210 . Additionally , there were regions where especially low robustness coincided with regions of low variability in vivo , such as in CA helices 2 and 3 , and also in the MHR ( Figure 11A ) . However , there were also significant discrepancies between the Shannon entropy and robustness profiles across CA . For example , in helices 5 , 6 , and 7 ( from residues 101 to 145 ) , Shannon entropy values were often quite high , but fitness measurements revealed this region to be highly genetically fragile . This finding strongly suggests that selective pressures in addition to fitness act on the HIV-1 CA sequence in vivo . One such selective pressure is likely imposed by the adaptive immune system , particularly cytotoxic T-lymphocytes ( CTLs ) that can recognize peptides derived from HIV-1 CA . Indeed , several studies have demonstrated that mutations in CA can be driven by selective pressures imposed by CTLs [43]–[46] . Moreover , some of these studies have indicated that immune responses to particular epitopes in CA are protective , in that their appearance correlates with slower disease progression . For example , the oft studied protective HLA-B*57-restricted epitopes KF11 ( residues 30–40 ) and TW10 ( residues 108–117 ) , and the B*27 restricted epitope KK10 ( residues 131–140 ) , all occur in regions of apparently extreme genetic fragility ( Figure 11A , Table 2 ) . In fact , a comparison of the DMFE values for the 41 library mutations that occurred in so-called protective epitopes [specifically QW11 ( residues 13–23 ) , IW9 ( residues 15–23 ) , KF11 ( residues 30–40 ) , TL9 ( residues 48–56 ) , EW10 ( residues 71–80 ) , TW10 ( residues 108–117 ) , DA9 ( residues 166–174 ) , KK10 ( residues 131–140 ) ] with 94 library mutations that occurred outside these epitopes , suggested that protective epitopes occur in regions of CA that are more genetically fragile than remaining CA sequence ( Figure 11B ) . Indeed , the mean fitness value for mutations occurring within these ‘protective’ epitopes was 7 . 9% of WT while the mean fitness value for mutations occurring outside these epitopes was 15 . 1% of WT . The scope of the mutagenesis carried out in this study , in which almost half of the amino acids in CA were individually mutated , coupled with an analysis of the frequency with which these mutations are present in natural populations , should allow generalized predictions of the fate of randomly introduced single nucleotide ( Figure 12A ) or single amino acid ( Figure 12B ) substitutions into CA . In the case of random nucleotide changes ( Figure 12A ) , 22% should give synonymous changes that , in the great majority of cases , should not affect fitness ( Figure 2 ) . If there was no negative selection pressure in vivo other than fitness , then 10% of nucleotide substitutions are expected to yield a nonsynonymous change that is sufficiently replication competent ( 40% of in vitro WT fitness ) to be potentially capable of flourishing in natural populations . However , if , as suggested by the data in Figure 9 , it is true that only a fraction of mutations that yield variants with >40% of WT in vitro fitness are capable of persisting in vivo , then only 4% of all nucleotide substitutions are expected to yield non-synonymous changes that would flourish in a natural setting . If only nonsynonymous nucleoside substitutions are considered ( Figure 12B ) , then 87% are expected to be non-viable , with most mutations ( 59% ) inducing a >5-fold attenuation of particle formation and a smaller fraction ( 21% ) resulting in the generation of particles ( at least 20% of the level of WT ) that are non-infectious . Only 13% of nonsynonymous mutations are predicted to be sufficiently replication competent ( at least 40% of WT fitness ) to be potentially capable of flourishing in vivo , based on measurements of intrinsic fitness . Given the finding that only a fraction of intrinsically fit mutants actually persist in vivo , these data suggest that only 5% of all CA amino acid substitutions are actually expected to thrive in a natural setting .
The goal of this study was to generate a reasonably sized sample of random mutations in the CA protein that might arise naturally during HIV-1 replication , and examine their biological effects . In so doing , we could determine the genetic robustness of HIV-1 CA and correlate the effect of amino acid substitutions in vitro with their occurrence in natural viral populations . Moreover , such a large library constitutes a resource for investigating various functions and properties of the HIV-1 capsid . Our results uncover a rather extreme genetic fragility in the HIV-1 CA protein , with a large fraction ( ∼70% ) of individual , random amino acid substitutions resulting in non-viable viruses ( <2% of WT fitness ) . Tables 4 and 5 place these findings in context . Specifically , Table 4 summarizes all previous studies of randomly mutagenized HIV-1 proteins ( and domains thereof ) , as well as all other randomly mutagenized viral proteins , and full viral genomes . Table 5 lists the genetic robustness of selected non-viral proteins ( it is not inclusive of all random mutagenesis studies but we were unable to find any more fragile than those listed ) . Viruses ( particularly ss ( + ) RNA viruses ) not only have much higher mutation rates than cellular organisms , but also have comparatively low tolerance to mutation ( Table 4 , 5 ) [61] , [62] . Even when compared to viral genomes or proteins ( e . g . enzymes , Table 5 ) that exhibit low robustness , HIV-1 CA is more genetically fragile than any other protein or virus for which this property has been measured . The extreme mutational fragility of the HIV-1 CA may be related to the requirement that it maintain a structure that enables several different sets of interactions . Within the mature hexameric CA lattice , each CA monomer must interact , via distinct interfaces , with at least three other CA monomers , and adopt a rather tightly constrained position with respect to its internal structure as well as other CA monomers within a hexamer [26] . However , some CA monomers , that have precisely the same amino acid sequence , must be capable of adopting slightly different structures and positions relative to other CA monomers in order to form the pentameric declinations that enable closure of the CA lattice [16] , [24] . Furthermore , a different set of CA:CA interactions are likely required during the formation of immature virions [15] . In addition to these constraints , the strength of the various CA:CA interactions in each of the different configurations probably needs to be finely tuned so that transitions from immature to mature virions during assembly , and appropriately timed disassembly , of the capsid can be accomplished in newly infected cells . Indeed , previous findings have suggested that appropriate capsid stability is required for HIV-1 infectiousness [63] . Additionally , CA interacts with a variety of host proteins , including cyclophilin A [23] , ABCE1 [22] and possibly with nuclear pore complex proteins [37] and CPSF6 [34]–[36] . In summary , the number of CA interactions and functions , the requirement for CA to adopt several different structures , and transition between them in an ordered manner at multiple , critical stages of replication , likely places tight constraints on sequence and reduces the number of possible mutations that lack a major fitness cost . Although lethal CA mutations could in principle affect any one of multiple steps in the HIV-1 life cycle , we found that most constitutively non-viable mutants exhibited attenuated particle formation . Moreover , all of the conditionally non-viable ts mutants exhibited attenuated particle formation at the non-permissive temperature . Interestingly , these deficits , at least for the conditionally non-viable mutants , could be rescued by inactivation of the viral protease . These results indicate that the critical role of CA in mediating efficient and accurate particle assembly , especially during Gag proteolysis , is the principal cause of its genetic fragility . How particle formation , budding and maturation are correctly orchestrated is not yet fully understood . Over a relatively short period of time ( 5–6 minutes ) [64] , thousands of Gag and Gag-Pol molecules coalesce and must be cleaved and rearranged through regimented and dynamically linked processes . Previous work has demonstrated that incorrect , premature activation of the HIV-1 protease can prevent particle formation [65] . Moreover , other perturbations of Gag ( e . g . partial nucleocapsid deletion ) can attenuate particle formation in a manner that is rescued by inactivation/inhibition of the viral protease [66] , [67] . One idea that could explain these findings is that delayed particle formation may result in more extensive proteolytic cleavage of Gag early in the assembly process , thereby attenuating particle formation [66] , [67] . The finding that many of the constitutively non-viable and all of the conditionally non-viable CA mutants showed evidence of perturbed Gag processing with additional ∼40 kDa protein species is consistent with the idea that CA mutations could similarly delay particle formation , resulting in mistiming of Gag proteolysis relative to budding and , therefore , protease dependent attenuation of particle formation . Within this framework , it is easier to understand how even subtle perturbation of Gag structure or multimerization caused by single amino acid substitutions could disrupt the fragile process of HIV-1 assembly , and thus how critical a role CA plays in orchestrating virion morphogenesis . The high frequency with which mutations in the N-terminal domain of CA result in diminished particle production is intuitively at odds with earlier findings that the N-terminal domain of CA is dispensable for particle formation [68] . It has already been hypothesized that a misfolded N-terminus ( caused by small changes in CA ) could be more disruptive to assembly processes than deletion [68] . Even without overt misfolding , the presence of a CA NTD that does not pack correctly or efficiently into a hexagonal lattice is likely to be more disruptive than NTD removal to other interactions that are essential for particle formation ( e . g . CA CTD-CTD interactions ) . Previous large-scale mutagenesis studies of the retroviral CA proteins have employed insertional mutagenesis of MLV CA ( using 12 amino acid peptide sequences ) [69] , [70] and alanine substitution of surface exposed residues in HIV-1 CA [56] . The fact that 12 amino acid insertions into MLV CA were uniformly lethal is consistent with the notion that retroviral CA proteins in general are sensitive to mutation , but insertions are obviously more likely to perturb CA structure than single amino acid substitutions . Indeed , a study of MLV CA mutants revealed that only 6 of 15 single amino acid substitutions ( 40% ) in the central portion of CA were lethal [71] . Interestingly this study found that the most frequent phenotype associated with single amino acid substitutions in MLV CA was the generation of non-infectious particles rather than impaired particle formation , perhaps suggesting a difference in the properties of MLV and HIV-1 CA proteins . Previously , the most extensive mutagenesis study of HIV-1 CA [56] generated 48 alanine-scanning mutants that targeted surface exposed residues , of which 44% were non-viable . This study also found that single CA mutations in both the CTD and NTD can cause attenuated or aberrant particle formation , as well as the generation of morphologically normal virions that are non-infectious [56] . The difference in the fraction of mutations that were lethal in [56] versus our study is consistent with our finding that viable CA mutants tended to be confined to surface-exposed residues while mutation of the inner core of the CA structure was rarely compatible with viability . Most notably , we found that sequences encoding CA helices , particularly those in the NTD , were especially genetically fragile . These sequences form the fundamental structure of the CA hexamer , and in some cases ( e . g . helices 1 , 2 and 3 ) , participate in CA:CA interactions within CA hexamers [26] . Mutations in CTD helices , and in the interhelical loops ( e . g . the CypA binding loop ) were less likely to cause lethal defects . This comparative tolerance of mutations in some portions of CA appears to correlate , at least partially , with conformational variation in crystal structures of CA hexamers [26] . In summary , CA NTD helices appear to be highly structurally constrained and particularly sensitive to mutations that might perturb the shape or stability of the CA NTD monomer or the CA hexamer . By correlating our fitness measurements with an analysis of the frequency with which mutations occur in natural populations , we hoped to make observations that neither approach alone could yield , and make predictions about the outcomes of hypothetical mutations . For example , all of the mutations that were present in our library and occurred at a frequency of greater than 3% in natural populations ( n = 9 ) were quite fit in vitro ( >40% of WT fitness ) . Conversely , all mutants that exhibited a fitness of <20% of WT ( n = 106 ) occurred at a frequency of <1% in natural populations . This finding indicates that there is some correlation between fitness and occurrence in natural populations , and suggests that below an ∼40% fitness threshold , mutations rarely confer a selective advantage within a human host . However , the correlation between mutant fitness and natural occurrence was clearly imperfect , indicating the impact of selective pressures other than fitness on naturally occurring sequences . Specifically , there was a lack of concordance in fitness and Shannon entropy profiles in some portions of the CA sequence , and it was striking that more than half of the mutations that had little or no impact on fitness , did not occur in natural populations ( ‘fit but rare’ mutants ) . This finding suggests the absence of a selective pressure driving the emergence of ‘fit but rare’ mutants , or perhaps the presence of a selective pressure against their emergence . Importantly , we were unable to demonstrate any defect in the ability of fit but rare CA mutants to infect primary cells , including primary T-cells and macrophages . One possible explanation for our findings is that fit but rare mutants are purged from natural populations because they confer sensitivity to adaptive or innate immune responses . Overall , our findings suggest that only ∼5% of all possible amino acid substitutions in CA result in viruses that can flourish in vivo . An obvious candidate for selective pressure that induces sequence variation in regions of high genetic fragility , and could potentially suppress ‘fit but rare’ mutants is that imposed by CTLs . Indeed , a number of examples of CTL escape mutations in protective epitopes are characterized by ( i ) their predictability ( i . e . recurrence of the same mutation in different infected individuals ) [72] , ( ii ) sometimes by their impact on viral fitness [45] , [73] , and ( iii ) co-occurrence with compensating mutations , suggesting an impact of epistatic interactions that alleviate compromised fitness induced by primary escape mutations [43] , [44] . All of these findings are consistent with , and indeed explained by , the finding that CA is highly genetically fragile , and suggest that the number of possible mutations that enable escape from CTL responses , with retention of fitness , is small . However , in considering these arguments , it is important to recognize that even though our sample of CA mutants should be representative of random variation , it is only a sample of the possible mutations that can arise in CA in natural populations . Analyses in cell culture suggest that a single cycle of HIV-1 replication ( incorporating transcription and reverse transcription steps ) generates 1 . 4 to 3 . 4×10−5 mutations per nucleotide [74] , [75] . For the 231 aa CA protein , this would result in approximately 0 . 0097–0 . 023 errors per replication cycle , or an amino acid substitution about once or twice per one hundred replication events . Importantly , the volume and rate of HIV-1 replication in vivo [76] means that every amino acid substitution that is accessible by a single nucleotide substitution is generated in nearly every infected individual , many times , every day . Thus , selective pressures imposed in vivo can act on an enormous pool of possible variants . The discrepancies between the fitness and Shannon entropy profiles in Figure 11 suggest that even in CA domains in which we found few or no fit mutants in our sample of 135 mutants , there do exist rare variants that can thrive in vivo , even if sometimes they exhibit compromised fitness or require the co-occurrence of compensating mutations . Previous work has indicated that the CTL responses to Gag , rather than other HIV-1 proteins , correlate best with reduced viral burden during HIV-1 infection [49] . Notably , so-called ‘protective’ CTL epitopes [44] tended to occur in regions of CA that were even more intolerant of mutation than CA as a whole . Additionally , Gag is among the most abundant viral proteins expressed in HIV-1 infected cells . Its abundance , and the comparative lack of plasticity in CA , likely both contribute to the relatively favorable clinical outcome in patients that mount strong immune responses to the Gag protein [49] . Nevertheless , it appears that the ability of viral populations to explore essentially all possible amino acid substitutions , partly overcomes the genetic fragility of CA and enables at least some degree of escape from immune responses , even those that target the most vulnerable areas of CA . Understanding the fitness effects of mutations in CA may facilitate drug or vaccine design . For example , one recent study done in the context of a possible vaccination strategy highlighted the observation that mutations in the most conserved residues in CA are not necessarily associated with the highest fitness costs [77] , as might otherwise be assumed . Thus , a better understanding of the constraints under which CA evolves may prove beneficial . Additionally , while CA has not yet been exploited as a target for drugs in the clinic , recent findings have indicated that it is possible to do so , at least in principle [78]–[82] . The genetic fragility exhibited by CA should reduce , but clearly not eliminate , the opportunities for drug resistance to occur , and it would be especially desirable to target CA domains that exhibit the greatest fragility . In spite of this , resistance to one recently identified CA-targeted inhibitor arises in vitro through mutations in regions of CA that exhibit high fragility [79] . Again , it appears that the sheer number of variants that can be ‘tested’ in natural viral populations can , at least sometimes , overcome the lack of robustness of HIV-1 CA . Nevertheless , the high genetic fragility exhibited by CA should facilitate attacks on HIV-1 through vaccination or therapy .
A mutagenized CA library was generated using the Genemorph II random mutagenesis kit ( Agilent ) using the following oligos 5′-GTA AGA AAA AGG CAC AGC AAG CGG CCG CTG -3′ and 5′- CTT GGC TCA TTG CTT CAG CCA AAA CGC GTG-3′ . The PCR product was cloned using a TOPO TA Cloning Kit and plasmid DNA was extracted from approximately ∼1×104 pooled , insert-positive colonies . After sequencing the amplicon from 20 clones to obtain a preliminary estimate of the mutagenesis frequency , this pooled plasmid DNA was digested using NotI and MluI and the CA-library insert subcloned into the pNHG ( JQ585717 ) derivative pNHGCapNM ( JQ686832 , Figure 1A ) . The NotI and MluI digested proviral plasmid was prepared in advance and generated >200-fold fewer colonies when ligated without insert than when ligated with insert . Furthermore , representative restriction digests indicated that aberrant restriction patterns , perhaps due to recombination , occurred at a frequency of only 1–2% of clones . Proviral plasmid DNA was extracted from individual cultures derived from 1056 colonies and subjected to sequencing and further analysis . Proviral plasmid DNA from the single mutants listed in tables 1 and 2 was freshly isolated , re-sequenced and analyzed by NotI and MluI restriction digest . The pNL4-3 Pr- plasmid is derived from pNHGCapNM and pNL4-3 ( NIH AIDS Research and Reference Reagent Program , Catalog No . 114 ) , and has NotI and MluI sites flanking CA as well as a point mutation in the protease active site , as has been previously described [83] . CA sequences harboring the temperature sensitive mutations were transferred from the pNHGcapNM vector to pNL4-3 Pr- by digestion with NotI and MluI . The adherent human cell lines , 293T and TZM-bl , were maintained in DMEM supplemented with 10% fetal calf serum ( FCS ) and gentamicin . Suspension MT-4 cells were maintained in RPMI with 10% FCS and gentamicin . For transfection experiments , 293T cells were plated at 2 . 5×104 cells/well in 96 well plates or 1 . 5×105 cells/well in 24 well plates . To measure the effect of CA mutations , and their temperature sensitivity , transfections were done the following day using polyethylenimine ( Polysciences ) , and either 100 ng of the WT or mutant NHGCapNM plasmids described above ( for 96 well plate experiments ) or 500 ng ( for 24 well plate experiments ) , or 500 ng of the pNL4-3 Pr- mutants ( all in 24 well plate experiments ) . Plates for all transfection experiments were placed at 35 , 37 or 39 . 5°C , as indicated , immediately upon addition of transfection mixture . PBMCs , CD4+ T cells , and macrophages were isolated from buffy coats ( from anonymous healthy blood donors and were purchased from the New York Blood Center ) using a Ficoll gradient . Primary CD4+ T cells were extracted using a RosetteSep Human CD4+ T-cell enrichment cocktail . Macrophages were isolated by adhesion to plastic and treatment with 100 ng/ml of granulocyte/macrophage-colony stimulating factor ( GM-CSF ) for 96 hours prior to infection . All primary cells were maintained in RPMI supplemented with 10% FCS , penicillin/streptomycin , and L-glutamine . Activation of PBMCs and CD4+ T cells was achieved by addition of phytohemagglutinin ( PHA ) at 5 µg/ml for 72 hours prior to infection , with addition of 25 U/ml of interleukin-2 at the time of infection . 293T cells were transfected with the proviral plasmids and given fresh medium 16 hrs later . At ∼40 h post-transfection , cell supernatants were collected and filtered ( 0 . 22 µm ) . Single-cycle infectivity was measured using MT-4 cells that were seeded in 96 well plates at 3×104/well and inoculated with a volume of filtered supernatant that was equivalent to an MOI of ∼1 for the WT viral clone . Dextran sulfate at 100 µM was added 16 hrs later to limit replication to a single cycle , and cells were fixed in 4% PFA 48 hrs after infection . Alternatively , harvested supernatants were subjected to a low-speed spin and a freeze-thaw cycle before addition to MT-4 or TZM-bl cells in 96 well plates , in the presence or absence of aphidicolin ( 2 µg/ml ) or at 35 , 37 , or 39 . 5°C , as indicated in figure legends . For spreading replication assays , MT-4 cells were inoculated with a volume of filtered supernatant that was equivalent to an MOI of ∼0 . 01 for the WT viral clone and fixed in 4% PFA at 72–80 hrs post-infection . FACS analysis for all infectivity and replication assays was carried out using a Guava EasyCyte instrument . The following mutations , some of which were part of either double or triple mutants , occurred in mutants that exhibited temperature sensitivity in spreading replication assay ( Figure S1 ) : S4C , L189M , I91F; Q6P , A78V; I15M , A78T; S16T , T48A; S33C , A92V; I91N , D163E; I91V , I124T; M96T; R100S , Q112L; R132G; R167Q; M214L; K227I . CA mutants encoding each individual substitution were generated , as necessary , prior to selection of the 11 ts mutants for analysis in Figures 4 and 5 . Infections of primary cells required different conditions . For single-cycle infections of PBMCs and CD4+ T cells , 0 . 1–1×106 cells were inoculated using virus generated from 293T cells , as described above , for the 11 rarely occurring mutants described in Figure 9C , the WT virus , and 4 randomly selected frequently occurring mutants from Figure 9B ( H87Q , I91V , E98D , T200S ) at an MOI of ∼5 , and were fixed in 4% PFA 36 hours later . For infection of macrophages , VSV-pseudotyped virus was generated in 293T cells for the same 16 mutants plus WT virus . Macrophages were infected at an MOI of ∼4 and fixed approximately 72 hours later . For spreading infection assays in PBMC , 1×106 cells were infected at an MOI of 0 . 1 , and cells fixed at the time points indicated in Figure 10C . All MOIs used for primary cells were reference values derived from titrations on MT-4 cells . Cell lysates and virions ( pelleted by centrifugation through 20% sucrose ) were resuspended in SDS sample buffer and separated by electrophoresis on NuPage 4–12% Bis-Tris Gels ( Novex ) . Proteins were subsequently blotted onto nitrocellulose membranes , probed with a primary anti-HIV p24 capsid antibody ( 183-H12-5C ) or anti-HIV p17 antibody ( VU47 Rabbit anti-p17 [84] ) , and then probed with a goat anti-mouse/anti-rabbit IRDye® 800CW secondary antibody ( LI-COR Biosciences ) . A LI-COR Odyssey scanner was used to detect and quantify fluorescent signals . A minimum of 3 separate Western blots was produced for each temperature sensitive mutant , including those in the protease negative background , and representative blots are shown . HIV-1 capsid hexamer structural analysis was done using MacPyMOL , with PDB reference 3GV2 [26] . Solvent accessibility surface area of residues was determined using UCSF Chimera . To prepare samples for thin-section electron microscopy , 293T cells were seeded in 6-well plates at 0 . 8×106 cells per well in duplicate . Transfections were done the following day with the addition of polyethylenimine to 2 µg of the WT or mutant NHGcapNM plasmids ( S33C , T48A , M96T , Q112L , R132G , L189M , G60W , K25I , L52F or N195S ) plus 0 . 5 µg of modified human tetherin ( delGI , T451 that is resistant to antagonism by HIV-1 Vpu [58] ) to enhance visualization of virions at the plasma membrane . Immediately after addition of transfection mixture , plates were placed at either 37 or 39 . 5°C , as indicated . After 48 hrs , supernatant was removed and cells were fixed in a solution of 2% paraformaldehyde ( PFA ) , 2 . 5% glutaraldehyde . One set of cells was then analyzed by FACS to compare transfection efficiencies ( which ranged from 23% for WT to 32% for L189M ) . The other set were fixed with 2 . 5% glutaraldehyde and 1% osmium tetroxide , and stained with 2% aqueous uranyl acetate . Fixed and stained cells were harvested into PBS and pelleted through 1% SeaPlaque agarose ( Flowgen ) at 45°C . The agar was set at 4°C and the cell pellets were cut into ∼2 mm cubes , which were dehydrated through a graded alcohol series and infiltrated with TAAB 812 embedding resin . After polymerisation , thin sections ( 120 nm ) were cut and examined in a JEOL 1200 EX II electron microscope . Numbers of virus particles associated with 150 randomly selected cells were counted for each sample . To isolate virions for cryo-electron microscopy , 10 cm plates of approximately 4×106 293T cells were transfected with 7 µg of the indicated plasmids ( NHGcapNM , R18G , K30N , G60W or M215V ) using polyethylenimine . After 48 hours , supernatant was collected and filtered and virions were pelleted by centrifugation at 14000 rpm through 20% sucrose . Virions were fixed in 10 µl of 2% PFA , 2 . 5% glutaraldehyde solution . Fixed aliquots of 3 ul of each sample were loaded onto freshly glow-discharged c-flat holey carbon grids ( CF-22-4C , Protochips , Inc . ) held at 4°C and 100% humidity in a Vitrobot vitrification robot ( FEI ) . Grids were blotted for 4 s prior to being frozen by plunging into a bath of liquid nitrogen-cooled liquid ethane . Vitrified specimens were imaged at low temperature in a JEOL 2200 FS cryo-microscope equipped with Gatan 626 cryo-stages . Low dose ( 10 e/Å2 ) , energy-filtered images ( slit width , 20 eV ) were recorded on a Gatan ultrascan 16-megapixel charge-coupled-device camera at a magnification of 50 , 000× . A set of 1 , 000 HIV-1 subtype B sequences isolates was obtained from the Los Alamos HIV sequence database ( www . hiv . lanl . gov/ ) . All sequences were sampled from distinct infections between from 1980 and 2009 . To minimize risks of sampling biases , multiple sequences from known transmission clusters were excluded . Sequences with frameshift mutations or stop codons that were likely to represent nonfunctional viruses or poor quality sequencing were excluded . Sequences were aligned using MUSCLE [85] , and PERL scripts were used to examine genetic variation in the resulting sequence alignment . An information-theoretic measure of diversity ( Shannon's entropy ) [86] was applied to quantify the amount of amino acid variation at each position in capsid .
|
The HIV-1 capsid protein ( CA ) is absolutely essential for viral replication and there is , therefore , intense evolutionary pressure for HIV-1 CA to conserve its functions . However , HIV-1 CA is also a key target of the host immune response , which should provide evolutionary pressure to diversify CA sequence . Genetic robustness , or fragility , is defined as the ability , or lack thereof , of a biological entity to preserve function in the face of sequence changes . Thus , it should be advantageous to HIV-1 CA to evolve genetic robustness . Here , we present the results of extensive , random mutagenesis of single amino acids in CA that reveal an extreme genetic fragility . Although CA participates in several steps in HIV-1 replication , the biological basis for its genetic fragility was primarily the need to participate in the efficient and proper assembly of mature virion particles . The extreme genetic fragility of HIV-1 CA may be one reason why immune responses to it correlate with better prognosis in HIV-1 infection , and suggests that CA is a good target for therapy and vaccination strategies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"immunodeficiency",
"viruses",
"virology",
"biology",
"microbiology"
] |
2013
|
Extreme Genetic Fragility of the HIV-1 Capsid
|
Epstein-Barr Virus ( EBV ) is an oncogenic γ-herpesvirus that capably establishes both latent and lytic modes of infection in host cells and causes malignant diseases in humans . Nuclear antigen 2 ( EBNA2 ) -mediated transcription of both cellular and viral genes is essential for the establishment and maintenance of the EBV latency program in B lymphocytes . Here , we employed a protein affinity pull-down and LC-MS/MS analysis to identify nucleophosmin ( NPM1 ) as one of the cellular proteins bound to EBNA2 . Additionally , the specific domains that are responsible for protein-protein interactions were characterized as EBNA2 residues 300 to 360 and the oligomerization domain ( OD ) of NPM1 . As in c-MYC , dramatic NPM1 expression was induced in EBV positively infected B cells after three days of viral infection , and both EBNA2 and EBNALP were implicated in the transactivation of the NPM1 promoter . Depletion of NPM1 with the lentivirus-expressed short-hairpin RNAs ( shRNAs ) effectively abrogated EBNA2-dependent transcription and transformation outgrowth of lymphoblastoid cells . Notably , the ATP-bound state of NPM1 was required to induce assembly of a protein complex containing EBNA2 , RBP-Jκ , and NPM1 by stabilizing the interaction of EBNA2 with RBP-Jκ . In a NPM1-knockdown cell line , we demonstrated that an EBNA2-mediated transcription defect was fully restored by the ectopic expression of NPM1 . Our findings highlight the essential role of NPM1 in chaperoning EBNA2 onto the latency-associated membrane protein 1 ( LMP1 ) promoters , which is coordinated with the subsequent activation of transcriptional cascades through RBP-Jκ during EBV infection . These data advance our understanding of EBV pathology and further imply that NPM1 can be exploited as a therapeutic target for EBV-associated diseases .
EBV is a human γ-herpesvirus that infects both epithelial cells and B lymphocytes , and it has been implicated in several human malignancies , including Burkitt's lymphoma ( BL ) , Hodgkin's lymphoma , lymphoproliferative disease in immune-suppressed patients , nasopharyngeal carcinoma ( NPC ) , and some cases of gastric cancer . The primary infection of B lymphocytes by EBV in vitro readily leads to the establishment of immortalized lymphoblastoid cell lines ( LCLs ) via the expression of a unique set of viral genes , including six EBV nuclear antigens ( EBNAs ) , three latency-associated membrane proteins ( LMPs ) , Bam A rightward transcripts , and small non-coding RNAs [1] . Among the EBV latency gene products , EBNALP , EBNA1 , EBNA2 , EBNA3A , EBNA3C and LMP1 are critical for LCL cell transformation and maintenance . During the initial stage of B cell infection by EBV , the W promoter ( Wp ) is exclusively employed to drive the transcription of EBNA2 and EBNALP genes , which are produced by the alternative splicing process . Switches in the usage of Wp to the C promoter ( Cp ) and activation of Cp by expression of EBNA2 and EBNALP subsequently leads to the concomitant expression of other EBNAs and LMPs that are essential for the latent infection and persistence of EBV {for reviews , see [1]} . The activation of cellular and viral genes by EBNA2 is coordinated with its biological role in supporting EBV-mediated conversion of resting B cells into LCLs [1] . In general , EBNA2 engages in transcription activation through interactions with cellular DNA-binding proteins , including RBP-Jκ , PU . 1 , and AUF1 , which are tethered to the cognate responsive elements [2] , [3] , [4] , [5] . The C-terminal acidic domain ( AD ) of EBNA2 recruits the basal transcription machinery , which then performs a major role during transcription activation by forming a transcriptional pre-initiation complex ( PIC ) [6] , [7] , [8] , [9] . The N-terminus harbors the second activation domain , which plays another role in promoter up-regulation and LCL cell maintenance [10] , [11] . Furthermore , the survival motor neuron protein ( SMN ) , co-activators , such as p300/CBP , PCAF , p100 and SKIP1 , and the hSWI/SNF chromatin remodeling complex have all been shown to facilitate EBNA2-dependent transcription via interactions with EBNA2 [12] , [13] , [14] , [15] , [16] , [17] . Although the mechanism by which EBNA2 activates its target genes has been intensively studied over decades , the host factors that are required in addition to the basal transcription machinery and the transcription co-activators that are involved in EBNA2-mediated transcription remain poorly understood . Nucleophosmin ( NPM1 , also known as protein B23 , NO38 , or numatrin ) was originally identified as an abundant nuclear phosphoprotein that resides in the nucleolus . It has attracted significant attention for its multifunction capabilities in nuclear shuttling , DNA-histone and nucleosome assembly , biogenesis of ribosomal RNA , regulation of cell growth , DNA repair , cell proliferation , and transformation [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] . The fact that NPM1 is frequently mutated , rearranged , and overexpressed in hematological disorders has led investigators to propose that NPM1 could be a proto-oncogene [21] . By contrast , NPM1 interacts with HDM2 to protect p53 from degradation , and its absence results in mislocalization and protein instability of the tumor suppressor ARF , suggesting an additional role for NPM1 in tumor suppression [26] , [27] . At the moment , NPM1 is believed to employ sophisticated mechanisms to engage in oncogenic and tumor suppressor pathways [28] . By acting as a molecular chaperone , NPM1 is able to modulate chromatin assembly and facilitate the DNA binding of the primary transcription factors to their responsive elements , which consequently promotes the transcription of their target genes [22] , [25] , [29] . In this study , we identified NPM1 as one of the cellular proteins bound to EBNA2 using a protein affinity pull-down assay followed by liquid chromatography-mass spectrometry ( LC-MS/MS ) analysis . The physical interaction mediated by EBNA2 and NPM1 was extensively characterized , which was proposed as the major determinant of PIC assembly . Finally , we provided striking evidence to show that ATP-charged NPM1 does indeed act as a chaperone to escort EBNA2 to couple with RBP-Jκ at the LMP1 promoter in order to initiate the subsequent EBNA2-dependent transcription , leading to the establishment and maintenance of EBV latency in B cells .
To explore the mechanism of EBNA2-mediated transcription in EBV-transformed B cells , we initially aimed to investigate the protein-protein interaction map of EBNA2 . We expressed and purified GST and GST fusion proteins with overlapping EBNA2 open reading frames ( ORFs ) ( GST- E2s ) , including amino acids ( aa ) 1–103 , 96–210 , 200–334 , 300–432 , and 426–465 ( AD ) , which covered almost the entire EBNA2 , from E . coli . Affinity matrices were used to pull down cellular proteins from IB4 cell lysates ( Figure 1 A–C ) . GST-E2 300–432 pulled down a unique protein band with a molecular weight approximately 33 kD ( Figure 1B ) . LC-MS/MS identified this band as NPM1 with peptide sequences covering 19% of the NPM1 ORF ( Figures S1 A and B ) . Proteins pulled down by the same bait were followed by immunoblotting with the NPM1-specific antibody , which further validated this interaction ( Figure 1C ) . To characterize the physical interaction mediated by EBNA2 and NPM1 in a physiologically relevant environment , we performed co-immunoprecipitation ( co-IP ) and immunofluorescence ( IF ) confocal microscopy analyses . In IB4 LCL , NPM1 and EBNA2 antibodies were used to immunoprecipitate approximately 0 . 8% of endogenous EBNA2 and NPM1 , and the protein complex formed by EBNA2 and NPM1 was not affected by RNaseA treatment ( Figure 1D ) , excluding the possibility that cellular RNAs might act as an intermediate bridge to link the two proteins together . Furthermore , EBNA2 and NPM1 mostly co-localized in the nucleoplasm of the lymphoblastoid cells ( Figure 1E , upper panel ) . HeLa cells transfected with both GFP-NPM1 and EBNA2 expression vectors also showed a pronounced nuclear co-localization ( Figure 1E , lower panel ) , indicating a substantial physical interaction mediated by EBNA2 and NPM1 that took place in vivo . To identify the core of the NPM1-binding domain located within EBNA2 aa 300–432 , GST and GST-E2 fusions that contain small portions of the EBNA2 ORF , including aa 300–335 , 335–360 , 300–360 , and 360–432 , were next used in a protein affinity pull-down assay ( Figure 2A ) . Approximately 2 . 5% of endogenous NPM1 was pulled down by GST-E2 300–360; however , neither GST nor the remaining GST-E2s was shown to possess any detectable NPM1 binding potency . GST-E2 300–360 was further utilized as the protein matrix to pull down ectopically expressed flag-tagged NPM1 ( FNPM1 ) and five truncated deletion mutants , including C1 ( 1–257 ) , C2 ( 1–194 ) , C3 ( 1–120 ) , N1 ( 120–294 ) , and N2 ( 194–294 ) [30] . Among the exogenous proteins , including FNPM1 and its mutants , approximately 2% of FNPM1 , C1 , C2 , or C3 was associated with GST-E2 300–360 , whereas neither N1 nor N2 was pulled down by the same protein bait ( Figure 2B ) . The N-terminus of NPM1 contains highly conserved amino acid residues that are typically found in members of the nucleophosmin family , which have been implicated in homotypic interactions [23] . In response to this information , the transfected flag-tagged oligomerization-defective mutants L102A and G105A and another oligomerization-competitive mutant , S106A [23] , were subjected to additional co-IP and protein affinity pull-down analyses . In each transfection assay , plasmid-expressed FNPM1 or three point mutants were expressed at similar levels and efficiently immunoprecipitated by the flag-specific antibody . Approximately 2% of endogenous NPM1 was co-immunoprecipitated with FNPM1 and S106A , while fairly small and barely detectable amounts of this protein were co-immunoprecipitated with L102A and G105A , respectively ( Figure 2C; upper panel ) . The interaction between exogenous FNPM1 and endogenous NPM1 was not affected by ectopically expressed EBNA2 ( Figure 2C; lower panel ) . Interestingly , our result showed that neither of the ectopically expressed NPM1 mutants was associated with GST-E2 300–360 ( Figure 2 B ) . Taken together , our data reveal that the conserved oligomerization domain ( OD ) of NPM1 and EBNA2 aa 300–360 mediate the protein-protein interaction between EBNA2 and NPM1 . To address the possibility that NPM1 has a role in EBV-associated malignancies , the expression levels of NPM1 from an equal amount of BL , NPC , and LCLs cell lysates were assayed by immunoblot analysis . We found that substantial amounts of NPM1 were expressed in AKATA BL cells regardless of EBV status . By contrast , only relatively low levels of NPM1 were observed in EBV-infected NPC cells and the amounts of NPM1 in EBV-negative NPC cells were at a barely detectable level ( Figure 3A ) . In addition to the hallmarks of the EBV latency III program , EBNA2 and its target gene c-MYC , a robust activation of NPM1 gene expression was also detected in four individual LCLs , while none of the proteins ( EBNA2 , c-MYC and NPM1 ) was detected in primary B cells . Quantitative real time PCR ( qPCR ) further identified that the mRNA levels of NPM1 were activated by 2- to 3 . 5-fold in LCLs in comparison to primary B cells ( Figure 3B ) . A transfected NPM1 reporter plasmid ( NPM1-Luc ) was consistently induced by EBNA2 to reach 4-fold activation , and EBNALP co-activated the EBNA2 effect to result in 8-fold total activation , which further up-regulated EBNA2-inducing NPM1 promoter activity by 2-fold ( Figure 3C ) . Furthermore , a 1 . 7-fold increase in the NPM1 protein expression level was detected in BJAB cells that were stably expressing EBNA2 ( BJAB-E2 ) in comparison with the basal level identified in control BJAB cells ( Figure 3D ) . These results indicate that EBNA2 and EBNALP have a major role in activating NPM1 transcription in EBV-transformed B cells . To verify that the induction of NPM1 in lymphoblastoid cells is directly associated with EBV infection , the expression patterns of NPM1 , EBNA2 , and c-MYC control in primary B cells were monitored by microscopy or flow cytometry-mediated IF analysis after 0 , 3 or 7 days of virus or mock infection . As expected , only the background signals of EBNA2 , c-MYC , and NPM1 were detected in mock-infected B cells from 0 to 7 days . EBNA2 , the key activator of EBV latent infection , began to be expressed in approximately 18% of B cells at 3 days post-infection ( dpi ) and increased to 26–30% at 7 dpi ( Figures 3 E , F , S2A–B , and Table S1 ) . The immunostained spots of c-MYC appeared in 14 . 86% of B cells at 3 dpi and increased to 27 . 06% at 7 dpi ( Figures 3F , S2B , and Table S1 ) . Similarly , NPM1-positive cells were observed in 12 . 76% of B cells and increased to 19 . 92% at 7 dpi ( Figures 3 E , F , S2A , and Table S1 ) . Notably , the c-MYC- or NPM1-expressing cells were all detected in EBV positively infected B cells , indicating an absolute dependence of c-MYC and NPM1 expression on EBV infection . Finally , we showed consistent expression of both c-MYC and NPM1 in a newly established LCL rather than in primary B cells ( Figures 3G and S2C ) , which further validated the hypothesis that NPM1 expression induction was highly associated with EBV infection . The fact that NPM1 has a role in transcription up-regulation via its associations with histone proteins inspired us to investigate whether NPM1 is involved in EBNA2-mediated transcription [22] . In transfection-mediated EBNA2-dependent transcription assays , the expression levels of EBNA2 were not affected by the co-transfected FNPM1 or any other NPM1 mutants ( Figures 4 A and B ) . EBNA2-inducing LMP1-Luc activity was maintained at 8- to 12-fold activation . We showed that neither the transfected FNPM1 nor any of the NPM1 mutants could affect the basal activity produced by the LMP1-Luc reporter plasmid alone . Ectopic expression of FNPM1 readily augmented the EBNA2-dependent LMP1-Luc activity to reach an average of 20- to 25-fold total activation , which induced a 2 . 5-fold co-activating effect ( Figure 4A ) . Similarly , the transfected C1 , C2 , or C3 elicited a 2 . 5-fold increase over the intrinsic EBNA2-inducing LMP1-Luc activity . By contrast , neither ectopically expressed N1 nor N2 was able to co-stimulate with EBNA2 , indicating that NPM1 N-terminal aa 1–120 contributed the major transcriptional up-regulating effect . Following the loss of the EBNA2 binding affinities found in the NPM1 mutants L102A , G105A , and S106A , we next investigated their potency in co-stimulation with EBNA2 . Although plasmid-expressed FNPM1 co-stimulated with EBNA2 by 2 . 5-fold , we found that all transfected NPM1 mutants lost the ability to up-regulate the EBNA2-inducing LMP1-Luc activity ( Figure 4B ) . These data reveal that NPM1-mediated self-association and interaction with EBNA2 are the prerequisite requirements for co-stimulation with EBNA2 . Both ChIP and streptavidin-agarose-mediated DNA pull-down analyses were further employed to gauge the abundance of NPM1 at the EBV latency-associated LMP1 promoter [31] . The amounts of EBNA2 and NPM1 bound to this cognate promoter were identified as 6% and 2% of the input DNA by PCR ( Figure 4C ) . In addition , the biotin-labeled LMP1 DNA pulled down approximately 2% of endogenous EBNA2 and RBP-Jκ and 5% of endogenous NPM1 ( Figure 4D ) , whereas neither of the proteins was precipitated by the biotin-labeled or non-labeled control DNA and the non-labeled LMP1 DNA . We then sought to explore whether NPM1 is essential for EBNA2-mediated transcription . The EBNA2-inducing LMP1-Luc activity was assayed in BJAB cells in which endogenous NPM1 was stably silenced by the lentivirus-expressed NPM1 shRNA ( shNPM1-1 or shNPM1-2 ) , or scrambled shRNA . BJAB cell lines that stably expressed each shRNA , which were designated as BJAB-shNPM1-1 , BJAB-shNPM1-2 , and BJAB-scrambled , or control BJAB cells were subjected to transfection-mediated reporter assays . The expression levels of the actin control were not affected by any of the transduced shRNAs , so the amounts of NPM1 detected in each cell line was expressed in relation to that of the control BJAB cells , giving 71% , 99% , and 0% NPM1 knockdown efficiency in the BJAB- shNPM1-1 , shNPM1-2 , and scrambled cell lines , respectively ( Figure 4E ) . With similar expression levels resulted by the transfected EBNA2 , the EBNA2-inducing LMP1-Luc activity could reach 6- and 8-fold activation in the control and BJAB-scrambled cell lines , whereas this EBNA2-dependent activity was reduced down to 3- and 1 . 5-fold activation in the BJAB-shNPM1-1 and BJAB-shNPM1-2 cell lines . In addition , the defect of EBNA2-mediated transcription caused by the stable NPM1 depletion could not be restored by the co-transfected EBNALP ( Figure 4F ) . Our data reveal the dominant importance of NPM1 in EBNA2-dependent transcription . The fact that NPM1 has been shown to enhance chromatin transcription by acting as a histone chaperone [22] inspired us to hypothesize that NPM1 is involved in recruiting EBNA2 onto the EBV latency-associated promoters . BJAB-E2 cells transduced with shRNAs , as described in Figure 4E , were used to monitor the changes in protein-protein interactions mediated by EBNA2 and RBP-Jκ by co-IP analysis . Endogenous EBNA2 , RBP-Jκ , and actin control were expressed at similar levels , and both RBP-Jκ and EBNA2 were efficiently immunoprecipitated by their corresponding antibodies from each cell lysate ( Figure 5A ) . Our result showed that a 58% knockdown of NPM1 decreased the amounts of co-immunoprecipitated RBP-Jκ and EBNA2 to 28% and 21% , whereas 80% depletion of NPM1 dissociated almost 100% of EBNA2 and RBP-Jκ from the immunoprecipitated protein complex . By contrast , the magnitude of the physical interaction between EBNA2 and RBP-Jκ that was identified in the scrambled shRNA-expressing cells remained similar to the level detected in control BJAB-E2 ( Figures 5A–C ) . Moreover , our results also indicated that NPM1 could associate with RBP-Jκ through its N-terminal OD ( Figures 5A and S3 ) , which further suggested that NPM1 acted as a connecting bridge to stabilize the interaction between EBNA2 and RBP-Jκ . To verify that NPM1 escorted EBNA2 to form a complex with RBP-Jκ at the EBV latency-specific LMP1 promoter , we next performed a ChIP assay using EBNA2 and LMP1-Luc reporter co-expressing BJAB cells ( BJAB-E2/LMP1-Luc ) that had been stably transduced with shRNAs as described in Figure 4E . Lentivirus-mediated expression of shNPM1-1 and shNPM1-2 effectively depleted endogenous NPM1 by 81% and 97% , while the scrambled shRNA remained ineffective in this cell line . Although EBNA2 , RBP-Jκ , and actin were expressed at similar levels in the control and in all shRNA-transduced BJAB-E2/LMP1-Luc cell lines , NPM1 deficiency caused a nearly complete dissociation of EBNA2 from the LMP1 promoter ( Figure 5D ) . By contrast , the promoter occupancy of RBP-Jκ was not altered by NPM1 depletion . Our results suggest that NPM1 has a role in the formation of a stable PIC . To verify that the involvement of NPM1 in EBNA2-dependent transcription is coordinated with the subsequent maintenance of lymphoblastoid cells , the expression levels of EBNA2 target genes and cell proliferation were assayed using IB4 LCLs that had been stably transduced with each type of shRNA as described elsewhere in this article . In addition to actin , we found that the expression levels of EBNAs , including EBNA2 , EBNALP and EBNA1 , and RBP-Jκ in the control cell line were similar to the levels of each protein identified in either the NPM1 shRNA or the scrambled shRNA expressing IB4 LCLs . As expected , the EBNA2 target genes , such as c-MYC and LMP1 , were expressed in a way that was inversely correlated with the NPM1 knockdown efficiency ( Figure 5E ) . Although cell growth was slightly impaired by the transduced scrambled shRNA , it remained similar to the phenotype caused by the control IB4 LCL . Strikingly , the cell proliferation of IB4 cells was halted when only 32% of the NPM1 expression was retained , whereas these cells died over time when an 89% knockdown efficiency of NPM1 was reached ( Figure 5F ) . Similar experiments were also conducted using the BJAB cell line and its derivatives as described in Figure 4E . We found that the patterns of cell growth were not altered by NPM1 depletion in the context of BJAB cells ( Figure 5G ) . These data strongly support the hypothesis that NPM1-assisted EBNA2-dependent transcription is linked to the concomitant maintenance of EBV-transformed B cells . To assess the possibility that the ATP-bound state of NPM1 [32] , [33] is involved in the PIC formation , the amounts of EBNA2 , RBP-Jκ , and NPM1 from IB4 cell lysates that were precipitated by ATP or streptavidin-agarose were identified by immunoblot analysis . Except for α-tubulin , we showed that approximately 0 . 4% of endogenous EBNA2 , RBP-Jκ , and NPM1 was pulled down by ATP-agarose . By contrast , neither of the proteins was precipitated by streptavidin-agarose ( Figure 6A ) . The ability to form a protein complex containing EBNA2 , RBP-Jκ , and NPM1 in IB4 cells was then monitored by co-IP assay after an ATP-depletion protocol . Strikingly , although ATP-depletion partially reduced the EBNA2 and BBP-Jκ expression levels , it vigorously dissociated each protein from the complex . By contrast , the stable protein complex remained in its original form in the control IB4 cells ( Figure 6B ) . A ChIP assay was carried out to further verify that ATP-depletion could impair the promoter occupancy of EBNA2 , RBP-Jκ , and NPM1 . In normal IB4 cells , a substantial level of occupancy of each protein at the LMP1 promoter was identified by both PCR and qPCR , which showed an average from 5% to 6% of the input DNA ( Figures 6C and S4A ) . The abundance of promoter-bound RBP-Jκ was not altered while both EBNA2 and NPM1 were almost completely dissociated from the LMP1 promoter in ATP-deficient cells . In addition , our result revealed that the amount of acetylated H3 ( H3ac ) at the control GAPDH promoter was not affected by ATP-depletion , accounting for 4% of the input DNA ( Figures 6C and S4B–C ) . To confirm that ATP-depletion should consequently lead to the abrogation of EBNA2-dependent transcription , the EBNA2-inducing LMP1-Luc activity was assayed in BJAB cells with 0 to 2 hours of ATP-depletion . Although the expression levels of the transfected EBNA2 and endogenous actin were not altered within the ATP-depletion time frame , a progressive loss of EBNA2-inducing activity was detected upon ATP-depletion in a time-dependent manner ( Figure 6D ) . On the other hand , we demonstrated that the intrinsic activity of the transfected SV40-Luc reporter plasmid was not affected by ATP-depletion ( Figure S4C ) . To exclude the possibility that our findings could have resulted from the nonspecific effects caused by cell ATP-deficiency , the plasmids of two ATP-binding mutants , K263R and K257/263R [32] , were used to confirm that the ATP-bound state of NPM1 is indeed required for the putative PIC formation following transfection-mediated ATP-agarose pull-down , Co-IP , and reporter assays . Our result showed that 2% of the transfected FNPM1 and three previously identified EBNA2 binding mutants L102A , G105A , and S106A ( Figure 2B ) were efficiently pulled down by ATP-agarose , whereas plasmid-expressed GFP , K263R and K257/263R did not possess any detectable ATP-binding affinity ( Figures 6E–F; S4E ) . The co-IP assay indicated that both ATP-binding mutants lost the ability to associate with EBNA2 , while the transfected FNPM1 remained effectively bound to EBNA2 ( Figure 6G ) . Furthermore , ectopic expression of the ATP-binding mutants did not cause any co-stimulating effects on the EBNA2-inducing LMP1-Luc activity , while the transfected FNPM1 vigorously augmented this EBNA2-dependent activity by 4-fold ( Figure 6H ) . Altogether , these results suggest that NPM1 needs to be in an ATP-bound state to induce the assembly of the PIC . Finally , we aimed to provide direct evidence supporting the idea that the recruitment of EBNA2 to the latency-specific LMP1 promoter relies on its interaction with NPM1 . To this end , the transfected FNPM1 and three types of EBNA2 binding mutants , including G105A , S106A , and K257/263R , were monitored for their ability to form a complex with EBNA2 at the LMP1 promoter in BJAB/LMP1-Luc cells , respectively . The co-transfected EBNA2 and FNPM1 , or its mutant derivatives , were expressed at similar levels ( Figure 7A ) . ChIP assays identified the accumulation of FNPM1 at the LMP1 promoter at approximately 20-fold above the background level found in the IgG control . In contrast , the amount of promoter occupancy identified for each transfected EBNA2-binding mutant was similar to that of the IgG control ( Figures 7A and S5 ) . We noted that shNPM1-2 suppressed NPM1 mRNA by targeting its 3′UTR , which allowed us to carry out the plasmid-mediated rescue experiment for the EBNA2-dependent transcription assay in the context of BJAB-shNPM1-2 cell line . The co-transfected EBNA2 , FNPM1 , and three EBNA2 binding mutants were expressed at similar levels . A complete loss of the EBNA2-inducing activity of LMP1-Luc in BJAB-shNPM1-2 cells was successfully restored by the co-transfected FNPM from the background level to 10-fold activation , which was comparable to the EBNA2-dependent transcriptional activity detected in BJAB-scrambled cells . In comparison , neither of the transfected EBNA2 binding mutants was able to rescue the EBNA2-dependent transcription defect of LMP1-Luc in BJAB-shNPM1-2 cells ( Figure 7B ) . These data imply that the ability of NPM1 to bind to EBNA2 is the most important factor that determines the EBNA2 abundance at its target promoter .
Due to a lack of intrinsic DNA binding activity , the stable association of EBNA2 with RBP-Jκ has been characterized in each EBNA2 target promoter , although the importance of the PU . 1 or AUF1 sites in the LMP1 promoter argues that the RBP-Jκ sites are insufficient to support all EBNA2-dependent transcription [1] . The potent interactions with both RBP-Jκ and SKIP that are mediated by the EBNA2 internal region aa 280–337 has long been considered to be a direct force that induces the formation of the PIC [2] , [13] . However , the mechanistic basis for this scenario has yet to be elucidated . In this study , we identified the domain that corresponds to EBNA2 aa 300–360 as the NPM1 binding module , which contains a partial overlapping region within the RBP-Jκ binding domain . Oligomerization mediated by the N-terminal region aa 1–120 of NPM1 is known to be coordinated with its major chaperone activity , while the ATP-bound state of NPM1 depends on the sumoylation of the K263 residue , which was previously implicated in protein stability , nuclear distribution , and cell proliferation [32] , [34] . Our data reveal that not only the N-terminal OD of NPM1 but also the oligomerized and ATP-bound states of NPM1 are required for maintaining the specific interaction with EBNA2 . The sophisticated functional profiles of NPM1 documented so far have led to the difficulties in the delineation of its biological picture . Nevertheless , the overexpression of NPM1 is found in many types of solid human tumors , including tumors of the thyroid , brain , liver , and prostate [34] . Induction of NPM1 by mitotic agent treatments in B or T lymphocytes further echoes its involvement in mitotic progression and cell proliferation [19] , [35] . Moreover , NPM1 has been identified as one of the c-MYC target genes , and it potentially enhances c-MYC-induced hyperproliferation and transformation via a direct protein-protein interaction [20] , [36] . Accordingly , we observed a pronounced induction of both c-MYC and NPM1 in EBV positively infected B lymphocytes throughout the duration of viral infection . This finding provides a more comprehensive understanding of why c-MYC is one of the major determinants of the EBV latency III program maintenance [37] , [38] . Our results at least partially indicate that EBNA2 and EBNALP play a role in the transactivation of NPM1 , and this effect could possibly be further augmented by the subsequent expression of c-MYC . Additionally , the high levels of NPM1 observed in EBV negatively infected AKATA or BJAB BL cells suggest that proliferating B cells constitute an additional means to induce NPM1 expression in the absence of an EBV infection . This assumption can also explain the observation that the NPM1-Luc reporter activity induced by EBNA2 and EBNALP in BJAB BL cells or the overexpressed level of enodogenous NPM1 in the EBNA2 stably expressing BJAB cell line was simply shown as an intermediate phenotype . The main breakthrough of this study is that ATP-charged NPM1 functions as a connecting bridge that links EBNA2 and RBP-Jκ together to form a complex at the EBV latency-specific promoter LMP1 , which implies that ATP-bound state of NPM1 is involved in the formation of a stable PIC . In the absence of either NPM1 or ATP , EBNA2 no longer binds to RBP-Jκ and will dissociate from the LMP1 promoter , while the DNA-binding potency of RBP-Jκ remains unaltered . Thus , the major contribution of NPM1 is that it acts as a chaperone to escort EBNA2 to its target genes . Our findings fill in missing parts of the EBNA2-dependent transcription model , which have been lacking from previous studies . Because neither NPM1 nor EBNA2 possesses bona-fide DNA-binding potency , the DNA-binding protein RBP-Jκ emerges as their intermediate instead . Nevertheless , EBNA2 alone remains insufficient in maintaining the interaction with RBP-Jκ and yet requires a prerequisite binding with NPM1 to stabilize the subsequent interaction with RBP-Jκ . In lymphoblastoid cells , the dominant importance of NPM1 in EBNA2-dependent transcription further supports this proposed scenario . Among the viral latent proteins that are involved in EBV-mediated immortalization of B cells , the expression of LMP1 appears to be highly dependent on EBNA2 . As in the EBNA2 target gene c-MYC , an enormous reduction in LMP1 expression was observed in NPM1 knockdown lymphoblastoid cells , while the expression levels of the remaining EBNAs , including EBNA1 , EBNA2 , and EBNALP , RBP-Jκ and the actin control were not altered . Most importantly , the robust impairments of cell growth caused by NPM1 knockdown in IB4 LCL emphasize the conspicuous biological role of NPM1 in EBV latent infection . According to our progress in this study , we have constructed the following new model for the EBNA2-dependent transcriptional cascades that are activated in EBV-immortalized B cells ( Figure 7C ) : ( 1 ) Entry of EBV into B cells occurs at the initial stage of viral infection . ( 2 ) EBNA2 and EBNALP are transcribed via activation of Wp . ( 3 ) Activation of NPM1 expression and other EBNA2 and EBNALP-dependent genes , such as c-MYC , CD21 , and CD23 . ( 4 ) Oligomerized NPM1 is charged by ATP to form the ATP-bound state ( which is the active form of NPM1 ) . ( 5 ) RBP-Jκ is recruited to the binding sites that are located in the EBNA2 target promoters . ( 6 ) The active form of NPM1 binds to EBNA2 , escorts it to form a complex with RBP-Jκ , and subsequently induces the assembly of a PIC . ( 7 ) The EBNA2 C-terminal AD mediates the recruitment of the basal transcription machinery ( BTM ) , transcription factors ( TFs ) , and transcription cofactors ( TCs ) . ( 8 ) EBNALP binds to EBNA2 and augments EBNA2-dependent transcription . ( 9 ) Expression of cellular and viral genes that are essential for the establishment and maintenance of the EBV latent infection occur . The involvement of NPM1 in viral replication has been documented in hepatitis delta virus , Japanese encephalitis virus , and adenovirus [39] , [40] , [41] . Until recently , NPM1 phosphorylation by v-cyclin-CDK6 was thought to be a critical event that was linked to KSHV latency [42] . In addition , the acetylation of NPM1 was further implicated in viral transactivation upon HIV infection [43] . Accumulated evidence implies that NPM1 seems to have co-evolved with the human viruses to facilitate diverse virus infection processes within the nucleus . The fact that EBNA2 mimics an activated Notch receptor to drive transcription through RBP-Jκ during B cell immortalization [44] suggests the possibility that NPM1 represents a prototype nuclear chaperone that is involved in Notch-dependent transcription , although this hypothesis has yet to be confirmed . Our studies open a new window for exploring the molecular models of EBNA2 and Notch-dependent transcriptional cascades . Moreover , our data additionally reveal NPM1 to be a potent drug target for EBV-associated diseases .
The expression vector of the GST-EBNA2 ( GST-E2 ) acidic domain ( AD ) was previously described [8] . The remaining GST-E2 expression plasmids were generated by subcloning the indicated EBNA2 flanking sequences , which were amplified by polymerase chain reaction ( PCR ) using the appropriate corresponding primer pairs , into the BamHI and EcoRI sites of pGEX-2TK ( GE Healthcare ) . The expression vectors of flag-tagged wild type NPM1 ( FNPM1 ) or its derivatives of truncated deletion mutants were gifts of Dr . Charles J . Sherff ( Howard Hughes Medical Institute and Department of Genetics & Tumor Cell Biology , St . Jude Children's Research Hospital ) . The flag-tagged expression plasmids of the NPM1 point mutants were generated using the QuikChange Site-Directed Mutagenesis Kit ( Stratagene ) instruction manual . The GFP-NPM1 expression vector was generated by inserting the entire coding sequence of NPM1 into the XhoI and BamHI sites of eGFP-C1 ( TakaraBio USA ) . The expression vectors of EBNA2 ( E2 ) and EBNALP ( LP ) and the reporter plasmids LMP1-Luc and CMV-βGal were described previously [10] , [45] . The reporter plasmid of NPM1-Luc was generated by subcloning the promoter fragment −910 to +49 into the KpnI and XhoI sites of the PGL3 vector ( Promega ) . IB4 is an EBV-transformed LCL [46] , whereas LCL1-3 are previously established cell lines from the laboratory . AKATA and BJAB are EBV latently infected type I and non-infected Burkitt's Lymphoma ( BL ) cell lines [47] , [48] . Both LCLs and BL cell lines were cultured in RPMI-1640 ( Life Technology ) supplemented with 10% fetal calf serum ( FCS ) ( R10 medium ) ( Biological Inc . ) , while HeLa and 293T cells were maintained in DMEM ( Life Technology ) supplemented with 10% FCS . For the ATP depletion assay , the culture medium for IB4 or BJAB cells was replaced with glucose ( − ) RPMI640 ( Life Technology ) supplemented with 10% FCS , 2 mM deoxyglucose ( Sigma ) , and 300 nM antimycin A ( Sigma ) and cultured for 2 hours or the indicated time before the assays were completed . The transfection procedure carried out in this study has been previously described [9] , [45] . The EBNA2 and LMP1-Luc reporter co-expressing BJAB stable clones were established by co-transfection of pSG5-EBNA2 ( E2 ) , LMP1-Luc , and pIRES-puro ( TakaraBio USA ) expression plasmids into BJAB cells and selected with 5 ng/ml puromycin . HeLa ( 2×105 ) were transfected with 0 . 5 µg of the GFP-NPM1 and EBNA2 expression vectors , whereas 293T cells ( 2×106 ) were transfected with 3 µg FNPM1 or its derivative mutant expression vectors and 3 µg E2 or control empty vector using Lipofectamine 2000 ( Life Technology ) following the manufacturer's protocol . For the EBNA2-inducing LMP1-Luc activity assay , appropriate amounts of the necessary expression vectors , LMP1-Luc reporter plasmid , and internal control CMV-βGal plasmid were co-transfected into BJAB cells or NPM1 knockdown cells under the indicated circumstances . Luciferase and β-Gal activities were assayed by Orion L ( Berthold ) . The data are represented as the mean ± the standard error of the mean ( SEM ) from three independent experiments . Whenever necessary , statistical comparisons were performed by one-way ANOVA variance analyses . A p-value of less than 0 . 05 was considered to be statistically significant . GST or GST-E2s were expressed in E . coli BL21 ( DE3 ) pLys strain ( Stratagene ) and purified by affinity chromatography analysis using GST-sepharose ( GE Healthcare ) , and each of the purified recombinant proteins was used as the protein bait to pull down proteins from IB4 cell lysates . Furthermore , cell lysates from BJAB cells that had been transfected with 30 µg of flag-tagged empty vector , wild type , or the indicated truncated deletion mutants of NPM1 were further employed for protein affinity pull-down assays using GST and GST-E2 aa 300–360 recombinant proteins . The cellular proteins that were bound to each of the GST-E2s were analyzed by SDS-polyacrylamide gel electrophoresis ( SDS-PAGE ) followed by LC-MS/MS ( Protech Taiwan ) or immunoblot analysis . Primary B cells were isolated from peripheral blood mononuclear cells ( PBMC ) using Dynabeads Untouched ( Life Technologies ) following the manufacturer's suggested protocol . EBV infectious particles were induced and purified from the B95-8 Z-HT cell line as previously described [49] . Primary B cells ( 5×104 per 100 µl ) were resuspended in RPMI 1640 supplemented with 15% fetal calf serum ( FCS ) , 2 mM L-glutamine , and penicillin/streptomycin and aliquoted into a 96-well plate . One hundred microliters of virus or PBS ( mock infection ) was used for each infection . Immunofluorescence ( IF ) analysis was performed according to the immunostaining protocol described previously [50] using the EBNA2 , NPM1 , and c-MYC-specific antibodies vC-20 ( Santa Cruz ) , C-19 ( Santa Cruz ) , and α c-MYC ( Millipore ) , respectively . In the co-immunostaining assay , rhodamine-conjugated goat anti-mouse and FITC-conjugated goat anti-rabbit ( Kirkegaard & Perry Laboratories , Inc . ) antibodies were used as fluorochromes , and DNA was counterstained with DRAQ5 ( Bio Status ) or DAPI ( Sigma ) . The nuclear co-localization of EBNA2 and NPM1 was visualized by confocal microscope ( LEICA TCS SP2 AOBS ) . For the virus infection assay , primary B cells infected with EBV or mock ( PBS ) were immunostained with the appropriate antibodies followed a counterstaining protocol with the corresponding secondary antibodies , which were conjugated with FITC or rhodamine . The immunostained cells were visualized using the Cell Imaging Station ( Life Technologies ) or analyzed by Guava flow cytometry ( Millipore ) . Lentiviral shRNA expression plasmids of NPM1 were purchased from the National RNAi Core Facility at Academia Sinica , Taiwan . The production of the shRNA-expressing lentiviral vectors was carried out by following the protocol suggested by the manufacturer . To silence NPM1 , 5×105 BJAB BL cells or IB4 cells per ml were cultured in six-well culture dishes and transduced with 1 ml of lentivirus supernatant in the presence of 8 µg/mL of polybrene . Each transfectant was replenished with new media 48 hours after transduction and maintained for another 48 hours , followed by selection with 5 ng/ml of puromycin . IB4 cells or EBNA2/LMP1-Luc co-expressing BJAB cells stably transduced with NPM1 shRNAs or scrambled shRNA ( 1×107 ) were subjected to an enzymatic shearing protocol ( Active motif ) suggested by the manufacturer followed by a subsequent chromatin immunoprecipitation ( ChIP ) analysis using antibodies for NPM1 ( 5E3; from Abcam ) , EBNA2 ( PE2; from Abcam ) , or RBP-Jκ ( from Genetext ) . The promoter occupancy of EBNA2 , NPM1 , and RBP-Jκ was monitored by PCR or quantitative PCR ( qPCR ) analysis for the corresponding promoter fragments . The primers for this assay are listed in Figure S4C . The ChIP assay kit for GAPDH control was purchased from Millipore . The protocol for qPCR was performed using the Eco Real-Time PCR system ( Illumina ) as previously described [31] . The promoter occupancy that was identified for each protein or IgG control was normalized by the input DNA and formulated as a percentage of the input DNA . On the other hand , the promoter occupancy of transfected FNPM1 or its mutants were normalized to the IgG control . The LMP1 promoter DNA or control oriP DNA was amplified by PCR using a pair of the corresponding biotin-labeled or non-biotin-labeled primers [31] . Cell lysates from 1×107 or 1×108 IB4 cells were prepared and subjected to a streptavidin ( Life Technology ) or an ATP-agarose ( Sigma ) -mediated pull-down assay . The proteins that were bound to each target DNA fragment or ATP-agarose were monitored by immunoblotting with the appropriate antibodies . Normal IB4 LCL and NPM1 knockdown IB4 cells ( 5×104 per 200 µl ) or BJAB cells and NPM1 knockdown BJAB cells ( 104 per 200 µl ) were aliquoted in triplicate into 96-well plates . Viable cells were counted using the trypan blue exclusion method every 24 hours for five consecutive days .
|
Epstein-Barr Virus ( EBV ) infects human B cells to establish a permanent infection in hosts , which can cause diseases ranging from infectious mononucleosis to a broad spectrum of human malignancies . The conversion of human primary B cells into indefinitely proliferating lymphoblastoid cell lines ( LCLs ) by in vitro EBV infection provides a suitable model for virus-mediated cellular transformation . Epstein-Barr nuclear antigen ( EBNA ) 2-mediated transcription is essential for the establishment and maintenance of EBV latent infection . In this report , we have extensively explored the mechanism by which EBNA2 activates the latency-specific LMP1 promoter to establish a permanent infection in B cells . We have identified and characterized the protein-protein interaction of EBNA2 with the nuclear shuttle protein nucleophosmin ( NPM1 ) in vivo and in vitro . In particular , we have determined that the expression of NPM1 is promptly induced upon EBV infection and that EBNA2 has a role in activating NPM1 gene expression . Furthermore , we have shown that oligomerized NPM1 is charged by ATP and binds to EBNA2 , which is crucial for its ability to stabilize its interaction with the DNA binding protein RBP-Jκ , which is in turn essential for supporting the transcriptional cascades of EBV latent infection . Our findings provide striking evidence to illustrate a new model for understanding EBV pathology .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"infectious",
"diseases",
"virology",
"biology",
"microbiology",
"viral",
"diseases"
] |
2012
|
The Nuclear Chaperone Nucleophosmin Escorts an Epstein-Barr Virus Nuclear Antigen to Establish Transcriptional Cascades for Latent Infection in Human B Cells
|
Type 2 Diabetes ( T2D ) is a highly prevalent chronic metabolic disease with strong co-morbidity with obesity and cardiovascular diseases . There is growing evidence supporting the notion that a crosstalk between mitochondria and the insulin signaling cascade could be involved in the etiology of T2D and insulin resistance . In this study we investigated the molecular basis of this crosstalk by using systems biology approaches . We combined , filtered , and interrogated different types of functional interaction data , such as direct protein–protein interactions , co-expression analyses , and metabolic and signaling dependencies . As a result , we constructed the mitochondria-insulin ( MITIN ) network , which highlights 286 genes as candidate functional linkers between these two systems . The results of internal gene expression analysis of three independent experimental models of mitochondria and insulin signaling perturbations further support the connecting roles of these genes . In addition , we further assessed whether these genes are involved in the etiology of T2D using the genome-wide association study meta-analysis from the DIAGRAM consortium , involving 8 , 130 T2D cases and 38 , 987 controls . We found modest enrichment of genes associated with T2D amongst our linker genes ( p = 0 . 0549 ) , including three already validated T2D SNPs and 15 additional SNPs , which , when combined , were collectively associated to increased fasting glucose levels according to MAGIC genome wide meta-analysis ( p = 8 . 12×10−5 ) . This study highlights the potential of combining systems biology , experimental , and genome-wide association data mining for identifying novel genes and related variants that increase vulnerability to complex diseases .
Insulin resistance is a common trait present in complex disorders such as type 2 diabetes ( T2D ) , obesity or metabolic syndrome ( MetS ) . Around 340 million people suffer from diabetes worldwide , 90% of whom have T2D ( http://www . who . int/diabetes/facts/en ) . Unlike type 1 diabetes , overt T2D is usually diagnosed several years after its onset due to its milder presenting symptoms , which in part explains why several devastating complications such as cardiovascular related diseases tend to develop soon after or have already arisen at the moment of the initial diagnosis . There has been growing interest in identifying genes and processes that could trigger insulin resistance beyond defects on the insulin signaling cascade itself . As a result , defective mitochondrial activity has been indirectly related to insulin resistance in insulin-targeted tissues , such as skeletal muscle [1] , [2] , [3] and liver [4] . In particular , patients with T2D and , more importantly , non-diabetic subjects with type 2 diabetic relatives showed mitochondrial dysfunction and lower expression of PPAR gamma co-activator 1 alpha and 1 beta ( PGC-1α and PGC1-1β ) , which are key regulators of mitochondrial biogenesis and function . In addition , subjects with early-onset type 2 diabetes typically show defective activation of PGC-1alpha in response to physical activity [5] , and similarly , morbid obese type 2 diabetic patients show a defective activation of mitochondrial gene expression in response to weight-loss surgery [5] . Whether there is a heritable component involved in the alterations in expression of mitochondrial genes/proteins in these common forms of T2D remains to be determined . Despite all of these efforts and lines of evidence , the mechanisms and the molecular contributors to the connection between mitochondria and the insulin signaling and resistance are still unknown . The availability of a wide range of functional interaction data , including metabolomics , genomics , transcriptomics and proteomics and the integration of all these data using systems biology approaches make it now possible to investigate in detail the molecular basis of the interaction between the insulin signaling cascade and mitochondrial biology in healthy and pathological scenarios , particularly in the context of T2D . In addition , and despite substantial progress achieved in the identification of candidate genes involved in specific complex processes or diseases through genome-wide association studies ( GWAS ) , for most diseases , including T2D , less than 10% of the heritability ( percentage of variance attributable to genetic variation ) can be explained by the identified genetic associations [6] . Some hypotheses suggest that a portion of the missing heritability stays behind multiple small effect size variants that have not yet reached genome-wide significance in GWAS meta-analyses when tested individually , due to insufficient sample sizes . If many of the modest effect variants are assumed to implicate genes that function in a limited number of biological processes , collective analysis of variants based on prior biological knowledge could substantially enhance association detection power . In that sense , the application of systems biology approaches to analyze GWAS data may have the potential to increase the chances of unraveling susceptibility genes or biological processes for complex diseases . In this study , we applied systems biology approaches to screen and identify novel candidate T2D genes . The search has been guided by the hypothesis that the functional components of the crosstalk between the insulin signaling pathway and the biology of the mitochondria may play a role in the etiology or the evolution of the disease . We have also generated and analyzed gene expression data on insulin resistance and mitochondria perturbed scenarios to support these candidate genes . We finally tested whether particular genetic variants in loci that contain the identified genes could be collectively associated with T2D .
In order to identify genes specifically involved in the crosstalk between the insulin signaling pathway and the mitochondria , we looked for all possible direct and indirect functional interactions between mitochondria and insulin signaling genes ( Figure 1 ) . We started by building reliable models and parts lists for these two systems . We first explored and manually filtered several public versions of the insulin signaling pathway to end up with a confident collection of 197 proteins/genes ( see Methods ) . At the same time , we extracted data from a database of nuclear and mitochondrial-encoded mitochondrial proteins ( MitoP2 ) to generate the corresponding list of 682 mitochondria genes [7] . Once both parts lists were constructed , we screened several large functional interaction databases to identify direct and indirect connections involving any of the protein/genes of each of the systems . We applied several filters and cutoffs to be able to isolate , from all available interactions , a reliable collection that will be used further in our study . For example , from protein-protein interaction ( PPI ) data , we only considered those protein pairs whose interactions were reported by two or more independent laboratories ( PPIhigh ) and whose pair of genes were reported to be expressed both in any of the insulin-sensitive tissues ( adipose tissue , muscle , liver and heart , [8] ) ; or any other PPI interaction reported only by a single laboratory , simultaneously expressed in any of the insulin-sensitive tissues and that also showed co-expression ( gene-expression correlation ) in a dataset of 427 healthy human liver samples [9] ( these interactions are here termed PPIcorr ) . As a third layer of functional interaction , we also linked those proteins observed to belong to the same protein complex as described in the CORUM protein complex database [10] . The fourth source of interaction consisted of pairs of genes coding for enzymes that participate in linked metabolic reactions , i . e . those reactions that are adjacent in a metabolic reaction map according to the Biochemical Genetic and Genomic ( BiGG_met ) and the Kyoto Encyclopedia of Genes and Genomes database ( http://www . genome . jp/kegg/kegg2 . html; KEGG_met ) [11] , [12] , [13] . Finally , we also included those interactions between genes coding for complexes or genes linked in a signaling pathway , as defined by KEGG ( KEGG_path ) [12] . This final functional interactome comprised 57 , 751 high confidence functional interactions involving 6963 genes , which represent a whole functional network of insulin-targeted tissues or cells . From the pool of selected high quality interactions ( affecting 6963 genes ) , we finally selected those interactions that , either directly or indirectly , provide a link between the mitochondrial and the insulin signaling cascade genes . We defined indirect interactions as those mediated by genes , termed linker or internode genes , that do not belong to either the insulin or the mitochondria parts list , but that are simultaneously connected to both systems . By applying these filters , we finally generated the mitochondria-insulin ( MITIN ) network consisting of 886 genes and a total of 1259 interactions , 70 direct ( Table S1 ) and 1194 indirect . The 70 direct interactions involved 44 insulin genes and 37 mitochondria genes , most of them showing only one evidence of interaction . Both the insulin and mitochondria genes that were directly connected were linked to a median of two genes from the other system . Direct connections showed heterogeneous sources of interaction: PPIhigh , PPIcorr , Corum Complexes , BiGG_met , KEGG_met , Kegg_pathway , contributed 41 , 9 , 13 , 2 , 3 , 12 links , respectively . Indirect interactions involved 286 linker internode genes ( Figure S1 , Dataset S1 , Table S2 and S3 ) . These internodes genes were connected to a mean number of 2 . 1 Insulin and 1 . 7 mitochondria genes and showed a mean of 2 . 6 and 2 . 0 lines of evidence of interaction with insulin and mitochondria , respectively . Regarding the 1194 indirect connections , PPI , PPIcorr , Corum Complexes , BiGG_met , KEGG_met , Kegg_pathway , contributed 570 , 472 , 1263 , 42 , 160 , 169 interactions , respectively . While the majority of the internode genes seem to be novel , as their bridging role connecting both systems has not yet been described , some of them have already been shown to interact with both systems , which constitutes an internal positive control of our underlying search methodology . For example , TRAF2 shows interactions within our MITIN network with four insulin and two mitochondrial genes ( Table 1 ) . Interestingly , other independent studies and approaches also identified five of these interactions . In particular with MAP3K1 ( MEKK1 ) , CAV1 ( caveolin-1 ) and MTOR ( mTOR ) , from the insulin signaling [14] , [15] , [16] and MAP3K5 ( ASK1 ) and CASP8 ( caspase-8 ) from the mitochondria [17] , [18] ( Figure 2 ) . Another example is NFKB1 , for which we found interactions with four insulin signaling and three mitochondrial genes . As above , NFKB1 has been also reported to interact with the IKBKB [19] , [20] , AKT2 [21] , MAP3K1 [22] and SOCS3 insulin genes , as well as to BCL2L1 [22] [23] and BCL2 [24] ( Figure 2 ) . The same MITIN network also allowed us to define which mitochondrial genes are more connected to insulin signaling , and vice-versa , either directly or indirectly . The top five insulin signaling genes most connected to mitochondria are NOLC1 , RPS6 , IKBKB , PKLR , SRC , with a total of 99 , 40 , 31 , 28 and 22 indirect connections with mitochondria , respectively . Similarly , the five most connected mitochondrial genes with the insulin cascade were TUFM , TP53 , SLC25A5 , POLG , ESR1 , with a total of 93 , 36 , 29 , 25 , and 19 indirect connections , respectively ( Table S4 ) . We next explored whether our collection of internode genes where enriched in particular functions or processes by querying the Molecular Signatures Database [25] . We found up to 148 functional signatures for which internode genes were significantly enriched ( 5 . 7×10−107<p value<4 . 41×10−6 , 1 . 94<Odds ratio<20 . 1; Table S5 ) . Besides several enriched categories related to translation , Reactome Regulation of Expression in Beta Cells ( p = 3 . 5×10−87 , Odds Ratio = 15 . 8 ) , Reactome Insulin Synthesis and Secretion ( p = 4 . 46×10−79 , Odds ratio = 14 . 0 ) , and Reactome Diabetes pathways ( p = 1 . 39×10−35 , Odds ratio = 5 . 5 ) were also highly enriched among our set of internode genes . No significant categories were found after correcting for multiple testing in a set of internode genes identified from a simulated network made of randomly generated interactions . In order to facilitate the selection of any of these genes for further studies , we have ranked them according to their number of connections to each of the systems . Hence , we provide a confident subset of 31 genes with at least three lines of evidence linking insulin signaling and mitochondria genes simultaneously ( Table 1 ) . As further support of the functional relationship between internode genes and both , the mitochondria and the insulin signaling pathway , we explored whether the expression of these identified internode genes is modified after perturbing each of the mitochondria or insulin signaling systems independently . To test the effect of the insulin signaling perturbation , we performed gene expression profiling of C2C12 differentiated myotubes that were either left untreated or treated with 100 nM insulin for 2 days in order to induce an insulin resistance state . This treatment resulted in the downregulation of the insulin receptor and subsequently significantly reduced insulin signaling cascades [26] . We used the gene set enrichment analysis method ( GSEA , [25] ) to look for enrichment of differential expression using our set of internode , mitochondria , and insulin genes as molecular signatures . Using the collection of all 6963 genes with identified interactions as a background , we found significant enrichment of upregulation within the internode genes ( Normalized Enrichment Score ( NES ) = 1 . 7; False Discovery Rate ( FDR ) = 0 . 0013 ) , while observed downregulation enrichment within the insulin signaling genes ( NES = −1 . 4; FDR = 0 . 028 ) ( Figure 3a ) . We also explored a second model of insulin signaling cascade perturbation through the analysis of transcriptome data from myotubes treated with RNAi against DOR ( also named Tp53inp2 ) . This gene is dysregulated in muscle of Zucker diabetic rats , participates in the myogenic differentiation and mediates a feed-forward loop between ecdysone receptor and the insulin signaling in flies [27] , [28] . In this model , we also found that there was an enrichment of upregulated internode genes ( NES = 1 . 4; FDR = 0 . 004 ) and enrichment of downregulated insulin ( NES = −1 . 35; FDR = 0 . 007 ) and mitochondrial ( NES = −1 . 36; FDR = 0 . 001 ) genes ( Figure 3c ) . In a parallel experiment we tested how perturbations of mitochondria affect the expression of the MITIN network genes . For this , we analyzed gene expression from the heart of Peroxisome-proliferator-activated-receptor γ coactivator 1 beta ( PGC-1β ) knock-out mice . PGC-1β is a co-activator that regulates mitochondrial biogenesis and function [29] , [30] , [31] , [32] . The analysis of heart gene expression of these mice showed an overrepresentation of upregulated genes within the internodes ( NES = 1 . 3; FDR = 0 . 02 ) , enrichment of upregulated genes within the insulin genes ( NES = 1 . 6; FDR = 0 . 0012 ) , and enrichment of downregulated mitochondria genes ( NES = −2 . 63; FDR = <0 . 0001 ) ( Figure 3b ) . Again , as a control from our experiment , randomly generated internode genes did not show any enrichment in any of these experiments ( Figure 3d ) . We next investigated whether any of these genes has been associated to phenotypes related to insulin resistance or energy metabolism . For this , we searched through the OMIM database ( http://www . ncbi . nlm . nih . gov/omim ) those internode genes that are involved in mendelian and complex disorders [33] . We found that , among all 286 internode genes , 191 ( 66% ) were in genomic loci associated to complex diseases or traits ( SNPs within 250 kb from internode gene were considered ) and 17 ( 6% ) were involved in mendelian diseases . Interestingly 53 of the genes ( 18% ) contained or were near polymorphisms associated to T2D or related traits such as obesity , adiposity , response to glucose challenge , hypertension or coronary artery disease ( Table S6 ) . 10 , 000 random simulations showed that finding 53 genes associated to T2D related traits was modestly more than what expected by chance ( p = 0 . 0535 ) . In contrast , the 10 , 000 random simulations also showed that we did not find more associations with any complex trait ( not restricting to T2D related traits ) , than would be expected by chance , suggesting that the enrichment for associations of the identified internode genes is specific for T2D and related metabolic traits . In order to further investigate the potential involvement of the internode genes in the etiology of T2D , we screened the DIAGRAM consortium GWAS dataset , which consisted on the largest T2D meta-analysis available at the time of the study ( DIAGRAM meta-analysis ) : 8 , 130 cases and 38 , 987 controls [34] . To analyze enrichment of associated genes within the internodes , we used MAGENTA [35] , a software specifically designed for large genome-wide association study meta-analyses , where individual genotypes are typically not available . We found that our internode gene list showed nominal enrichment for modest to strongly associated genes within the top 5% of T2D scores , with 18 genes observed , including three already confirmed T2D associated SNPs [34] , [36] , [37] , compared to the 12 expected by chance ( p = 0 . 0549 , Table 2 ) . These results were robust to the enrichment cutoff used ( p = 0 . 0368 when testing for enrichment above the 97 . 5th percentile of all gene scores; 6 genes expected above cutoff , 11 observed ) . Unlike the collection of internode genes , no significant enrichment for T2D associations was found for gene-sets belonging only to the insulin signaling ( p = 0 . 71 ) or to the mitochondrial ( p = 0 . 52 ) systems . The insulin and mitochondria genes directly interacting with each other were also not enriched for T2D associations ( p = 0 . 53 ) . To further support the involvement of at least some of these 18 internode SNPs in glucose metabolism regulation , we also computed how the best associated SNPs in the 18 regions increased the risk of altered glycemic traits , available from MAGIC consortium datasets [38] , [39] , [40] , [41] , [42] , using an approximation approach developed by Toby Johnson [43] . Among the seven traits tested , we found a significant association risk score for fasting glucose ( p = 8 . 12×10−5 including the 18 top ranked SNPs and p = 0 . 004 including 15 out of the 18 SNPs not previously associated with T2D ) . In order to evaluate the probability of finding such a highly statistical p-value , when using the top T2D associated genes ( and best local SNPs ) we ran MAGENTA on 10 , 000 simulated random gene-sets , and extracted for each simulation the p-values of the most significant SNP per gene for all genes that ranked above the 95th percentile . The empirical p-value was then calculated as the frequency of random gene-sets whose p-values were smaller than the one obtained with the real data and whose effect size was higher than 0 . We found that 8 . 12×10−5 is significantly lower than what one can expect by chance ( p = 0 . 0144 ) , confirming the association of our set of internode genes , not only with T2D , but also to fasting glucose levels . To further explore the involvement of the internode genes associated with T2D ( see above ) in related metabolic traits we explored several available GWA meta-analyses pertaining to obesity-related traits from the GIANT consortium [44] , [45] , seven glycemic traits from MAGIC datasets [38] , [39] , [40] , [41] , [42] , and cardiovascular disease traits from the ICBP consortium [46] . We found that in 10 of the 18 internode genomic loci with modest to strong associations , there was at least one SNP showing association ( p<10−5 ) to one of these metabolic traits . For example , rs6453220 , located in the IQGAP2 intron , was associated to circulating glycated hemoglobin ( p = 4 . 19×10−6 ) and rs13107325 , located upstream of NFKB1 , was strongly associated with diastolic blood pressure ( p = 7 . 53×10−7 ) , body mass index ( p = 1 . 37×10−7 ) , high density lipoprotein levels ( p = 7 . 2×10−11 ) , and systolic blood pressure ( p = 2 . 57×10−7 ) .
Understanding the molecular basis of insulin resistance is essential for the early diagnosis , treatment and prevention of T2D and related co-morbidities , such as hyperlipidemia or cardiovascular disease . In this study we explored the molecular basis of insulin resistance beyond the known role of insulin signaling genes , and , implicitly screened for novel candidate T2D genes . Based on published evidence that connects the function of the mitochondria with insulin resistance and T2D [5] , [47] , [48] , [49] , we hypothesized that there are genes responsible for the crosstalk between the mitochondria and the insulin signaling system , which makes them good candidates for T2D . By screening and filtering a variety of available functional interaction data , we have first generated a conservative network ( MITIN ) containing all genes involved in or connected to the insulin signaling or mitochondrial systems , not only through PPI but also based on interactions of other nature , including co-expression , protein complexes , and signaling and metabolic interactions . From there , we then selected a fraction of 286 internode genes that show connections to genes of both systems and are , therefore , likely to be involved in the functional crosstalk between the insulin signaling cascade and the mitochondria . We have examined these genes at different levels to validate their bridging role and their potential implication in T2D or co-morbidities . In order to provide a more stringent list amenable to low throughput molecular biology experiments in future studies on insulin resistance and diabetes , we ranked these genes on the basis of their level of connectivity to insulin and mitochondrial genes and generated a high confidence subset of 31 genes showing three or more functional connections to each of the systems . While there are no reported confirmatory data for the majority of the 286 internode genes , some have been already found to be linked to both systems , and even to T2D and related metabolic processes . For example TRAF2 [14] , [15] , [17] , [18] , NFKB1 [19] , [20] , [21] , [22] , [23] , [24] ( Figure 2 ) and SMAD3 [50] , which show multiple connections to insulin signaling and to mitochondrial genes in our MITIN network , have also been described elsewhere to interact with genes of both systems . In addition , variants near the NFKB1 gene have been associated to T2D based on the DIAGRAM dataset ( best nearby SNP p-value = 1 . 6×10−5 ) , while SMAD3 has been recently found to protect against diet-induced obesity as well as coronary artery disease [50] , [51] . Other genes that also emerge as connecting internode genes in our MITIN network , such as the chaperone HPSP90AA gene , have not been previously described as linked to the insulin or the mitochondrial systems , but have been linked to insulin resistance conditions and hence to T2D [52] , [53] . On top of the previous knowledge on some of the internode genes , we provide here further evidence that supports the robustness of our search strategy and of this collection of genes as potential molecular connectors of these systems , as well as insulin resistance or T2D candidate genes . First , the 286 internode genes showed significant enrichment of functional categories , like “regulation of beta cell development” ( p = 2 . 1×10−79 ) , “insulin synthesis and secretion” ( p = 3 . 4×10−79 ) and “diabetes pathways” ( p = 1 . 9×10−35 ) . Second , experimental models of mitochondria and insulin signaling perturbation caused a significant upregulation of the internode genes . This could be the result of direct regulation or a mechanism that compensates these perturbed metabolic scenarios . In all cases , the expression analyses helped us to confirm that these genes are indeed functionally connected to both systems . Furthermore , the deregulation of these internode genes under experimental conditions of insulin resistance suggests their involvement in T2D . Encouraged by our positive functional and expression results supporting the connecting role of the internode genes and their impact on T2D , we went one step further and used the MITIN network as a basis for the identification of genetic signatures associated with T2D , contributing to unraveling its missing heritability . We tested for enrichment of T2D associations within the newly identified internode genes , by analyzing the results from the DIAGRAM GWA meta-analyses [34] using MAGENTA to define gene association scores and enrichment of gene associations [35] . We found enrichment of T2D variants within this group of genes , involving 18 associated genes compared to the 12 that were expected by chance ( p = 0 . 0549 ) . Our study also confirms the absence of significant signal when we tested insulin signaling and mitochondria gene-sets for enrichment of T2D associations . This is in agreement with previous studies , where no enrichment was found for mitochondrial or insulin signaling genes [34] , [35] , and suggests that the genes involved on the crosstalk between the insulin and mitochondria networks are more susceptible to harbor T2D risk variants than those that belong to either the insulin cascade or the mitochondria alone . The best local SNP in each of the 18 top ranked regions showed a combined risk score of increased fasting glucose levels according to MAGIC consortium data-sets ( p = 8 . 12×10−5 ) . Also supporting these results , several variants in the internode genomic regions identified by MAGENTA were also associated with many metabolic related quantitative traits , as reported by the MAGIC [38] , [39] , [40] , [41] , [42] , GIANT [44] , [45] and ICBP [46] consortia ( Table 2 ) . Interestingly , the best-associated SNP in four of the 18 genes were among the 43 already validated loci of susceptibility for T2D , which in the former reports were assigned to ZBED3 , BCL11A , PRC1 , and KCNJ11 genes , based only in proximity [34] , [36] , [37] . Taking into account the intrinsic challenge in linking an associated variant to its causal gene , we cannot exclude that these SNPs may be proxies for causal variants affecting our group of identified internode genes . Accordingly , recent findings suggest that a fraction of regulatory variants can be more than 500 Kb away from their regulated gene and that a single locus can expand more than 1 Mb , and even contain more than one independent causal variant [54] , [55] , [56] . Among the top 18 top ranked internode genes identified by MAGENTA analyses of T2D GWAS meta-analysis , there are independent lines of evidence suggesting the involvement on the development of T2D or insulin resistance . For example , two members of the IQ-motif-containing GTPase-activating protein ( IQGAP ) family , scaffold proteins involved in a wide range of cellular and signaling processes , including cytoskeletal organization , cell adhesion , and tumorigenic processes [57] , [58] , appear in the top 95th percentile for association with T2D according to MAGENTA analysis . IQ motif containing GTPase activating protein 2 ( IQGAP2 ) , the second ranked gene according to MAGENTA analysis , contained an intronic low frequency SNP ( rs6453220; MAF = 0 . 05 ) , which was strongly associated with glycated haemoglobin according to MAGIC WGA-meta-analyses ( Hb1Ac; p = 4 . 19×10−6 ) , providing more evidence that variants in IQGAP2 may contribute to insulin resistance . In addition , another gene of the same family , IQGAP1 ( top four according to MAGENTA ) , was recently reported to bind the target of rapamycin complex 1 ( mTORC1 ) having a potential negative feedback loop role upstream mTORC1/S6K1 AKT1 activation [59] . Furthermore , IQGAP1 associates with PKA and AKAP79 in pancreatic Beta cells , suggesting a role in the Beta-cell development and physiology [60] . It is also worth mentioning that IQGAP1 was also found upregulated in our chronic insulin treatment experiment ( fold change 1 . 4; FDR<0 . 01 ) and the Tp53inp2 RNAi treatment in myotubes experiment ( fold change = 1 . 33; FDR = 0 . 1 ) . These results , together with the general role of scaffolding proteins as hubs of signaling pathways further supports the implication of the IQGAP protein family in the insulin signaling and the mitochondrial systems crosstalk and its association to T2D . RAB4A ( Best SNP p value = 3 . 5×10−5 ) is a GTPase that regulates glucose transporter GLUT4 [61] , and is suggested to participate in metabolic remodeling in the diabetic heart [62] . Finally , breast cancer anti-estrogen resistance 1 ( BCAR1 ) , ( Best SNP p value = 6 . 61×10−5 , distance from gene = 16 . 5 Kb ) is another gene that deserves attention , as is connected to 10 insulin genes , according to our network: CRK , SRC , PTPN1 , PTK2 , CRKL , PIK3R1 , GRB2 , PTPRF , RHOA and PTPRA . Interestingly , a SNP in an intronic region 16 Kb upstream this gene was reported to be strongly associated with type 1 diabetes [63] . In summary , this study contributes to untangling the molecular basis linking the mitochondria and the insulin signaling systems and provides a subset of novel T2D candidate genes for further genetic , molecular and clinical studies . This study also constitutes a proof of concept of the utility of combining several integrative systems biology approaches with the analysis of gene expression and large GWA meta-analyses to uncover novel associations with complex diseases of otherwise hidden candidate genes .
We constructed a consensus insulin pathway from several public resources , including ( Biocarta; www . biocarta . com , Kegg [12]; www . genome . jp/kegg/ , and PID; [64]; http://pid . nci . nih . gov/ ) and a commercial resource ( Biobase; www . biobase . de ) . This pathway was manually curated and refined by the participation of molecular biologists in the field . In order to select the parts lists that compose mitochondrial proteins or genes , we have selected a total set of 900 proteins from the mitoP2 database ( www . mitop2 . de/; [7] ) . As it was done for the insulin pathway , the set has been manually curated by the participation of the expert groups in the consortium . To allow for transferability of the results to other species , we have identified each mouse orthologous gene/protein for all involved proteins . To identify protein-protein interactions we used a non-redundant set of 23 protein interaction datasets and only included those interactions reported independently by two different laboratories ( PPIhigh ) [8] . For the gene co-expression analysis , we used the dataset of Schadt et al . [9] , which consists of expression data of 427 healthy human liver samples and constituted the largest insulin-sensitive human transcriptome dataset . We evaluated the overlap between gene co-expression in liver and low confident PPIs ( those reported only by a single lab ) to provide a new source of high confident interactions . Third , we added those interactions that pertained to the CORUM complex database [10] , considering that two genes are functionally linked if they both pertain to a common complex . The fourth source of interaction consisted of pairs of genes coding for those enzymes that participate in linked ( or consecutive ) metabolic reactions as described in KEGG or BiGG databases [11] , [12] , [13] . Finally , we also considered those interactions between genes coding for complexes or genes linked any signaling pathway , as defined by KEGG [12] . We used the Molecular Signatures Database from the Broad Institute ( [25]; http://www . broadinstitute . org/gsea/msigdb ) and for a total of 6770 gene sets , we computed an enrichment score based on a Chi-Square test . The corrected significant p-value after applying Bonferroni's correction for all the tests was 4 . 41×10−6 . We only considered gene sets that had at least 10 genes within the group of internode genes . All statistical analyses were performed using Bioconductor ( Gentleman et al . , 2004 ) . Microarray data was normalized via quantile normalization and summarized to probeset expression estimates via robust multi-array average ( RMA ) ( Irizarry et al . , 2003 ) using the function rma from the oligo package . All the newly generated data was deposited in the Gene Expression Omnibus ( GEO ) ( http://www . ncbi . nlm . nih . gov/geo ) database ( GSE3932 ) . We used gene set enrichment analysis ( GSEA ) ( Subramanian et al . , 2005 ) as implemented in the Bioconductor library phenoTest [65] to assess the degree of association between gene expression and the following signatures: insulin , mitochondria and internodes . As indicated in Subramanian et al . [25] , P-values were computed restricting attention to simulated ES with the same sign as ESobs . All chemicals and reagents were purchased from Sigma-Aldrich , ( Poole , UK ) . Briefly , C2C12 cells were cultured in Dulbecco's modified Eagle media ( DMEM ) supplemented with 10% Fetal bovine serum , and penicillin/streptomycin . To induce differentiation media was replenished by DMEM containing 2% ( v/v ) of horse serum with penicillin/streptomycin . Myotubes between days 4 and 7 following the induction of differentiation were used for experiments . For chronic insulin treatment cells were either left untreated or incubated with 100 nMinsulin in DMEM for 48 h in fusion medium to induce an insulin resistance state . Medium was changed every 24 h . Hearts were quickly collected and snap frozen in liquid nitrogen from wild-type and PGC-1β KO on a mix background ( sv129 and C57BL/6 ) generated as previously reported [31] . Animal procedures were performed in accordance with the UK Home Office regulations and the UK Animal Scientific Procedures Act [A ( sp ) A 1986] . Animals were housed in a temperature-controlled room with a 12-h light/dark cycle . Food and water were available ad libitum . Lentiviruses encoding scrambled or DOR siRNA were used as reported [27] . Fifteen million C2C12 myoblasts grown on 12-well plates were transduced at moi 100 and cells were amplified during 5–7 days . Transduced cells ( GFP-positive ) were then sorted with a MoFlo flow cytometer ( DakoCytomation , Summit v 3 . 1 software ) , obtaining between 93%–99% GFP-positive cells . Confluent C2C12 myoblasts previously infected with lentiviruses encoding scrambled RNA or DOR siRNA were allowed to differentiate in 5% horse serum-containing medium for 4 days . Total RNA was purified and microarrays were performed by using an Affimetrix platform . We used the latest DIAGRAM T2D GWA meta-analysis comprising 8 , 130 cases and 38 , 987 controls [34] and the MAGENTA software was used to test for enrichment of associations in the 286 internode genes [35] . Briefly , we assigned to each gene a set of SNPs that lie within 500 Kb upstream and downstream of the gene's most extreme transcript boundaries . This boundaries were based on the fact that a fraction of regulatory variants can be up to 500 Kb distal to their regulated gene and that a single locus may harbor more than one causal variants , and extend to more than 1 Mb from the locus top hit [54] , [55] , [56] . For each gene , a score was assigned based on the most significant SNP , followed by correction for confounders , including gene size , number of independent SNPs , and linkage disequilibrium-based properties . Once all the association scores were computed , MAGENTA tested for over-representation of genes in a given gene set above a predetermined gene score rank cutoff , which in this case was the 95th percentile of all gene scores . The enrichment is evaluated against a null distribution of gene sets of identical set size that were randomly sampled from the 6963 genes that constitute our complete interactome based on all identified functional interactions . We computed how the best associated SNPs in the 18 regions could collectively increase the risk of altered glycemic traits available from MAGIC consortium datasets [38] , [39] , [40] , [41] , [42] . We used the method described in [43] . An unweighted genetic risk score was defined for each individual as the sum of the number of risk increasing alleles at each of the 18 SNPs of interest . If one had access to individual-level data , association between SNP score and glycemic traits could be tested using the usual approach . However , when the risk score involves SNPs in linkage equilibrium , it was shown [43] that association between risk score and trait can be assessed using meta-analysis results only , without going back to individual-level data . The effect of the risk score on the phenotype is estimated bywhere is the meta analysis effect size for SNP j , and aj is the inverse of the standard error estimate of . The assumption of no Linkage Disequilibrium ( LD ) is required for the contribution of each SNP to be independent and for the standard error estimate to be valid . P-value for the risk score association can be assessed using the ratio of the SNP score effect estimate divided by its standard error , and assessing the significance of the ratio by comparing it to the standard normal distribution . This large sample procedure will result in valid p-values under the null hypothesis of no relationship between the trait and variants included in the risk score .
|
It has been shown that the crosstalk between insulin signaling and the mitochondria may be involved in the etiology of type 2 diabetes . In order to characterize the molecular basis of this crosstalk , we mined and filtered several interaction databases of different natures , including protein–protein interactions , gene co-expression , signaling , and metabolic pathway interactions , to identify reliable direct and indirect interactions between insulin signaling cascade and mitochondria genes . This allowed us to identify 286 genes that are associated simultaneously with insulin signaling and mitochondrial genes and therefore could act as a molecular bridge between both systems . We performed in vitro and in vivo experiments where the insulin signaling or the mitochondrial function were disrupted , and we found deregulation of these connecting genes . Finally , we found that common variants in genomic regions where these genes lie are enriched for genetic associations with type 2 diabetes and glycemic traits according to large genome-wide association meta-analyses . In summary , we reconstructed the network implicated in the crosstalk between the mitochondria and the insulin signaling and provide a list of genes connecting both systems . We also propose new potential type 2 diabetes candidate genes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"systems",
"biology",
"genome-wide",
"association",
"studies",
"genomics",
"functional",
"genomics",
"genetics",
"biology",
"computational",
"biology",
"metabolic",
"networks",
"signaling",
"networks",
"genetics",
"and",
"genomics"
] |
2012
|
Identification of Novel Type 2 Diabetes Candidate Genes Involved in the Crosstalk between the Mitochondrial and the Insulin Signaling Systems
|
Yersinia pestis causes pneumonic plague , a disease characterized by inflammation , necrosis and rapid bacterial growth which together cause acute lung congestion and lethality . The bacterial type III secretion system ( T3SS ) injects 7 effector proteins into host cells and their combined activities are necessary to establish infection . Y . pestis infection of the lungs proceeds as a biphasic inflammatory response believed to be regulated through the control of apoptosis and pyroptosis by a single , well-conserved T3SS effector protein YopJ . Recently , YopJ-mediated pyroptosis , which proceeds via the NLRP3-inflammasome , was shown to be regulated by a second T3SS effector protein YopK in the related strain Y . pseudotuberculosis . In this work , we show that for Y . pestis , YopK appears to regulate YopJ-mediated apoptosis , rather than pyroptosis , of macrophages . Inhibition of caspase-8 blocked YopK-dependent apoptosis , suggesting the involvement of the extrinsic pathway , and appeared cell-type specific . However , in contrast to yopJ , deletion of yopK caused a large decrease in virulence in a mouse pneumonic plague model . YopK-dependent modulation of macrophage apoptosis was observed at 6 and 24 hours post-infection ( HPI ) . When YopK was absent , decreased populations of macrophages and dendritic cells were seen in the lungs at 24 HPI and correlated with resolution rather than progression of inflammation . Together the data suggest that Y . pestis YopK may coordinate the inflammatory response during pneumonic plague through the regulation of apoptosis of immune cells .
Acute bacterial pneumonia is the result of active colonization of the airspace in the lungs combined with host inflammation that is unable to resolve due to host-pathogen interactions as well as progressing host- and microbial- induced injury . Resident macrophages in the lungs play an important role in orchestrating the mucosal immune response and subsequent tissue repair following infection [1] . Alveolar and interstitial macrophages act as sentinel cells and react to pathogen-associated molecular patterns following bacterial invasion of the lung mucosa by activating pro-inflammatory cytokine production and phagocytosis . Following chemotaxis , neutrophils are the primary mediators of bacterial clearance . After neutrophils destroy invading extracellular bacteria , interstitial macrophages activate a resolution program , allowing efferocytosis and clearance of apoptotic neutrophils [2] . Activated alveolar macrophages retain a pro-inflammatory role and apoptosis of these macrophages signals the down-regulation of inflammation and induction of tissue repair . When efferocytosis by interstitial macrophages does not occur , increased severity of pneumonia results while , conversely , treatment of mice with apoptotic macrophages is protective against lethality [3] , [4] . Apoptosis provides both pro- and anti-inflammatory signals and each is necessary to prevent bacterial pneumonia . Yersinia pestis is a Gram negative bacterium that causes bubonic , pneumonic , and septicemic plague [5] . Expression of multiple virulence factors together allow evasion and manipulation of the host innate immune system and rapid replication in the midst of a massive pro-inflammatory response [6] . Pulmonary infection of mammals progresses as an acute bronchopneumonia , with an initial delay in inflammatory responses thought to be important to successful infection [7] , [8] . Near 48 hours post-infection ( HPI ) , bacteria multiply rapidly and cause host cell pyroptosis and necrosis and a large pro-inflammatory response . Neutrophils form coalescing foci of inflammation that are unable to combat the massive bacterial growth , and acute bronchopneumonia rapidly overtakes the host . Depletion of neutrophils early following infection increases the sensitivity of mice to pneumonic plague suggesting that neutrophils are , at least initially , active against Y . pestis [9] . The type III secretion system ( T3SS ) is responsible for the injection of anti-host effector proteins , collectively known as Yops , into phagocytic and other mammalian cells and is essential for virulence [10] . Upon host cell contact , insertion of a translocation pore into the plasma membrane allows injection of Yops from the bacteria to the host cytosol [11] , [12] , [13] . Each Yop likely has a distinct role in the pathogenesis of infection because Y . pestis strains lacking individual effector proteins have varied virulence defects ranging from mild to severe [14] , [15] , [16] , [17] , [18] , [19] . Yop-mediated host cell death and modulation of inflammatory responses have been demonstrated and are important to the pathogenesis of plague . The T3SS effector protein YopJ is a deubiquitinase and acetyltransferase whose activity blocks signaling by mitogen activated protein kinase , prevents NF-κB activation and leads to decreased production of anti-apoptotic proteins and pro-inflammatory cytokines [20] , [21] , [22] , [23] . Yet in spite of these immune-modulating activities , YopJ is largely dispensable for virulence [18] . Moreover , the evolution of Yersinia towards increased virulence includes reductive secretion of YopJ suggesting selective pressure against YopJ activity for the development of plague [24] , [25] . Translocation of effector Yops is believed to be controlled by YopK , whose activity has been shown to modulate the size of the translocation pore in red blood cells and epithelial cells [26] , [27] . YopK is translocated into host cells where it localizes to the plasma membrane [27] , [28] . In addition , recent evidence supports a role for YopK in modulating inflammatory responses . Insertion of the translocation pore by a Y . pseudotuberculosis strain lacking most of the Yop effectors resulted in activation of caspase-1 and the inflammasome , subsequent production of pro-inflammatory cytokines and pyroptotic cell death [29] , [30] , [31] , [32] . YopK was therefore proposed to remain associated with the translocation pore where it could suppress inflammasome activation [33] . Nevertheless , in vivo , YopK was dispensable for caspase-1-mediated host defense suggesting additional function for YopK in Y . pseudotuberculosis virulence . Y . pestis is believed to evade inflammasome activation at least in part , due to the presence of a non-canonical lipopolysaccharide structure , not present in Y . pseudotuberculosis , which does not stimulate Toll-like receptor 4 signaling [34] , [35] . In this work , we found that YopK mediates inflammatory responses in a murine model of pneumonic plague . We show that YopK is necessary for YopJ-dependent activation of apoptosis of monocyte-like macrophages but for alveolar macrophages , deletion of YopK resulted in increased apoptosis . Using small molecule inhibitors , we show that YopK activity in RAW 264 . 7 cells is independent of caspase-9 , acting upstream of caspase-8 in the extrinsic apoptosis pathway . In vivo , YopK is necessary for virulence , and its deletion caused a 5 , 000-fold increase in lethal dose . YopK-dependent modulation of macrophage apoptosis was observed during the first 24 hrs of infection and led to recruitment of macrophages , prevention of dendritic cell migration and subsequent progression of inflammation and disease . In contrast , lung inflammation and congestion were observed early but were quickly followed by a resolution program in the absence of YopK . Pro-inflammatory cytokine production was not affected by YopK in vitro . Nevertheless , elevated levels of MCP-1 persisted throughout the yopK infection in vivo and the MCP-1 receptor , CCR2 , was involved in host defense . Together , the data suggest that YopK modulates inflammation through the control of macrophage and perhaps other immune cell apoptosis for the rapid development of pneumonic plague .
We generated a deletion of the yopK open reading frame in Y . pestis KIM D27 using homologous recombination and the suicide plasmid pCVD442 [36] . The resulting strain was analyzed by PCR and Western blot , confirming the deletion of yopK and the presence of all three virulence plasmids ( data not shown ) . This mutant strain was also found to be highly attenuated in an intravenous model of septicemic plague , with a calculated LD50 of 1 . 5×105 CFU , a 1 , 500-fold increase over the LD50 of the wild type ( WT ) Y . pestis KIM D27 strain ( Figure 1A ) [14] . Previous data on the function of YopK employed in vitro infection models of epithelial cells where increased injection of effector Yops has been demonstrated using cell fractionation and reporter assays examining a subset of Yops [26] , [27] . However , it is known that during infection , Y . pestis targets phagocytic cells [37] . We therefore sought to characterize the yopK mutant in vitro using macrophage infection models to better understand the potential impact of YopK in plague . Towards this end , we employed functional assays and first measured the induction of apoptosis in RAW 264 . 7 cells , a mouse monocyte-macrophage cell line . Caspase-3-dependent apoptosis of macrophages is caused by the T3SS effector YopJ , which has been shown to have increased translocation in the absence of yopK [27] . Thus , we anticipated that more caspase-3 activation would be observed in the yopK mutant due to the increased effector Yop translocation . Unexpectedly , however , the yopK mutant did not cause increased cleavage of caspase-3 but rather significantly less cleavage was observed ( Figure 1B ) . This phenotype could be complemented by expressing YopK on a low copy plasmid from its native promoter . In contrast , Y . pestis-induced lysis of RAW 264 . 7 macrophages was not significantly affected by YopK , and 25–30% LDH release was observed following infection by WT and yopK mutant Y . pestis ( Figure 1C ) . Host-cell lysis was dependent on the T3SS , however , and only 5% LDH release was observed following infection of RAW macrophages with Y . pestis lacking the T3SS virulence plasmid pCD1 . We also tested whether MH-S cells , a murine alveolar macrophage cell line , induced increased amounts of YopK-dependent caspase-3 activity following Y . pestis infection . In sharp contrast to RAW 264 . 7 cells , MH-S cells infected with Y . pestis yopK had significantly increased caspase-3 cleavage , consistent with increased translocation of YopJ due to the deregulation of the translocation pore ( Figure 1D ) . When yopK expression was complemented on a recombinant plasmid , however , levels of caspase-3 cleavage were significantly decreased to levels below that caused by the WT strain . This result is also consistent with a role for YopK as a regulator of pore size in the plasma membrane of MH-S cells where increased expression and translocation of YopK would be predicted to result in decreased injection of Yops . Furthermore , LDH release by infected MH-S cells was overall low , and only 3–5% LDH release was observed following infection by WT and yopK mutant Y . pestis ( Figure 1E ) . This low level of MH-S cell lysis was dependent on the T3SS as less than 1% LDH release was observed following infection by pCD1 mutant bacteria . Previous data indicated that YopK amino acid residue D46 was important for function whereas residue T45 was dispensable [38] . We therefore engineered these point mutations into the yopK complementing plasmid and tested them for the ability to complement the yopK mutant in inducing apoptosis of RAW macrophages . Indeed , mutation of D46 resulted in loss of YopK function and no complementation of apoptosis was observed ( Figure 1F ) . In contrast , T45 was dispensable and this mutant fully complemented apoptosis of RAW macrophages . Together , the genetic data support a role for YopK in the induction of apoptosis in RAW macrophages . Y . pestis yopK has previously been shown to translocate increased amounts of YopJ in CHO cells , yet in RAW macrophages , the yopK mutant induced decreased caspase-3 activation [27] . This result suggests that YopK may be necessary for YopJ activity in specific cells or that YopK has an independent role in inducing apoptosis that might be additive to YopJ . However , previously published data suggests apoptosis of Y . pestis-infected RAW macrophages requires YopJ [39] . To confirm that apoptosis of RAW macrophages requires YopJ in our assay , we infected RAW macrophages with Y . pestis carrying a point mutation in the catalytic site of YopJ ( C172A ) which abolishes its ability to induce apoptosis [31] . Consistent with previous reports , YopJ is necessary for apoptosis of macrophages and , upon infection with the yopJ mutant strain , no caspase-3 activation was observed ( Figure 2A ) . Therefore , while YopK is necessary for maximal apoptosis in RAW cells , YopJ is absolutely required . Previously , inhibition of caspase-8 and caspase-9 revealed that Y . pseudotuberculosis YopJ induces apoptosis through both the intrinsic and extrinsic pathways [40] . We therefore asked whether YopK was necessary for both of these activities . Caspase-3 cleavage can be activated by caspase-8 whose auto cleavage from pro-caspase-8 occurs following formation of a complex involving the adaptor protein FADD which is normally initiated through binding of Fas or TNF receptors [41] . YopJ is necessary for inducing formation of the DISC in the absence of receptor binding [42] . In lymphocytes and perhaps other immune cells , formation of the FADD-caspase-8 complex , the death-inducing signaling complex ( DISC ) leads to caspase-3 activation directly , while in most cells , caspase-3 activation requires amplification through caspase-9 and the intrinsic apoptosis pathway [43] . We therefore treated RAW macrophages with inhibitors to caspase-8 and -9 and measured the impact on Y . pestis-induced activation of caspase-3 . Pre-treatment of macrophages with caspase-9 inhibitor Z-LEHD-FMK partially blocked caspase-3 activation and 30% remained following infection of RAW cells by WT Y . pestis ( Figure 2B ) . This suggests that Yersinia-induced caspase-3 activation in RAW macrophages is partially dependent on caspase-9 . In contrast , the caspase-9 inhibitor had a more severe effect on apoptosis caused by the yopK mutant bacteria where caspase-3 activation was reduced to background levels . This suggests that the caspase-3 activity caused by the yopK mutant may be due solely to caspase-9-dependent signaling whereas when YopK is present , the extrinsic , caspase-9-independent pathway may be induced . Consistent with this hypothesis , pre-treatment of macrophages with caspase-8 inhibitor ( IETD ) abolished caspase-3 activity following infection by both WT and yopK mutant Y . pestis suggesting caspase-8 is absolutely required for caspase-3 activation following Y . pestis infection ( Figure 2C ) . Together , the data suggest that YopK acts upstream of caspase-8 to modulate YopJ-dependent apoptosis of RAW macrophages . Following Y . pestis infection , inhibitors of c-jun N-terminal kinases ( JNK ) increase caspase-3 activation of RAW and J774 macrophages in a manner dependent on the acetylation activity of YopJ [44] . We asked whether this required YopK . SP600125 pre-treatment caused a 50% increase in caspase-3 activation of RAW macrophages following WT infection , similar to previously published data ( Figure 2E ) . In contrast , SP600125 had no effect on infection by the yopK mutant as no detectable differences were observed between treated and untreated macrophages . Thus , like YopJ , YopK is required for the JNK inhibitor to induce apoptosis in infected RAW macrophages . Taken together the inhibitor data suggest that Y . pestis induces YopJ-dependent apoptosis of RAW macrophages through the intrinsic and extrinsic pathways , with a requirement for YopK in the extrinsic pathway . Because macrophages are key mediators of the innate and adaptive immune systems , and their apoptosis is likely to have profound effects on these systems in vivo , we were interested in understanding how the contribution of YopK to apoptosis affected inflammatory responses during pulmonary infection . Since the Y . pestis KIM D27 parent strain is a non-pigmented mutant , it is attenuated and cannot cause pulmonary disease [45] . In order to better understand the role of YopK in pathogenesis , we studied the fully virulent Y . pestis CO92 strain which is capable of causing primary pneumonic plague . Towards this end , we generated a yopK null mutation in Y . pestis CO92 using the strategy described above . This strain was characterized by PCR , confirming the absence of yopK and presence of pCD1 , pPCP and pMT1 ( data not shown ) . We measured the LD50 of this strain following intranasal infection of BALB/c mice as 1 . 5×106 CFU , a 5 , 000-fold increase in lethal dose over that of the parent CO92 strain ( Supplemental Figure S1A ) . This increase appears more severe than the LD50 measured in the septicemic plague model and , together with previously published data , suggests that YopK has an essential function for all forms of plague [14] . Importantly , the LD50 for both CO92 and KIM D27 yopK mutant strains is less than that observed following pulmonary infection of Y . pestis lacking the type III secretion system ( pCD1− ) , a strain that cannot cause disease in mice indicating the virulence defects observed in our strains are not due to second site mutations that result in loss of T3SS activity . Following infection with lethal challenge doses of yopK , some mice succumbed to disease in 3 to 4 days post-infection , a similar time course as wild type CO92 at 100-fold lower challenge dose , while about 10% died later ( 5 to 14 days post-infection ) . The mice that succumbed to disease early appeared to be suffering from primary pneumonic plague , defined as multi-focal bacterial colonies in the lungs and neutrophil congestion in the parenchyma ( Supplemental Figure S1B–C ) . Those mice that succumbed to disease later appeared to have developed and succumbed to septicemic plague , with fewer bacteria in the lungs , interstitial pneumonia , and massive liver and spleen necrosis and inflammation ( Supplemental Figure S1D–F ) . Bacterial titers and severity scores for lungs , liver and spleen of moribund mice were determined ( Supplemental Table S3 ) . These mice had in common little to no damage to or colonization of the spleen , but an overall systemic infection . To identify host responses that lead to clearance of yopK mutant Y . pestis , we next performed a 72 hour time course experiment . Due to the large attenuation of the yopK mutant , comparisons with wild type are subject to misinterpretation because of the growing differences in clinical state and bacterial load in the tissues over time . We therefore studied two challenge doses: one equivalent to 0 . 67 LD50 for yopK ( 1×106 CFU ) and the second equivalent 30 LD50 for wild type ( 1×104 CFU , a dose which causes no symptoms when used for the yopK mutant ) . This allowed us to assess similar clinical states of infection caused by both strains of bacteria . At 6 hours post-infection ( HPI ) , increased bacterial load for yopK was observed compared to WT due to the 100× higher inoculating dose ( data not shown ) . At 24 HPI , however , both strains colonized the lungs at similar levels ( Figure 3A ) . Strikingly , at this time point , pathology in the lungs of both wild type and yopK infected mice were similar with mild to moderate infiltration of neutrophils and little tissue damage ( Figure 3B , D ) . In contrast , at 72 HPI , the divergence of the two strains was notable , with wild-type bacteria replicating to high titer while yopK mutant bacteria were undergoing clearance . Congestion in yopK-infected lungs at 72 HPI was mild , and either interstitial or consisted of lymphocytes around blood vessels near the airways , while in WT-infected lungs , inflammation was more severe and predominantly neutrophils were present ( Figure 3C , E ) . Wild type-infected mice also had visible bacterial colonies in the hepatic sinusoids of the liver as well as the red pulp of the spleen while the yopK mice had smaller inflammatory foci in the liver and spleen ( Supplemental Figure S2A–D ) . Together , the pathology of the yopK infection suggests that these bacteria may cause an alteration of the host inflammatory response compared to that caused by wild type Y . pestis . In order to identify the cellular infiltrate in the lungs during the later stage of infection , we analyzed these tissue sections by immunohistochemistry . Formalin fixed lung tissues from 72 HPI were stained with NIMP R14 ( a monoclonal antibody that recognizes Ly6G/6C on the surface of monocytes and neutrophils ) , F4/80 , CD3 or B220 and analyzed by microscopy . Wild type bacteria induced an inflammatory response in the lungs that appeared primarily composed of Ly6G/6C+ cells ( Supplemental Figure S3 ) suggesting active neutrophil recruitment , while in the liver and spleen , fewer neutrophils were observed ( Supplemental Figure S4 ) . In contrast , fewer Ly6G/6C+ staining cells were present in yopK infected mice , and instead , F4/80+ , CD3+ and B220+ stained cells were found especially in the lungs indicating recruitment of macrophages , T cells and B cells . Together , the results suggest a dramatic difference in the inflammatory response between WT- and yopK-infected mice at 72 HPI . Correlation between divergence in inflammatory cell recruitment and reduction in bacterial load suggests that the primary role of YopK in virulence may be immunomodulatory . To determine if YopK-dependent caspase-3 activation occurs in vivo , we also stained lung tissue sections from 6 , 24 and 72 HPI with labeled antibodies to cleaved caspase-3 and analyzed them by microscopy . Initial assessment revealed relatively frequent caspase-3 positive macrophages in both WT- and yopK-infected lungs ( Figure 4A–B ) . We counted caspase-3 positive macrophages in 10 non-overlapping fields of lungs at 40× magnification . At 6 HPI , a modest , but significant increase in caspase-3-positive staining was observed following yopK infection compared to WT suggesting an increase in apoptotic cells ( Figure 4C ) . This observation is similar to that observed in MH-S cells , however it is important to note that at this time point , there are increased numbers of yopK bacteria compared to WT which may be responsible for this observation . In contrast , at 24 HPI when bacterial titers of WT and yopK were highly similar , lungs harvested from mice infected by WT contained greater than 2-fold more macrophages undergoing apoptosis compared to yopK infected mice . This suggests that YopK may induce increased apoptosis of macrophages in vivo , correlating with the progressing development of pneumonia in WT-infected mice . In contrast , at 72 HPI , mice infected by WT bacteria ( now progressing towards lethality ) , displayed significantly reduced caspase-3 positive macrophages compared to 24 HPI . At this time point , the yopK-infected mice have begun to resolve inflammation and likely harbor many fewer bacteria than the WT-infected mice . Overall , these results are consistent with a role for YopK in modulating macrophage apoptosis early following infection , as seen in vitro . We quantified the number of lung and alveolar macrophages and dendritic cells by flow cytometry . Lung homogenates of BALB/c mice were prepared at 24 HPI and stained with CD11c and F4/80 ( Figure 5A–C ) . Alveolar macrophages ( CD11c+F4/80+ ) appeared similar in number between naïve mice and WT- or yopK-infected mice ( Figure 5D ) . Strikingly , lung macrophages ( CD11c−F4/80+ ) were increased in WT- but not yopK-infected mice while dendritic cells ( CD11c+F4/80− ) were decreased in yopK-infected mice . Together , the data suggests that YopK activity in vivo leads to recruitment of macrophages , resulting in progressing bronchopneumonia , while the absence of yopK caused an alternative inflammatory program that correlated with bacterial clearance and no disease . YopK-induced apoptosis of macrophages would be expected to impact bacterial clearance directly , contribute to neutrophil-induced tissue injury and/or modulate inflammatory responses [46] , [47] . To distinguish between these possibilities , we sought to determine the overall impact of YopK on immunomodulation during infection . Towards this goal , we analyzed serum cytokines from mice infected with WT and the yopK mutant for a panel of 13 cytokines and chemokines . When comparing WT- and yopK-infected mice at 24 HPI , we found similar , low levels of pro-inflammatory cytokines in the serum ( Figure 6A–D , Supplemental Table S5 ) . As the infection progressed , WT-infected mice expressed significantly increased amounts of pro-inflammatory and anti-inflammatory cytokines and chemokines compared to the yopK-infected mice . Together these data suggest that YopK modulates inflammatory responses but may not affect secretion of inflammatory cytokines or chemokines . Notably , however , yopK-infected mice continued to express moderate levels of MCP-1 when most other pro-inflammatory cytokines were no longer detectable . We therefore stained the formalin-fixed lung sections with antibody to CCR2 , the receptor for MCP-1 , present on many leukocytes , to determine if CCR2+ cells are recruited to the lungs as the yopK infection is being cleared . While the yopK-infected lungs stained positive for CCR2 as early as 24 HPI , the WT-infected lungs harbored very little CCR2+ staining ( Figure 6E–F ) . Likewise , at 72 HPI there were increased numbers of CCR2+ cells in the perivascular regions of the yopK-infected lungs , while WT-infected lungs continued to harbor little to no CCR2+ staining cells ( Figure 6G–H ) . Importantly , the observed changes in the pro-inflammatory cytokine response and inflammatory cell recruitment occurred after the onset of increased macrophage apoptosis , consistent with a model whereby YopK-mediated depletion of macrophages influences inflammation . Furthermore , we also observed increased B220+ cells in the perivascular regions of yopK-infected lungs at 24 and 72 HPI as compared to WT ( Figure 6I–L ) . In contrast , although Gr1+ cells were present in mice infected with either WT or yopK mutant Y . pestis at 24 HPI , by 72 HPI , Gr1+ cells predominated in WT-infected tissues , with yopK-infected mice having mainly Gr1+ cell debris ( Figure 6M–P ) . Together the results suggest that during pulmonary infection by Y . pestis yopK , recruitment of CCR2+ cells correlates with bacterial clearance and resolution of inflammation . We recently showed that C57BL/6 Ccr2−/− mice develop pneumonic plague with similar kinetics as wild type mice suggesting that CCR2+ cells may play only a minor role during WT infection [48] . However , CCR2 has been shown to be important to host defense against Y . pestis mutants lacking YopM [49] . To determine the relevance of CCR2 signaling to host defense against Y . pestis yopK , we tested C57BL/6 mice lacking Ccr2 for sensitivity to pulmonary infection . Groups of 4–6 wild type C57BL/6 were used to determine the lethal dose for C57BL/6 mice , since this is the parent strain background of our Ccr2−/− mutant mice . These experiments led to a measured LD50 of 4 . 3×106 CFU of Y . pestis yopK following intranasal infection , which is approximately 3-fold higher than that measured for BALB/c suggesting C57BL/6 mice may be less susceptible to Y . pestis yopK in this model ( data not shown ) . We challenged WT and Ccr2−/− mice with 5×106 CFU by intranasal infection of Y . pestis yopK and survival was monitored for up to 14 days . At this challenge dose , 80% of the WT mice survived the infection while Ccr2−/− mice were significantly more sensitive ( Figure 7A ) . Knock-out mice that succumbed to infection did so within 4 days , a time course typical of primary pneumonic plague . We examined bacterial load and histopathology in the lungs of WT and Ccr2−/− mice following intranasal infection with 5×106 CFU Y . pestis yopK at 72 HPI . Only three of nine Ccr2−/− mice survived to that time point , two of which had cleared the infection and the third mouse harbored very high bacterial titer in the lungs , while for WT , more mice survived to 72 HPI , with the surviving mice exhibiting a wide range of titers ( Figure 7B ) . Lungs from moribund Ccr2−/ . mice contained neutrophils and large bacterial colonies in the alveoli suggestive of primary bronchopneumonia while those of wild type mice had smaller bacterial colonies and less inflammation ( Figure 7C–D ) . Together the data suggest that CCR2 signaling helps to limit congestion in the lungs infected by Y . pestis lacking YopK . To further confirm that the effect of YopK in vivo was not due to inhibition of the inflammasome or secretion of pro-inflammatory cytokines , we measured production of MCP-1 and IL-1β following infection of macrophages in vitro . Since RAW and MH-S cell lines showed different cell death phenotypes when infected by the yopK mutant , we tested each for secretion of MCP-1 and IL-1β at 8 HPI following pre-treatment of macrophages with IFN-γ . No detectable differences in MCP-1 secretion were found between RAW or MH-S cells infected with T3SS-competent strains containing or lacking YopK ( Figure 8A–B ) . Levels of IL-1β produced by RAW and MH-S macrophages were independent of infection and no detectable differences in IL-1β secretion were observed between infected and not infected cells ( Figure 8C–D ) . Together with the data presented in figure 1B and 1E , Y . pestis YopK does not appear to prevent or cause the induction of the host inflammasome . Overall , the data support a model whereby the function of YopK in vivo may be to regulate early apoptosis of immune cells which subsequently reprograms the inflammatory response .
The TNF receptor super family is a group of cell surface proteins including TNF-receptor and Fas that are involved in inducing programmed cell death to an extracellular ligand , the extrinsic apoptosis pathway [50] . Yersinia infection has been shown to induce the assembly of the pro-caspase-8-containing death receptor complex , known as the DISC , independent of TNF or Fas receptor signaling , thereby activating caspase-8 ( Figure 9 ) [42] . YopJ is essential for activation of apoptosis due to its ability to shut down production of anti-apoptotic proteins and to stimulate assembly of an atypical DISC . YopK is located at the plasma membrane , associated with the translocation pore , where it could regulate , either directly or indirectly , YopJ activity in promoting DISC assembly or downstream signaling . Resident and recruited macrophages in the lung are believed to be different in their response to apoptotic stimuli , with monocyte-like , recruited macrophages highly sensitive to extrinsic apoptosis while resident alveolar macrophages are somewhat resistant to apoptosis due in part to the expression of NF-κB-induced pro-survival factors [51] . Two populations of macrophages are resident in the mammalian lungs: interstitial and alveolar . Interstitial lung macrophages more closely resemble monocytes , with a greater capacity for immune regulation than alveolar macrophages , and , due to their location in direct contact with epithelial and endothelial cells , they can have pathological impact as well [52] . In contrast , alveolar macrophages exhibit greater functional activity such as increased cytotoxicity , phagocytosis and reactive oxygen species production . MH-S is an alveolar macrophage cell line whereas RAW 264 . 7 phenotypically more closely resembles monocytes , thus it is conceivable that the differences we observed in vitro may have important consequences in vivo [52] , [53] . Although the yopK mutant causes a rapid change to the inflammatory response in vivo , we did not identify YopK-dependent pro-inflammatory cytokines produced by infected macrophages in vitro . This is especially relevant concerning the proposed function of YopK as a regulator of translocation of effector Yops , a model that predicts increased injection of at least 3 immunomodulators: YopJ , YopH and YopM by yopK mutant Yersinia . Although YopJ has well-characterized activity in the prevention of NF-κB activation , suppression of TNF-α secretion and induction of apoptosis in vitro and in vivo , yopJ mutant Y . pestis exhibited little to no virulence defects in rodent models of plague . Our data suggests that YopK may be important as a regulator of YopJ-mediated apoptosis , and that the dysregulation of apoptosis that occurs in the absence of YopK leads to changes in inflammatory responses in vivo that affect the establishment of infection . YopH has been shown to suppress MCP-1 and other pro-inflammatory cytokines in vitro [54] . Yet , we did not find decreased MCP-1 secretion between WT and yopK Y . pestis infection of RAW or MH-S cells , both of which produced MCP-1 in the absence of infection . Together , the data are accumulating that support a critical role for YopK in modulating inflammatory responses through the control of immune cell death . CCR2+ cells are known to be broadly important to host defense against bacterial infections [55] . CCR2 is expressed by many inflammatory cells and allows for their recruitment to the lungs following infection by many respiratory pathogens . Further , pulmonary infection by Y . pestis results in robust MCP-1 secretion ( the chemokine for CCR2 ) in the lungs [56] . Thus , successful respiratory infection and pneumonia that occurs during Y . pestis infection requires evasion of early CCR2 responses . Our data and others indicate the T3SS is critical for CCR2 evasion , with three effectors , YopH , YopM , and now YopK implicated in modulating recruitment of CCR2+ cells . YopH was shown to suppress pro-inflammatory responses in macrophages , and may delay CCR2-signaled recruitment of immune cells [19] . Recently , YopM was also shown to suppress CCR2 inflammatory responses that would otherwise contribute to bacterial clearance from the liver [49] . Thus CCR2 is a critical mediator of host defense against Y . pestis whose effect is blocked by the T3SS . Many immune cells express CCR2 , albeit to varying levels , on the cell surface including monocytes , macrophages , dendritic cells , natural killer T cells , and some B cells . Our observations of the yopK infection suggest that all of these cells may be recruited through MCP-1 signaling and one or more of its receptors . CCR2/MCP-1 have been implicated in both bacterial clearance and resolution of inflammation . Although a role for macrophages and monocytes in resolution of inflammation via phagocytosis of apoptotic neutrophils following bacterial clearance is well known , a role for B cells in resolution has not yet been elucidated . The yopK pulmonary infection model may therefore provide a novel system for studying the role of B cells and other lymphocytes in the resolution of inflammation .
Strains and plasmids used in this study are listed in Supplemental Tables S4 and S5 , respectively . All Yersinia pestis CO92 strains used were routinely grown fresh from frozen stocks and streaked for isolation onto heart infusion agar ( HIA ) plates containing 0 . 005% Congo Red and 0 . 2% galactose to identify bacteria that retained the pigmentation locus [57] , [58] . For pneumonic plague challenge , a single red pigmented colony was used to inoculate heart infusion broth ( HIB ) containing 2 . 5 mM CaCl2 and grown 18–24 hrs at 37°C , 120 rpm . All handling of samples containing live Y . pestis CO92 was performed in a select agent authorized biosafety level 3 laboratory under protocols approved by the University of Missouri Institutional Biosafety Committee . Non-pigmented Y . pestis strains were routinely grown fresh from frozen stocks on HIA , followed by aerobic growth at 28°C in HIB overnight prior to use in experiments . Where indicated , ampicillin ( 100 µg/ml ) or chloramphenicol ( 20 µg/ml , KIM D27 strains only ) was added to media for selection of plasmids . Primers used in this study are listed in Supplemental Table S5 . Deletion of the open reading frame ( ORF ) of yopK in KIM D27 and CO92 was achieved by allelic exchange using the suicide vector pCVD442 and 1 , 000 bp upstream and downstream of yopK [36] . Candidates carrying the deletion of the yopK ORF were confirmed by PCR and Western blot and tested for the presence of all three virulence plasmids ( pCD1 , pMT1 and pPCP1; data not shown ) . Y . pestis CO92 pCD1− was also generated using pCVD442 , confirmed by PCR and found to be avirulent in a mouse model of bubonic plague ( data not shown ) . Complementation of the yopK mutation in Y . pestis KIM D27 was achieved by amplifying the yopK promoter , open reading frame and putative transcriptional terminator from Y . pestis CO92 by PCR and cloning into a derivative of the low copy plasmid pHSG576 [59] . Restriction enzymes were purchased from New England Biolabs ( Ipswich , MA ) . E . coli DH5α was generally used for cloning and S17-1λpir was used for propagating pCVD442 plasmids [60] , [61] . Mutations to residues T45 and D46 of YopK were created using PCR mutagenesis of the complementing plasmid ( pDA15 ) . Point mutations were introduced in primers that were used to amplify pDA15 with Phusion polymerase ( New England Biolabs , Ipswich , MA ) . PCR products were gel purified using QIAquick Gel Extraction kit ( Qiagen , Germantown , MD ) , followed by ligation and transformation into E . coli . Mutants were verified by DNA sequencing . RAW 264 . 7 or MH-S alveolar macrophages ( American Type Culture Collection , Manassas , VA ) were grown in DMEM with phenol red ( Invitrogen , Carlsbad , CA ) and 10% FBS ( Hyclone , Logan , UT ) or RPMI 1640 ( Invitrogen , Carlsbad , CA ) with 10% FBS and 0 . 05 mM 2-mercaptoethanol ( Millipore , Billerica , MA ) , respectively . Culture media for propagating macrophages was supplemented with 1% penicillin/streptomycin ( Invitrogen , Carlsbad , CA ) . Caspase-3 activity was assessed as previously described [62] , [63] . Briefly , RAW 264 . 7 or MH-S ( ATCC , Manassas , VA ) cells ( 1×106 ) were seeded in 12 well plates in DMEM with phenol red and 5% FBS and incubated overnight at 37°C with 5% CO2 . Y . pestis strains were grown in HIB overnight , diluted 1∶10 in HIB containing 2 . 5 mM CaCl2 and grown 2 hours at 28°C . Cultures were shifted to 37°C for 1 hour , and added to RAW 264 . 7 or MH-S cells at a multiplicity of infection ( MOI ) of 20 . Plates were centrifuged for 5 min at 41×g and incubated for 3 . 5 hours , or the indicated times , at 37°C and 5% CO2 . After incubation , cells were harvested , lysed and processed using the EnzCheck Caspase-3 Assay Kit according to manufacturer's recommendations ( Invitrogen , Carlsbad , CA ) . Contribution of JNK , caspase-8 , and caspase-9 to caspase-3 cleavage by Y . pestis KIM D27 and KIM D27 yopK was assessed using the assay described above . One hour prior to infection , fresh media was added to the wells containing the indicated inhibitor , either SP600125 ( 10 µM , Sigma Aldrich , St . Louis , MO ) , IETD ( 50 µM , Sigma Aldrich , St . Louis , MO ) , or Z-LEHD-FMK ( 50 µM , R & D Systems , Minneapolis , MN ) . Caspase-3 activation was detected using EnzChek ( Invitrogen , Carlsbad , CA ) according to the manufacturer's protocol; values were normalized to untreated cells infected with either KIM D27 or KIM D27 yopK . Cytotoxicity was assessed by lactate dehydrogenase ( LDH ) release as previously described [62] . Briefly , cultures of Y . pestis strains were grown as described above . RAW 264 . 7 cells or MH-S cells ( 1×106 ) were seeded in 12-well plates as described above and incubated overnight at 37°C , 5% CO2 . Bacteria were added at an MOI of 20 , plates centrifuged for 5 min at 41×g and incubated for 4 hours , or the indicated times , at 37°C , 5% CO2 . Duplicate wells of macrophages that were not infected were then detergent lysed to determine the maximum amount of LDH signal; 100 µl of supernatant from each sample was removed and analyzed using the CytoTox-One Homogenous Membrane Integrity Assay according to manufacturer's instructions ( Promega , Madison , WI ) . LDH activity was calculated as the amount present in the media divided by the maximum LDH released from detergent-lysed cells . As indicated in the text , infected animals were euthanized , blood was collected by cardiac puncture , and organs removed and divided in half for histology and analysis of bacterial loads . Small lung lobes were analyzed for bacterial titer; the large lung lobe was inflated in situ with 10% formalin for histological analysis . All tissues were fixed for at least 96 hours and then stained with hematoxylin and eosin ( H&E ) . Pathology was assessed by analysis by a veterinary pathologist and scored blindly for severity . Tissues that had been fixed in 10% formalin as described above were sectioned onto slides for immunohistochemical analysis . Slides were stained with anti-rat F4/80 ( Serotec , Oxford , UK ) , anti-rat NIMP R14 ( Ly6G/6C ) , ( Santa Cruz Biotechnology , Santa Cruz , CA ) , anti-rat caspase-3 ( Trevigen , Gaithersberg , MD ) , anti-mouse CD3 ( Dako , Carpinteria , CA ) , anti-mouse B220 ( CD45R ) ( Trevigen , Gaithersberg , MD ) , or anti-mouse CCR2 ( Abcam , Cambridge , MA ) and detection was achieved by secondary staining with biotinylated rabbit anti-rat IgG and HRP-streptavidin ( DAKO , Carpenteria , CA ) . Staining and detection were carried out according to the manufacturer's guidelines . For enumeration of caspase-3 positive alveolar macrophages , slides were blinded and 10 non-overlapping fields were assessed at 40× magnification . To analyze inflammatory cells in the lung , mice were euthanized at 24 HPI and the chest cavity was opened to expose the lungs and trachea . 0 . 5 ml of 2 mg/ml collagenase dispase ( Roche Diagnostics , Indianapolis , IN ) in 1× PBS ( Gibco , Invitrogen , Carlsbad , CA ) was injected through the trachea into the lungs with lungs remaining in situ for 3–5 minutes . Lungs were then removed from the mouse and homogenized in 1 ml additional collagenase dispase solution . Lung homogenate was incubated in collagenase dispase solution for 45 minutes at 37°C shaking at 150 rpm . After incubation , homogenate was strained through a 40 µm filter to remove tissue debris and red blood cells were lysed with a 1× ACK lysis solution for 90 seconds . Cells were resuspended in 100 µl FACS buffer ( 1× PBS with 1% FBS ) for 20 minutes using TruStain ( BioLegend , San Diego , CA ) which blocks Fcγ receptor to prevent non-specific binding of antibody . Cells were then stained with anti-CD11c PE ( BioLegend , San Diego , CA ) at 1∶200 and anti-F4/80 FITC ( BioLegend , San Diego , CA ) at 1∶100 . Cells were washed in FACS buffer and then fixed in 4% paraformaldehyde . Samples were run on a FACSCalibur ( BD Biosciences , San Jose , CA ) and analyzed with FlowJo software ( FlowJo , Ashland , OR ) . Forward and side scatter were used to gate out cellular debris which typically includes dead cells and residual red blood cells . Following euthanasia , blood was collected and organs removed and divided in half as described above . For bacterial titers , tissues were homogenized in sterile PBS , serially diluted and plated in triplicate on HIA . Plates were incubated for 48–72 hours at 26°C . Blood from BALB/c mice was collected at the indicated times post-infection and centrifuged to remove cells . Serum samples were stored frozen at −80°C until analyzed . Cytokine analysis was performed using ELISA or the Premix 13-plex kit ( Millipore , Billerica , MA ) according to manufacturer's instructions and analyzed by Illuminex using IS 100 software ( Qiagen , Camarillo , CA ) . RAW 264 . 7 or MH-S cells ( 1×106 ) were seeded in a 12-well plate as described above and incubated overnight at 37°C , 5% CO2 . Cells were washed and fresh media was added containing 100 U/ml IFN-γ ( Abcam , Cambridge , MA ) followed by incubation for 4 hours prior to infection . Bacterial cultures were prepared as described above then added to macrophages at a MOI of 20; control cells were not infected . Plates were centrifuged at 41×g for 5 minutes at room temperature then incubated at 37°C , 5% CO2 . After 8 hours , 1 ml of supernatant was removed and frozen at −80°C until analysis . Each supernatant was diluted 1∶10 in 1× assay diluent according to manufacturer's directions and analyzed for the presence of MCP-1 and IL-1β ( BioLegend , San Diego , CA ) by ELISA . Cytokine and chemokine concentrations were calculated according to manufacturer's directions using a standard curve . Data from all replicates were analyzed for statistical significance . Survival data were evaluated by the Gehan-Wilcoxon test; all other data were evaluated by Student t-test or one-way ANOVA with Tukey post-test using GraphPad Prism ( GraphPad Software , La Jolla , CA ) .
|
In this work , we studied the mechanism whereby bacteria manipulate innate immune responses by controlling host cell death . Yersinia pestis , the causative agent of plague , requires effector Yops of the Type III Secretion System ( T3SS ) to evade the innate immune system during infection . We show that Yersinia induces apoptosis of macrophages through two distinct mechanisms , each through the activity of the well-characterized T3SS effector YopJ , yet regulated in an opposing manner through the activity of a second effector protein YopK . In a murine pneumonic plague model , we found evidence that YopK regulates apoptosis of macrophages during the early stage of infection , leading to uncontrolled inflammation and disease . In contrast , the absence of YopK-regulated apoptosis allowed recruitment of lymphocytes and CCR2+ immune cells which led to bacterial clearance and resolution of inflammation . Together the data suggest that Yersinia YopK modulates apoptosis of immune cells to control the inflammatory response during plague .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"gram",
"negative",
"emerging",
"infectious",
"diseases",
"immunity",
"innate",
"immunity",
"microbial",
"pathogens",
"immune",
"defense",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"bacterial",
"pathogens"
] |
2013
|
Early Apoptosis of Macrophages Modulated by Injection of Yersinia pestis YopK Promotes Progression of Primary Pneumonic Plague
|
Bacillus anthracis causes three forms of anthrax: inhalational , gastrointestinal , and cutaneous . Anthrax is characterized by both toxemia , which is caused by secretion of immunomodulating toxins ( lethal toxin and edema toxin ) , and septicemia , which is associated with bacterial encapsulation . Here we report that , contrary to the current view of B . anthracis pathogenesis , B . anthracis spores germinate and establish infections at the initial site of inoculation in both inhalational and cutaneous infections without needing to be transported to draining lymph nodes , and that inhaled spores establish initial infection in nasal-associated lymphoid tissues . Furthermore , we found that Peyer's patches in the mouse intestine are the primary site of bacterial growth after intragastric inoculation , thus establishing an animal model of gastrointestinal anthrax . All routes of infection progressed to the draining lymph nodes , spleen , lungs , and ultimately the blood . These discoveries were made possible through the development of a novel dynamic mouse model of B . anthracis infection using bioluminescent non-toxinogenic capsulated bacteria that can be visualized within the mouse in real-time , and demonstrate the value of in vivo imaging in the analysis of B . anthracis infection . Our data imply that previously unrecognized portals of bacterial entry demand more intensive investigation , and will significantly transform the current perception of inhalational , gastrointestinal , and cutaneous B . anthracis pathogenesis .
Bacillus anthracis is a sporulating Gram-positive bacterium that causes the disease anthrax . The three forms of anthrax reflect the route by which the infection is initiated: cutaneous , gastrointestinal , and inhalational [1 , 2] . Anthrax is characterized by both toxemia , which is caused by secretion of two toxins , and septicemia , which is associated with bacterial encapsulation . The capsule consists of poly-γ-d-glutamic acid attached to the peptidoglycan of the cell wall and inhibits phagocytosis , functions as a non-immunogenic surface , and is vital for full virulence [3–5] . The toxins , lethal toxin and edema toxin , modulate host immune responses and at high doses can cause death [6 , 7] , but elimination of toxin production does not alter virulence in a mouse model of B . anthracis infection [8 , 9] . Surprisingly , relatively little is known about how spores enter the host or how the dynamics of infection are affected by the route of infection , and no animal model exists for gastrointestinal anthrax at this time . Cutaneous anthrax , the most common yet least lethal form of anthrax in humans , is believed to initiate through abrasions in the skin [10] . Gastrointestinal anthrax is generally considered to be the primary route of infection of livestock , can occur in humans , and is caused through the ingestion of contaminated food [2 , 11] , yet the means by which B . anthracis crosses membrane barriers to establish infection remains unknown . The current model of inhalational anthrax consists of uptake of spores by alveolar macrophage , then transport of these spores to the draining mediastinal lymph nodes , where the spores germinate and establish infection within the lymphatics to ultimately disseminate systemically [10 , 12 , 13] . To identify the portals of initial entry and growth and to better define the differences associated with these three forms of B . anthracis infection , a model using in vivo bioluminescent imaging ( BLI ) was developed . BLI consists of detecting photons emitted from a cell expressing luciferase—in this case , a bacterium—within the body of a host animal [14] . BLI analysis of B . anthracis infection provides the advantages of tracking a dynamic infection in a dynamic fashion , monitoring the entire mouse ( allowing the discovery of as yet-unknown locations of bacterial growth ) , and detecting subclinical infections . Thus , BLI permits a kinetic and global view of bacterial dissemination that allows the synthesis of an integrated infection model from inoculation to death . We found that , contrary to the current view of B . anthracis pathogenesis , B . anthracis spores germinate and establish infections at the initial site of inoculation in both inhalational and cutaneous infections without needing to be transported to draining lymph nodes . Furthermore , we found that Peyer's patches in the mouse intestine are the primary site of bacterial growth after intragastric inoculation , establishing , to our knowledge , the first animal model of gastrointestinal anthrax . Our data imply that previously unrecognized portals of bacterial entry demand more intensive investigation .
All experiments were performed with a bioluminescent derivative of B . anthracis strain 9602P ( described previously in [8] ) , which is a derivative of the highly virulent natural human isolate 9602 [15] , but does not produce the protective antigen component of the toxins . Bacteria were grown on brain–heart infusion ( BHI ) agar ( Difco , http://www . bd . com/ds ) unless otherwise noted . Spores were produced and purified on Radioselectan ( Renografin 76%; Schering , http://www . schering . de ) using previously described methods and stored in sterile deionized water [16] . The luxABCDE operon was PCR amplified from pXEN5 ( pAUL-A Tn4001 luxABCDE Kmr from the Xenogen Corporation , http://www . xenogen . com ) [14] using the following oligonucleotides: LuxAF1BamXho-CGCGGATCCTCGAGCAGATGAAGCAAGAGGAGGACTC , which contains a BamHI followed by an XhoI site ( underlined ) ; LuxER1Bam- GGCGGATCCGTCGACTTAACTATCAAACGCTTCGG , which contains a BamHI site . This PCR product was ligated with the pGEM-T Easy vector kit ( Promega , http://www . promega . com ) and transformed into Escherichia coli strain TG1 ( pIG3–40 ) . The pIG3–40 plasmid was subsequently purified , and the luxABCDE insert was removed via BamHI ( New England Biolabs , http://www . neb . com ) digestion and ligated into plasmid pAT113 [16] , a conjugative shuttle vector that has no Gram-positive origin of replication ( pIG4–2 ) . The pag promoter , consisting of 1 kb of sequence upstream of the protective antigen coding sequence , was PCR amplified from B . anthracis DNA using the following oligonucleotides: PPagF1EcoR1-CGAGAATTCGATGAAAATGGTAATATAGCG , which contains an EcoRI site , and PPagR1Xho- AGGGCTCGAGGTATAAAATTAAATTTATATTATATTG , which contains an XhoI site . This PCR product was cut with EcoRI and XhoI , ligated into pIG4–2 , cut by equivalent enzymes , and transformed into TG1 ( pIG6–19 ) . Subsequently , pIG6–19 was introduced into B . anthracis 9602P by heterogramic conjugation using previously described methods and the conjugative E . coli strain HB101 ( pRK212 . 1 ) [16] , and selected on BHI plus 5 μg/ml erythromycin , yielding strain BIG19 . Insertion into the upstream sequence of pag was verified by both PCR and the production of light when grown on capsulation plates in 5% CO2 at 37 °C [17] . The stability of the lux integration was assessed by diluting cultures 1:1 , 000 in R-media with 0 . 6% sodium bicarbonate ( R-Bic ) [18] at 37 °C and 5% CO2 two times daily for 1 wk , then detecting the presence of the lux insertion by erythromycin resistance and luminescence . All clones contained the insert after 1 wk of passage . Determination of bioluminescence in comparison to growth phase was performed by growing BIG19 in BHI plus 0 . 5% glycerol media overnight at 37 °C with agitation . Subsequently , this culture was diluted into R-Bic to an optical density ( OD ) of 0 . 05 , and then incubated at 37 °C in 5% CO2 with agitation . At the indicated times , samples were removed and imaged for luminescence in a 96-well plate ( Nunc , http://www . nuncbrand . com ) with the IVIS 100 system as described below , and the OD600 was measured with a spectrophotometer ( Ultrospec 3300 pro; Amersham Biosciences , http://www . gelifesciences . com ) . Six- to 10-week-old BALB/cJ mice ( Charles River Laboratories , http://www . criver . com ) were maintained under specific pathogen-free conditions at the Pasteur Institute in compliance with European animal welfare regulations . Cutaneous infections were performed under light anesthesia by injecting 10 μl of spore suspension in PBS into the dermis of the right ear with a 0 . 5-ml insulin syringe ( Becton Dickinson , http://www . bd . com ) [19] . For the intravenous infection experiments , capsulated bacteria were prepared by germinating spores in BHI for 5 min at room temperature , centrifuging , and incubating in R-Bic induction media at 37 °C in 5% CO2 with agitation for 1 . 5 or 5 h . The length of bacteria in each sample was quantified by binding formalin-fixed bacteria to polylysine-coated glass cover slips , acquiring digital microscopic images , and then measuring the length of the bacteria in each image . Injection cultures were washed with RPMI-Glutamax ( Invitrogen , http://www . invitrogen . com ) , and 1 . 4 × 107 and 2 . 0 × 106 colony-forming units ( CFU ) of the 1 . 5- and 5-h cultures , respectively , were injected . Seven times less CFU of the 5-h culture was injected to account for the fact that the 5-h bacteria were approximately seven times longer than the 1 . 5-h bacteria , and thus CFU were not an accurate measure of bacterial mass . Spores in 100 μl of PBS were inoculated intragastrically by inserting a 5-cm length of polyethylene catheter that was attached to a needle and syringe down the esophagus and into the stomach of lightly anesthetized mice , changing the catheter for each mouse . Alternately , 18G animal feeding needles ( Popper and Sons , http://www . popperandsons . com ) were used for intragastric inoculation . Aerosol infections were performed by delivering spores for 20 min using a Raindrop nebulizer ( Tyco , http://www . tyco . com ) designed to produce 1-μm droplets of a solution containing 1 × 108 spores/ml water in an all-glass , nose-only aerosol chamber [20] . An average 3 . 8 ± 0 . 3 log10 ( mean ± standard deviation [SD] ) spores were found in the lungs 2 h post-infection . Intranasal inoculation was performed in lightly anesthetized mice by introducing 1 × 105 spores in 20 μl of PBS upon inhalation into the right nostril . An average of 4 . 5 ± 0 . 1 log10 ( mean ± SD ) spores were found in the lungs 2 h post-infection . Intratracheal inoculation of 1 × 105 spores was performed using plastic tubing as previously described [21] . An average of 3 . 8 ± 0 . 2 log10 ( mean ± SD ) spores were found in the lungs 2 h post-infection . For CFU determination , organs were immediately placed in ice-cold saline solution and then homogenized in a chilled glass tube , and dilutions were plated on BHI plates . Differentiating spores from vegetative bacteria was determined after heat treatment of 65 °C for 30 min to kill the vegetative bacteria . Lymph node location and nomenclature were based on previously defined standards [22] . Lungs were isolated by cutting below the tracheobronchal bifurcation where the bronchus enters the lobes of the lung . Images were acquired using an IVIS 100 system ( Xenogen ) according to instructions from the manufacturer . Analysis and acquisition was performed using Living Image 2 . 5 software ( Xenogen ) . Unless otherwise noted , mice were anesthetized using a constant flow of 2 . 5% isofluorane mixed with oxygen using an XGI-8 Gas Anesthesia System ( Xenogen ) , which allows control of the duration of anesthesia . Images were acquired with a binning of 16 . Luminescent signals from the exterior of mice or cultures in plates were acquired for 1 min , whereas luminescence of internal organs during dissection was integrated for 10 s . All other photographic parameters were held constant . Quantifying photons per second emitted by each organ was performed by defining regions of interest corresponding to the organ of interest , while subtracting the background as defined by a non-infected animal in the same photograph . Tissues were removed from mice with luminescence in the organ of interest and immediately placed in 4% buffered formalin ( Labonord , Templemars , France ) for standard microscopic analysis . Skulls of mice were decalcified in a solution of 4% buffered formalin and 10% trichloroacetic acid for approximately 2 mo for analysis of the nasal passages . Serial 5-μm sections were stained with hematoxylin and eosin and/or Gram [23] . Samples were analyzed by two independent observers , one of whom was a trained pathologist . Statistical analysis and graphing was performed using GraphPad Prism 4 software ( http://www . graphpad . com ) . The Pearson correlation coefficients ( r ) were used to determine the direction and magnitude of covariation between CFU and photons/s-organ , since each X and Y value were independent measures and are assumed to follow a Gaussian distribution . The two-tailed p-values , α , and n are indicated on each graph in Figure 1F .
A capsulated non-toxinogenic strain of B . anthracis was chosen for study because the capsule is vital for virulence and is associated with dissemination [3–5] , while elimination of toxin production does not alter virulence in mouse models of infection [8 , 9] . Bioluminescent B . anthracis were constructed by integrating the luxABCDE genes from Photorhabdus luminescens [14] under the control of the pag promoter , the gene for protective antigen , a toxin component that is highly induced under in vivo conditions [18] . These bacteria were strongly luminescent ( Figure 1A ) , whereas dormant spores were not ( unpublished data ) , reflecting the necessity of bacterial metabolic activity for light production . Maximal luminescence was observed during mid-log phase growth and decreased in late-log and stationary phase ( Figure 1B ) . Spores derived from these bioluminescent bacteria were used in subsequent murine bacterial dissemination models . Cutaneous infection is the most common form of anthrax in humans and is often limited to the site of initial infection [1] . A cutaneous ear infection model [19] allowed efficient enumeration of CFU and high sensitivity . Luminescence at the site of injection was detected within 120 min ( Figure 1C ) . This shows that spores rapidly germinate at the site of injection and quickly become metabolically active , and that the pag promoter is induced; these results are similar to those obtained in tissue culture models [24 , 25] , but unlike previous studies that suggested that germination occurs in the draining lymph nodes [7 , 10] . Within 24 h of cutaneous injection of 500 spores , bioluminescence was detected in both the ear and the superficial parotid lymph node draining the ear , and was also visible in the spleen and right lung 12 h later ( Figure 1D ) . Dissection confirmed that the superficial parotid lymph node , spleen , both lungs , and the blood were luminescent ( Figure 1E ) . The left lung , though luminescent , was more difficult to detect from the exterior because the heart partially blocked the exit of photons . No other organs harbored luminescent B . anthracis . Differences in pigmentation , organ orientation , and tissue depth affect how light exits the body , necessitating determination of the relationship between luminescence and quantity of bacteria for each organ of interest . Luminescence correlated with CFU in the ear , superficial parotid lymph node , spleen , and right lung ( Figure 1F ) . We analyzed only the right lung because the heart blocked a large portion of the left lung's emitted light . The minimum detectable quantity of bacteria in each organ was as follows: ear , 3 × 103; superficial parotid lymph node , 4 × 103; spleen , 2 × 105; and right lung , 2 × 104 . Thus , an increase in luminescence at a particular location can be interpreted as an increase in the bacterial load . Lung luminescence late in infection was unexpected from cutaneous infections . This may indicate that bacteria growing in distal sites enter the blood and are trapped in the lungs [5] , which is the first fine capillary bed encountered upon the return of blood from the periphery . Bacterial entry into the blood was simulated by intravenous injection . Luminescent capsulated bacteria grown 5 h in inductive media , averaging 20 . 0 ± 5 . 5 μm ( mean ± SD ) in length and injected intravenously , were detected solely in the lungs from 2 min post-injection until death ( Figure 2A and 2B ) , and killed the mice within 40 min . Bioluminescence was not detected in any other organs , suggesting that a majority of the bacteria were immediately trapped in the lungs . In contrast , light emitted by bacteria grown 1 . 5 h , averaging 3 . 2 ± 0 . 7 μm in length , was initially below the detection threshold . After 1 h , bioluminescence was detectable in the spleen , liver , and lungs . These data suggest that bacteria from the 1 . 5-h culture were initially widely distributed throughout the mouse and were eventually localized within the spleen , liver , and lungs . This implies that the size of bacteria entering the blood circulation and/or growth phase influences the dissemination of bacteria to the lungs and other organs . Anthrax is associated with polygastric ruminant livestock , and natural gastrointestinal infection occurs when food contaminated with B . anthracis spores is ingested [1] . Likewise , gastrointestinal anthrax occurs in monogastric hosts , such as humans [2 , 15] , yet there are no published animal models of gastrointestinal anthrax . Intragastric inoculation of spores in the monogastric mouse using a rigid feeding needle caused infection of the laryngopharynx ( Figure 3A ) , suggesting that abrasions in the upper alimentary tract acted as portals of infection . Intragastric inoculation with flexible plastic tubing caused true gastrointestinal infection that initiated as a defined point of luminescence in the abdomen from approximately 24- to 33-h post-inoculation that suddenly progressed systemically ( Figure 3B ) . Histological examination of the initial site of infection showed that bacterial growth occurred in the Peyer's patch ( Figure 3C and 3D ) , and could occur in Peyer's patches throughout the small intestine ( unpublished data ) . Very little of the original dome architecture was maintained , with only a few lymphocytes at the periphery . Bacterial rods , numerous polymorphonuclear cells , necrosis , and hemorrhage were evident ( Figures 3C and S1 ) . Similar lesions were observed in the spleen and lung ( Figure S1 ) . Interestingly , vegetative bacteria could be isolated from the feces 4 h post-inoculation ( Figure 3E ) , but only spores were found 8 h post-inoculation . These data suggest that spores germinate in the gastrointestinal tract , vegetative bacteria transiently colonize the lumen , and spores are continuously shed for up to 48 h post-inoculation after a majority of the initial inoculum has been excreted . As in the cutaneous model , the ensuing systemic infection spread first to the draining jejunal lymph nodes , then the spleen , and finally the lungs ( Figure 3B–3D ) . Inhalational anthrax has a high mortality rate and is the form of disease that is targeted when spores are weaponized [26] . Dissemination kinetics of aerosol , intranasal , and intratracheal inoculation of spores were studied . Mice infected with aerosolized spores developed an infection in the nasal cavity by 22 h post-infection , followed by the mandibular lymph nodes 5 h later ( Figure 4A and 4B ) . When bioluminescence was first visible in the nasal cavity , it was not found in the lungs and mediastinal lymph nodes ( Figures 4B and 5B ) , despite the effective delivery of spores into the lungs ( see Materials and Methods ) and the presence of a small quantity of heat-resistant spores in the tracheobronchal mediastinal lymph nodes ( Figure 4C ) . Intranasal inoculation of spores also caused infection of the nasal cavity by 29 h post-infection , followed by the mandibular lymph nodes 15 h later ( Figure 5A and 5B ) . Histologic analysis revealed that bacteria were growing in nasal-associated lymphoid tissues ( NALTs ) when bioluminescence was first detected in the nasopharynx ( Figure 5C ) . Inoculation of spores by intubation of the trachea caused two distinctly different patterns of infection . In the first form , the primary site of early luminescence was the mandibular lymph nodes ( Figure 6A ) , likely representing initiation of infection by spores that were expelled from the lung by coughing and/or the ciliary escalator [27] . The second form of infection initiated in the larynx within 22 h post-infection and progressed to systemic disease within 9 h ( Figure 6B and 6C ) . Intubation of the trachea ( via the larynx ) may cause abrasions that allowed direct mucosal infection , as previously noted [12] , and which we also observed in the gastrointestinal model ( Figure 3A ) . As in the cutaneous and intragastric models , the bioluminescence in all inhalational models progressed from the site of initial infection to systemic distribution ( Figures 4A and 5A ) .
Real-time analysis of capsulated B . anthracis infection has allowed the synthesis of a coherent model of murine infection from initial infection to death that modifies the current paradigm of B . anthracis dissemination; primary bacterial growth is limited to locations associated with the inoculation site , followed by dissemination to the draining lymph nodes , then the spleen , and ultimately the lungs and blood . In agreement with J . M . Ross's seminal study of inhalational infection of guinea pigs [12] , our data indicated no bacterial growth within the lungs at early stages of infection in inhalation models . These results , along with our observation that intravenous bacteria rapidly associate with the lungs , support the hypothesis that vegetative bacteria arrive in the lungs via the blood [5 , 28] , rather than in situ growth from spores . Unexpectedly , our results showed that metabolically active bacteria were found solely in the nasopharynx and its associated lymph nodes early in infection , which is contrary to previous studies that suggested that bacteria multiply primarily in the mediastinal lymph nodes at the initiation of infection [12 , 13] . These previous studies note that there were incisions or trauma to the airways caused by experimental technique , which makes interpretation of early events in the infection difficult , as we found that abrasions were preferential portals of infection ( Figures 3A , 6B , and 6C ) and may allow bacterial access to the lymph nodes via direct transport to the lymphatics across the damaged epithelial barrier . Indeed , in Ross's study , spores in transit to the regional lymph node are only observed in infections initiated through intratracheal intubation , a technique that she noted to cause tracheal injury , and were not observed in transit when a non-invasive spore aerosol was used [12] . BLI analysis of B . anthracis infection provided the advantage of differentiating these unintended sites of infection from those infections initiated in targeted tissues . Furthermore , other studies that support mediastinal growth have derived conclusions from autopsy of animals after death caused by the infection [28 , 29] , which could confound interpretation , since we observed a sudden increase in bacteria in the lungs late in infection . The bacteria found in the mediastinal lymph nodes late in infection or at death could otherwise be explained as originating from the population of bacteria that spread to the lungs hematogenously . Regardless of the method of spore delivery ( cutaneous , gastrointestinal , intravenous , or by inhalation ) we found that the lung eventually contained large numbers of bacteria that may subsequently drain into the mediastinal lymph nodes . Supporting this interpretation , Zaucha et al . noted significant mediastinal lymph node involvement in rabbits infected subcutaneously [28] . Thus , our investigation also implies that the timing of when tissue samples are taken is vital , indicating an additional advantage of using dynamic models of infection such as BLI . Our findings demonstrate that a severe infection with highly metabolically active bacteria develops in the nasopharynx of mice at a time in the infection when low to undetectable quantities of spores are in the mediastinal lymph nodes . Thus , it is reasonable to question the relevance of the bacteria found in the mediastinal lymph nodes to the subsequent disseminated disease when at an earlier time-point a much larger infection is occurring in the nasopharynx . Moreover , large quantities of spores can persist in the lungs for long periods without causing disease ( [30]; unpublished data ) , also bringing into question the relative efficacy of initiation of infection from lung tissues . Additionally , we observed bacteria growing within the NALT , which may indicate that bacteria can directly invade tissues of the nasopharynx without needing to be transported by alveolar macrophages to the lymphatics , as previously postulated [31] . Indeed , there has been growing evidence that B . anthracis can directly infect non-phagocytic cells [32] . Therefore , our results suggest that B . anthracis–host interactions in the nasopharynx , and in particular with the NALT , are critical and should be the focus of future research . The data obtained through gastric inoculation of spores reflect many of the characteristics of human gastrointestinal anthrax , including two forms of infection ( oropharyngeal and lower gastrointestinal ) , intestinal abscesses , hemorrhage , necrosis , and epithelial barrier breakdown [2] . To our knowledge , this is the first establishment of a mouse model of gastrointestinal B . anthracis infection . This model should prove particularly useful in addressing questions regarding B . anthracis pathogenesis since mice offer a wide variety of immunologic and genetic tools that permit the determination of specific host factors involved in host–pathogen interactions . Furthermore , we observed that while abrasions to the mucosa were preferentially infected , gastrointestinal infection also occurred in the absence of mucosal damage , albeit with slower kinetics . Indeed , it has been hypothesized that the tendency of anthrax outbreaks to begin after drought conditions is due to abrasions in the mucosa caused by the consumption of sharp dried plants [11] . Similarly , preexisting lesions may predispose humans to infection [28 , 29] . The means by which B . anthracis invades the Peyer's patch in the absence of mucosal damage remains unknown , but may include uptake by M cells or dendritic cells [33] . In vivo bioluminescent imaging of B . anthracis has numerous possible applications . Some of these applications include real-time visualization of the effects of antibiotic therapy at different stages of B . anthracis infection , which is of particular interest because anthrax diagnosis often occurs at late stages of the disease , leading to ineffective treatment [11] . Another potential application is the analysis of different anti-anthrax vaccination strategies to determine where , when , and how efficiently the induced protective adaptive immunity terminates infection . Taken together , the data presented in this manuscript suggest that the current paradigm of B . anthracis infection should be revisited . Unexpectedly , we found that B . anthracis spores germinate and establish infections at the initial site of inoculation in both inhalational and cutaneous infections , and that mice initially develop infections in Peyer's patches upon intragastric inoculation of spores . Since the elimination of toxin production does not affect virulence in the mouse model of infection [8 , 9] , one would predict that our models are valid representations of infections with toxinogenic capsulated strains , though whether differences will be seen in the presence of both toxin production and capsule synthesis remains to be determined experimentally .
The National Center for Biotechnology Information ( http://www . ncbi . nlm . nih . gov ) accession numbers for the genes and gene products discussed in this paper for B . anthracis are protective antigen , pagA ( AF268967 ) ; lethal factor , lef ( M29081 ) ; Edema factor , cya ( M24074 ) ; capsule synthesis operon , capBCAD ( D85765 ) , and capE , which can be found on the sequence of plasmid pX02 ( AF188935 ) . The number for the P . luminescens luciferase operon , luxCDABE , is M90093 .
|
Anthrax is caused by Bacillus anthracis , a bacterial pathogen that forms spores , dormant bacteria that are highly resistant to destruction . Infections initiate from the introduction of spores into airways or damaged skin , or from the consumption of contaminated food . Within the host , spores germinate , then bacteria secrete toxins that cripple the immune response and sheath themselves in a capsule that prevents them from being phagocytosed . We strove to determine in real space and time where and when spores introduced by these three routes of infection germinate and how bacteria subsequently disseminate in a mouse model . This was achieved through the development of light-emitting B . anthracis that could be tracked inside a living mouse . Contrary to current models , our studies indicated that spores germinated in situ in the skin , the intestines , and the nasal passages without needing to be transported to lymph nodes . Furthermore , bacteria disseminate from initial sites of infection in a similar fashion , first to the draining lymph nodes , then the spleen , and finally the lungs and blood . These findings imply that spore interactions with local sites of entry are critical in the development of systemic disease and that disruption of these interactions may offer new methods of anthrax prevention .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion",
"Supporting",
"Information"
] |
[
"mus",
"(mouse)",
"infectious",
"diseases",
"microbiology",
"eubacteria"
] |
2007
|
Primary Involvement of Pharynx and Peyer's Patch in Inhalational and Intestinal Anthrax
|
Despite a high current standard of care in antiretroviral therapy for HIV , multidrug-resistant strains continue to emerge , underscoring the need for additional novel mechanism inhibitors that will offer expanded therapeutic options in the clinic . We report a new class of small molecule antiretroviral compounds that directly target HIV-1 capsid ( CA ) via a novel mechanism of action . The compounds exhibit potent antiviral activity against HIV-1 laboratory strains , clinical isolates , and HIV-2 , and inhibit both early and late events in the viral replication cycle . We present mechanistic studies indicating that these early and late activities result from the compound affecting viral uncoating and assembly , respectively . We show that amino acid substitutions in the N-terminal domain of HIV-1 CA are sufficient to confer resistance to this class of compounds , identifying CA as the target in infected cells . A high-resolution co-crystal structure of the compound bound to HIV-1 CA reveals a novel binding pocket in the N-terminal domain of the protein . Our data demonstrate that broad-spectrum antiviral activity can be achieved by targeting this new binding site and reveal HIV CA as a tractable drug target for HIV therapy .
Highly active antiretroviral therapies ( HAART ) against human immunodeficiency virus type 1 ( HIV-1 ) have proven in recent years to be extremely effective at reducing viral load and significantly delaying disease progression [1] . However , there remains a pressing need to discover and develop new classes of HIV inhibitors . The virus continues to acquire resistance to currently administered antiretroviral drugs and the rate of transmitted resistance is increasing [2] , [3] . The discovery of compounds that inhibit the replication of HIV-1 via new mechanisms offers the best hope of generating drugs that are active against all HIV-1 variants in the clinic . The potency of these compounds would not be affected by mutations that confer resistance to existing therapies [4] . The capsid protein ( CA ) of HIV-1 plays critical roles in both late and early stages of the viral replication cycle and is widely viewed as an important unexploited therapeutic target [4] , [5] , [6] . At the earliest stages of particle assembly , the interactions between CA domains of the Gag polyprotein help drive the formation of immature particles at the membrane of host cells [7] . After the release of immature particles from infected cells , proteolytic processing of the Gag polyprotein is completed , leading to capsid assembly and formation of the mature virus . During assembly , the viral RNA genome is packaged into a capsid particle composed of a lattice of CA protein hexamers that form a distinct fullerene cone shaped particle [8] . After virus fusion with a target cell , the core is released into the cytoplasm and CA is thought to undergo a controlled disassembly reaction in order for reverse transcription of the viral genome to occur properly [9] . The HIV-1 CA protein has attracted increased interest as a drug discovery target in recent years . A small molecule , CAP-1 , and two versions of a peptide inhibitor , CAI and NYAD-1 , have been described that target HIV-1 CA in vitro and appear to interfere with CA function in infected cells [10] , [11] , [12] . In addition , high resolution structural data on the hexameric lattice that forms the full core structure has been reported [13] , [14] . These structures illustrate the distinct roles and importance of inter-subunit interfaces in the CA complex and have shed some light on the potential mechanisms of previously reported CA assembly inhibitors . Here we describe a novel series of antiviral compounds that target HIV-1 CA in infected cells and appear to interfere with both the viral uncoating process and the formation of infectious particles . Mechanism-of-action and in vitro resistance studies of this series are described . A high resolution co-crystal structure has been determined and illustrates a novel binding pocket in the N-terminal domain ( NTD ) of HIV-1 CA that is distinct from any previously described . We demonstrate that targeting this new binding pocket with small molecules results in broad-spectrum antiviral activity . This study provides the starting point , where structure-based drug design is a viable option , for the development of a new class of HIV therapeutics .
PF-1385801 was identified as a hit in a high throughput screen for inhibitors of HIV replication [15] . Several analogs of this compound demonstrated activity in antiviral assays using the MT-2 T-cell line and HIV-1 NL4-3 . PF-1385801 inhibited HIV-1 replication with a 50% effective concentration ( EC50 ) of 4 . 5 µM and exhibited a 50% cytotoxic concentration ( CC50 ) of 61 µM , resulting in a therapeutic index ( TI , CC50/EC50 ) of 14 ( Fig . 1 ) . More potent analogs were subsequently designed and synthesized , as detailed in Fig . 1 . A key issue in the development of novel HIV drugs is the potential therapeutic spectrum ( consistency of activity across multiple strains of HIV ) . To address this , two of the compounds were evaluated in antiviral assays using peripheral blood mononuclear cells ( PBMCs ) infected with a diverse set of HIV-1 lab strains and clinical isolates covering 6 clades and both X4 and R5 tropic viruses ( Fig . 2 ) . PF-3450074 and PF-3759857 were active against all strains of HIV-1 tested with median EC50 values of 0 . 207 ( range 0 . 113 to 0 . 362 µM ) and 1 . 17 ( range 0 . 51 to 3 . 17 ) µM , respectively ( Fig . 2 and Tables S1 and S2 ) . In addition , PF-3759857 was active against HIV-2 with an EC50 of 4 . 7 µM . This tight spectrum of activities against HIV-1 compared well with the marketed drugs , AZT , a nucleoside analog , and efavirenz ( EFV ) , a non-nucleoside reverse transcriptase inhibitor ( NNRTI ) ( Fig . 2 and Tables S3 and S4 ) . To positively identify the antiviral target of this class of inhibitors , PF-1385801 resistant viral variants were selected in in vitro serial passage experiments . Sequence analysis of cDNAs derived from resistant viral variants selected in the presence of PF-1385801 revealed a single mutation , T107N , located in the NTD of the CA protein . No mutations associated with resistance selection were identified in the integrase or reverse transcriptase coding sequences . Recombinant HIV-1 NL4-3 virus encoding the T107N substitution in HIV-1 CA exhibited an 11-fold reduction in susceptibility to PF-1385801 when compared to wild-type NL4-3 ( Table 1 ) . Similar reductions in susceptibility were observed for PF-3450071 ( 6-fold ) and PF-3759857 ( 10-fold ) , while no shift was observed for EFV ( NNRTI ) or the integrase inhibitor AG-110079 . In subsequent in vitro resistant virus selection experiments using a compound with a larger TI , PF-3759857 , additional substitutions were identified in CA ( T107N , H87P , Q67H , K70R and L111I ) , after 53 days in culture . Virus isolated from the final passage of this experiment showed a >60-fold reduction in susceptibility to PF-3759857 and recombinant NL4-3 virus encoding the five substitutions ( 5M ) exhibited a >40-fold reduction in susceptibility to PF-3450074 ( up to the cytotoxicity limit of the compound ) . Wild-type NL4-3 , the T107N mutant , and the 5M mutant virus all exhibited similar levels of infectious virus production from transfected cells and comparable levels of viral replication in reporter gene-based infection assays ( in the absence of compound ) , suggesting that the substitutions do not significantly impair replication capacity . However , more detailed studies , such as direct competition assays are required to accurately determine if the substitutions affect replication fitness . These results demonstrate that mutation ( s ) in HIV-1 CA ( e . g . T107N ) are sufficient to confer resistance to this new class of inhibitors and demonstrate that HIV-1 CA is the target in infected cells . Although additional work is needed to determine the individual contributions of the mutations identified in addition to T107N , it appears that more than one mutation is required to generate high levels of resistance to this inhibitor class . While it is a definitive tool for target identification , it should be noted that in vitro selection can be used to generate resistant mutants to any known antiviral and does not accurately predict the clinical barrier to resistance , which is a function of several in vivo parameters , including the pharmacokinetics of the compound . To determine the stage of the replication cycle targeted by this new class of compounds , PF-3450071 and PF-3450074 were evaluated in single cycle infection assays . Such assays monitor the early steps of infection up to the integration of the viral cDNA into the host cell chromosome and expression of that sequence . PF-3450071 and PF-3450074 were tested in parallel infections with single-cycle HIV-1 virus packaged with either wild type HIV-1 envelope ( NL4-3 pseudovirus ) or vesicular stomatitis virus glycoprotein ( VSVG ) , an envelope that allows viral entry by an alternative mechanism ( VSV pseudovirus ) . In addition , three compounds with known mechanisms , AMD3100 ( HIV CXCR4/entry inhibitor ) , EFV ( NNRTI/early mechanism ) and NFV ( protease inhibitor ( PI ) /late mechanism ) were evaluated in the assays . PF-3450071 and PF-3450074 were active against both the NL4-3 and VSV pseudoviruses ( Table 2 ) . This inhibition profile was similar to EFV . The profile was distinct from that of AMD3100 , which inhibited the HIV-1 envelope NL4-3 pseudovirus , but not the VSVG pseudovirus , and NFV which was not active in the assay ( Table 2 ) . These data indicate that PF-3450071 and PF-3450074 act early in the HIV-1 replication cycle at a step following HIV-1 envelope-mediated entry . To begin dissecting the early events of infection , DNA was isolated from single cycle infections of MT-2 cells in the presence or absence of inhibitors and analyzed by quantitative PCR . Primer sets were used that detect total viral cDNA produced , 2-LTR circles ( a nuclear episomal form of the viral cDNA ) , or provirus ( the viral genome integrated into host cell chromosomes ) as described previously [16] . PF-3450074 inhibited the accumulation of total viral cDNAs ( 9% of control ) and as result inhibited the levels of integrated provirus DNA measured ( 2% of control ) ( Fig . 3a ) . This profile was similar to that observed for the RT inhibitor EFV , indicating that PF-3450074 either inhibits reverse transcription or a step prior to it . In contrast , the integrase inhibitor , AG-110079 , resulted in a 4-fold increase in 2-LTR circle accumulation and minimal reduction in total viral cDNA ( 78% of control ) ( Fig . 3a ) . As the profile for PF-3450074 did not match that of the integrase inhibitor control , integrase inhibition can be ruled out as a possible mechanism . To further analyze the mechanism of action , the inhibition profile of PF-3450071 was compared to that of AMD3100 and EFV in synchronized time-of-addition experiments , where inhibitors are added a various time points following infection to monitor when the susceptible step has occurred . The viral entry inhibitor , AMD3100 , showed a dramatic early loss in activity , even within the first two hours of infection . The NNRTI , EFV , retained the majority of inhibitory activity ( >78% ) when added up to 4 hours after infection and lost activity thereafter , with a mid-point around 6 hours after infection . PF-3450071 displayed a profile that was distinct from both comparators . The compound maintained the majority of its inhibitory activity ( >84% ) when added up to 2 hours after infection , and lost significant levels of activity when added 3 or more hours after infection ( Fig . 3b ) . These results suggest that this new class of HIV-1 inhibitor targets an early step in the virus replication cycle , after viral entry but before reverse transcription , possibly viral uncoating . Consistent with these data , PF-3450074 did not inhibit recombinant HIV-1 RT in standard biochemical assays ( IC50>100 µM ) . Virus production-infectivity assays were conducted to determine if compounds affect the late stages of viral replication . Infectious virus was expressed from transfected producer cells in the presence or absence of compound and the infectivity of the resulting virus was tested by diluting the supernatant on to an indicator cell line . As infectious virus is produced from a transfected DNA , bypassing the early stages of the replication cycle , only inhibitors targeting the later stages of viral replication ( i . e . post-integration ) should show activity in the assay . We tested two compounds ( PF-3450071 and PF-3450074 ) in the viral production-infectivity assay using NFV ( PI ) and EFV ( NNRTI ) as late stage and early stage inhibitor controls . Both PF-3450071 and PF-3450074 inhibited the production of infectious virus with EC50 values of 0 . 78 and 0 . 33 µM , respectively ( Table 2 ) . As expected the HIV-1 PI , NFV , inhibited infectious virus production while the RT inhibitor , EFV was not active in the assay ( Table 2 ) . Notably , the overall levels of Gag proteins released into the supernatant of transfected cells were not affected by PF-3450074 , as measured by p24 ELISA ( Fig . 3c ) . Thus , this series does not inhibit HIV-1 particle production but renders the nascent particles noninfectious . Western blot analysis of viral supernatants with antibodies directed against the HIV-1 CA protein ( p24 ) showed that PF-3450071 did not affect HIV-1 Gag proteolytic processing ( Fig . S1 ) , which demonstrates that the compounds do not inhibit HIV-1 protease or virion maturation in the manner described for inhibitors such as Bevirimat ( PA-457 ) [17] and PF-46396 [18] . Collectively , these data suggest an effect of the series on proper assembly of fully processed CA protein in nascent viral particles . To observe the effects of this class of compound on the morphology of nascent viral particles , PBMCs were infected with HIV NL4-3 either in the absence or presence of PF-3450074 at 7 µM ( ∼10× EC50 ) . The cells and resulting viral products were fixed and treated for transmission electron microscopy ( TEM ) . In untreated infections , approximately 66% of particles ( 78/118 observed ) were native-like , as defined by being ∼100nm in diameter and having a distinct central density representative of a mature capsid core ) . A close in view of a native-like particle from an untreated culture ( Fig . 4a ) clearly illustrates the conical capsid core in the center of the virion . A wider view shows the high level of uniformity among the particles produced in the untreated infections ( Fig . 4b ) . Both native-like and apparent immature particles are present at consistent shape and size . In contrast , native-like particles were not observed in PF-3450074-treated infections . The particles produced in PF-3450074-treated infections lack a clear central density representative of a mature capsid ( Fig . 4c ) . Consistent with the observation that these compounds do not affect p24 levels , the number of particles in PF-3450074-treated infections appears similar to untreated , however the morphology and size of particles is highly variable ( Fig . 4d ) . The simplest explanation to reconcile the data above involves a model where this new class of inhibitors targets both early ( viral core uncoating ) and late ( viral core assembly ) events in the replication cycle by affecting CA – CA interactions and thus core stability . To study the effects on capsid assembly , we evaluated the inhibitors in in vitro CA multimerization assays [10] . Such assays can be used to measure the effect of compounds on the rate of formation of higher order CA multimers or tubes that are widely thought to represent many aspects of native core structure . Addition of PF-3450074 results in a significant increase in the rate of CA multimerization ( Fig . 3d ) . In contrast , a structural analogue with no antiviral activity , PF-4159193 ( Fig . S2 ) , did not affect the kinetics of CA multimerization , indicating that this profound effect is correlated with antiviral activity and not a general physical effect of this series . As we have reproduced with the CAI peptide ( Fig . 3d ) , all of the previously reported HIV CA assembly inhibitors decreased the rate of multimerization in this assay [10] , [11] , [12] . These data demonstrate indirectly that PF-3450074 interacts with HIV-1 CA and further suggest a mechanism that is fundamentally distinct from previously reported HIV CA inhibitors . To further understand the mode of action of this novel class of compounds , we determined the crystal structure of HIV-1 CA NTD protein in complex with PF-3450074 using a CA protein construct that contained a single glycine residue in place of the cyclophilin binding loop ( residues 87–99 ) . Although the cyclophilin binding loop is important for viral infection , it is not required for proper HIV CA protein folding and in vitro multimerization function [8] . Binding affinity of PF-3450074 to the crystallographic construct ( Kd = 3 . 42 µM ) is similar to that observed for both full length wild type CA ( Kd = 2 . 79 µM ) and isolated wild type NTD ( Kd = 2 . 24 µM ) , as measured by isothermal titration calorimetry . Using this construct , the co-crystal structure of PF-3450074 was solved with the NTD of HIV-1 CA to 1 . 8 Å resolution . The structure of the complex showed that the overall fold of the CA protein is the same as previously described CA structures [19] , [20] , [21] and illustrates that neither the compound , PF-3450074 , nor the loop deletion caused any significant shifts in the protein structure ( Fig . 5a ) . PF-3450074 occupies a preformed pocket in the HIV-1 CA NTD bounded by helices 3 , 4 , 5 and 7 ( Fig . 5a ) . The R1 and R2 aromatic moieties of the compound occupy two hydrophobic sub-pockets and provide most of the key interactions which anchor the compound to the NTD ( Fig . 5b ) . The indole substituent protrudes from the NTD R3 sub-pocket close to Lys70 . The binding site for PF-3450074 is distinct from the sites targeted by CAP-1 and CAI/NYAD-1 [10] , [11] , [22] , [23] , [24] . These results confirm that PF-3450074 directly binds HIV CA and are consistent with this class of inhibitors acting via a unique mechanism . Coordinates are stored in the Protein Data Bank ( PDB ID code 2XDE ) .
We describe a novel class of inhibitors that target HIV CA by a unique mechanism that interferes with both early and late events in the viral replication cycle . HIV CA plays an essential role in several stages of viral replication and is viewed as an important , yet unexploited target for therapeutic intervention [4] , [5] , [6] . This new series demonstrates that small molecules targeting HIV CA can have potent broad-spectrum antiviral activity . We demonstrate directly that HIV CA is the antiviral target of these inhibitors in infected cells by showing that mutations in HIV CA confer resistance to several members of the series . EM analysis shows that the series profoundly affects the morphology of nascent HIV particles . We demonstrate that the compounds affect CA protein multimerization in vitro and we have elucidated details of the novel compound binding site on HIV CA by solving a co-crystal structure of a compound from the series bound to the NTD of the protein . Our data strongly suggest that this new series of inhibitors targets HIV CA function during both the virion uncoating and viral core assembly processes . Previous studies have described two molecules that target HIV-1 CA assembly in vitro , CAP-1 ( a small molecule ) and CAI ( a dodecapeptide ) [10] , [11] , [12] . CAP-1 acts in the late stage of viral replication and does not inhibit HIV-1 infection when added to pre-formed HIV-1 particles . A cell-permeable derivative of CAI , NYAD-1 , inhibits formation of both immature and mature HIV-1 virus particles as well as early events in the replication cycle at low micromolar concentrations . The properties of the new class of CA inhibitors described in this study are clearly distinguished from those of other HIV-1 inhibitors , including previously described CA inhibitors . Unlike CAP-1 , the small molecules described here inhibit both early and late events in the HIV replication cycle . In addition , PF-3450074 did not inhibit Gag particle production from HIV-1 transfected cells , suggesting that the compound series does not affect immature particle assembly . This is in contrast to the effects on immature particle assembly reported for NYAD-1 . A co-crystal structure of a representative compound ( PF-3450074 ) demonstrated a new binding site on HIV-1 CA distinct from those described for CAP-1 or CAI . Furthermore , PF-3450074 increased the rate of HIV-1 CA multimerization in vitro , while CAI and CAP-1 decreased the rate of CA multimerization in the same assay . While this does not necessarily define the action of these compounds on replicating virus , it does suggest a fundamentally different mechanism of inhibition from that of previously described CA inhibitors . The proposed mechanism of action for both the early and late stage activities of this new class of inhibitors involves a direct effect on higher-order structures of HIV CA , in assembly and uncoating . Although the present data do not indicate whether the compounds enhance or inhibit the uncoating process , either effect is likely to interfere with proper reverse transcription [25] . HIV-1 capsid mutations proximal to the PF-3450074 binding pocket have been described that either destabilize or enhance the stability of viral cores and result in specific postentry defects in virus replication [24] . It is possible that such mutations and the compounds described in this study have analogous effects on inter-subunit capsid interactions . To gain further insights into the mode of action of PF-3450074 , we generated a model of an assembled capsid hexamer in complex with PF-3450074 ( Fig . 6a ) based on superpositioning of published assembled capsid structures [13] , [14] with the structure of the PF-3450074/CA complex . In the model , the R3 indole group which protrudes from the NTD in our structure localizes to the interface between capsid monomers in an assembled capsid and sits directly between the NTD of one capsid monomer and the C-terminal domain of another , making contacts to Tyr-169 , Leu-172 , Arg-173 , Gln-179 , and Lys-182 ( Fig . 6b ) . This suggests the R3 indole group of PF-3450074 could play a critical role in modulating inter-subunit interactions . Both the CA NTD contact residues described by the co-crystal structure and these putative C-terminal contacts are well conserved across viral strains ( Tables S5 and S6 ) . This is consistent with the broad-spectrum antiviral activity observed for this series . Although the sum of our results suggests a mechanism that affects interactions between capsid monomers , the early stage activity is consistent with other models . Cyclophilin A plays a critical role in the early stages of HIV-1 replication through interactions with the viral capsid [26] . Also , capsid-binding restriction factors such as the tripartite motif containing ( TRIM ) proteins prevent the infection of many primate cells with HIV or SIVs from other species [5] , [9] . Thus , based on our data , we cannot dismiss the possibility that , during early infection , this new series might affect specific capsid-host protein interactions that mediate the viral uncoating process . A detailed study of the early stage activity of PF-3450074 has demonstrated direct destabilization of the HIV-1 capsid and a dependence on cyclophilin A , indicating that the compound induces premature uncoating of the virus , potentially through a mechanism similar to that of TRIM restriction [27] . In this study , we identify a new binding site on HIV-1 CA that can be targeted by small molecule inhibitors resulting in broad-spectrum antiviral activity . In addition , we describe the discovery and characterization of a novel series of compounds that act at this site and inhibit the virus at two points in the replication cycle . This series should serve as a good starting point for the development of a new class of HIV therapeutics through structure-based drug design or other approaches . The broad spectrum activity of this series is particularly exciting and highlights this novel mechanism as a significant therapeutic opportunity .
HeLa CD4 LTR/beta-Gal , MT-2 , PM1 , CEM-SS and HEK 293 cells as well as pNL4-3 HIV-1 infectious clone , HIV-1 IIIB , HIV-1 RF , HIV-1 BaL , and all primary isolates were obtained through the National Institutes of Health ( NIH ) AIDS Research and Reference Reagent Program , Bethesda , MD . JC53BL cells were sourced from Tranzyme . Efavirenz ( EFV ) was kindly provided by DuPont Merck ( Wilmington , DE ) . Nelfinavir ( NFV ) , 135137 , PF-3450074 , and PF-3759857 were synthesized by Pfizer Inc . As described [18] , host cells were infected with HIV-1 NL4-3 , HIV-1 IIIB , HIV-1 RF . The cytopathic effect was measured using XTT reagent and the therapeutic index ( TI ) calculated by dividing the CC50 ( mock infected cells ) value by the EC50 . PHA-stimulated PBMC's were incubated for 1 hour with virus at an moi of 0 . 001–0 . 01 . They were plated in 96-well plates at 5×105 cells/ml and incubated for 5 days at 37°C . 10 µL of the supernatant was then transferred to a plate containing 40 µL of JC53BL cells at 0 . 5×105 cells/ml . After 2 days , the cells were processed for β-galactosidase activity using the FluorAce kit ( Bio-Rad ) . PF-3794231 was tested against clinical isolates at the Southern Research Institute ( Frederick , MD ) as previously described [18] . The single-cycle infectious HIV-1 reporter viruses were generated as previously described [28] by co-transfecting HEK293 cells with an HIV-1 NL4-3 single-cycle infectious cDNA ( pNL4-3deltaEnv ) and an NL4-3 or VSV-G envelope expression vector . Half-log dilutions of test compounds were added to HeLa CD4 LTR/beta-Gal target cells , seeded in 96-well plates at a cell density of 1×104 cells per well in DMEM containing 10% FBS . Compound-treated or compound-free target cells were then infected with the HIV-1 single-cycle infectious virus and after 72 hours measured for the induction of beta-galactosidase Data were expressed as the percent of reporter gene activity in infected compound-treated cells relative to that of infected , compound-free cells . Envelope-deleted NL4-3 cDNA ( pNL4-3deltaEnv ) was co-transfected into HEK 293 cells with an HIV envelope expression vector as previously described [29] . Half-log dilutions of test compounds were then added to transfected cell cultures 3 hrs after transfection . The supernatants were harvested 72 hrs after transfection and infectious virus production was subsequently quantified by measuring the induction of the beta-galactosidase reporter gene after 100-fold dilution into fresh medium and incubation in the presence of HeLa CD4 LTR/beta-Gal target cells for 72 hours . Viral variants resistant to PF-1385801 or PF-3759857 were selected as described previously [18] . To construct NL4-3 recombinant virus containing the Q67H , K70R , Q87P , T107N and L111I amino acid substitutions identified in the serial passage studies , viral cDNAs that had been TOPO cloned for sequence analysis were digested with BssHII and ApaI and ligated back into a wild type pNL4-3 background . The effect of amino acid substitutions in HIV-1 Gag sequences on PF-3759857 susceptibility was measured in HIV Replication assays as described previously [15] . Proteins were expressed in E . Coli BL21 ( DE3 ) in 2YT containing kanamycin . Cells were harvested following induction with IPTG and growth overnight at 20°C . CA1-231 was purified as described [30] . For CA1-146/Δ87-99G , cells were resuspended and lysed in buffer A ( 50 mM TrisHCl pH 7 . 4 , 150 mM NaCl , 10 mM imidazole , and 2 mM β-mercaptoethanol ) . The lysate was clarified by centrifugation , filtered then applied to a His trap nickel affinity column ( Sigma ) , eluting with buffer A supplemented with 300 mM imidazole . The eluted material was further purified on a Sephacryl 100HR column in 50 mM TrisHCl pH 7 . 4 , 150 mM NaCl and 2 mM β-mercaptoethanol . Multimerisation assays were performed as previously described [10] . Compound was added to 30 µg of full-length CA protein in 50 mM sodium phosphate buffer , pH 8 . 0 in a volume of 30 µl . Capsid assembly was initiated by addition of a concentrated NaCl solution ( 50 µl 5 M NaCl in 50 mM sodium phosphate , pH 8 . 0 ) . Optical density was monitored on a Molecular Devices SpectraMax spectrophotometer at 350 nm every 20 s for 1 h . PBMCs were incubated with 7 µM PF-3450074 ( approx . 10× IC50 ) , for 30 minutes before being infected with mock virus ( medium alone ) , or with NL4-3 at an MOI of 1 . At 72 hours post-infection the supernatants were removed and the cells were fixed for 60 min in MacDowell's fixative ( 4% ( v/v ) paraformaldehyde , 1% glutaraldehyde in 0 . 1M phosphate Milliong buffer , pH 7 . 3 ) at 4 C . The cells were rinsed in 0 . 15M Phosphate Sörensen buffer with 0 . 2% ( v/v ) NaCl ( pH 7 . 3 ) and suspended in warm 3 . 5% agar in water . The agar blocks were cooled to 4°C and washed three times for 10 minutes in 0 . 15M Phosphate Sörensen buffer with 0 . 2% ( v/v ) NaCl ( pH 7 . 3 ) . The agar blocks with the cells were stained using 2% Osmium tetroxide in 0 . 3M phosphate Sörensen buffer ( pH 7 . 4 ) for 1 h at 4°C . The blocks were washed 4 times for 10 minutes in sterilized water at room temperature before they were dehydrated in graded ethanol ( 50% , 70% , 90% , 100% , 100% ) for 15 minutes at each concentration at room temperature . The blocks were treated with propylene oxide three times for 15 minutes in room temperature before the resin was infiltrated for 1 hour at room temperature with a 1∶1 propylene oxide∶Epon mix ( Epon contains epoxy embedding medium 20 mL ( Epon 812 resin ) ) and 12 ml MNA ( methyl nadic anhydride ) and 9 ml DDSA ( dodecenyl succinic anhydride ) ) before the resin was treated in epon overnight at room temperature . The resin was then embedded with Epon and BDMA 3% for 1 hour at room temperature . The resin was polymerised for 48h at 58–60°C . Approximately 1 micron thick semi-thin sections were cut and stained in toluidine blue and observed in light microscopy . Subsequently , ultra-thin sections of 70–90 nm were cut on an Ultracut E Reichert microtone , and stained with uranyl acetate and lead citrate . Observations were made using a Jeol 1200 EX II electron microscope . Isothermal titration calorimetry ( ITC ) was performed using the VP-ITC calorimetric system ( GE Healthcare ) . Protein solutions for ITC were dialyzed against buffer A ( 50 mM TrisHCl pH 7 . 5/150 mM NaCl/2 mM β-mercaptoethanol ) . The dialyzed protein solution ( 15 µM ) in the calorimetric cell ( 1 . 4274 ml ) was titrated at 25°C with ligand ( 200 µM ) in the buffer A using 1×2 µL , followed by 25×10 µL injections . Heat evolved was obtained from the integral of the calorimetric signal and heat of dilution was negligible in titrations of the ligand into buffer only . Analysis was carried out with Origin 5 . 0 software ( GE Healthcare ) . Binding parameters such as the number of binding sites ( n ) , the binding constant ( Ka , M−1 ) , and the binding enthalpy ( ΔHa , kcal/mol of bound ligand ) were determined by fitting the experimental binding isotherms . Protein was concentrated to 30 mg/ml and inhibitor added to 5 mM from a 100 mM DMSO stock solution . After standing for 2 hours , hanging drop crystallizations were set up consisting of 2 µl of protein and 1 µl of well solution ( 20% PEG 8000 , 100 mM phosphate-citrate pH 4 . 2 and 200 mM sodium chloride ) . Crystals grew overnight at room temperature and were frozen for X-ray data collection following addition of 60% NDSB containing 40% ethylene glycol ( 4 µl ) . Data was collected at the ESRF ( ID14-4 ) on an ADSC Q315 detector , integrated with mosflm [31] and scaled with CCP4 package SCALA [32] . The structure was solved by molecular replacement using MOLREP [32] with a truncated model of the N-terminal CA . The structure was refitted using QUANTA version 2000 . 1 ( Accelrys Inc . , San Diego , CA ) and refinement was carried out using REFMAC [33] . Data collection and refinement statistics are shown in Table S7 . Coordinates are stored in the Protein Data Bank ( PDB ID code 2XDE ) .
|
Although the current standard of care for Human Immunodeficiency Virus ( HIV ) is high , viral resistance has emerged to every drug currently in the clinic , in some cases rendering the entire class ineffective for patients . A new class of antiretroviral drugs would be effective against strains of HIV-1 that are resistant to any existing drug and would expand the therapeutic options available to patients . Capsid is the primary structural protein of HIV and a critical part of the viral replication cycle , both in the assembly of viral particles and in the infection of host cells . We report a new class of antiretrovirals that targets HIV-1 capsid and demonstrate that it is active at two critical stages in the viral replication cycle . These compounds were consistently effective against a range of clinical strains of HIV-1 , from various sub-types , as well as HIV-2 . Finally , the compounds bind in a unique pocket on capsid that has not previously been highlighted as a drug binding site . We believe this new class of antiretrovirals can serve as a starting point for the development of a new generation of HIV-1 therapeutics and , more generally , underscores the potential of capsid as a therapeutic target .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] |
[
"virology/virion",
"structure,",
"assembly,",
"and",
"egress",
"virology/immunodeficiency",
"viruses",
"virology/viral",
"replication",
"and",
"gene",
"regulation",
"virology/antivirals,",
"including",
"modes",
"of",
"action",
"and",
"resistance"
] |
2010
|
HIV Capsid is a Tractable Target for Small Molecule Therapeutic Intervention
|
Screening of herbal remedies for Cl− channel inhibition identified Krisanaklan , a herbal extract used in Thailand for treatment of diarrhea , as an effective antidiarrheal in mouse models of secretory diarrheas with inhibition activity against three Cl− channel targets . Krisanaklan fully inhibited cholera toxin-induced intestinal fluid secretion in a closed-loop mouse model with ∼50% inhibition at a 1∶50 dilution of the extract . Orally administered Krisanaklan ( 5 µL/g ) prevented rotavirus-induced diarrhea in neonatal mice . Short-circuit current measurements showed full inhibition of cAMP and Ca2+ agonist-induced Cl− conductance in human colonic epithelial T84 cells , with ∼50% inhibition at a 1∶5 , 000 dilution of the extract . Krisanaklan also strongly inhibited intestinal smooth muscle contraction in an ex vivo preparation . Together with measurements using specific inhibitors , we conclude that the antidiarrheal actions of Krisanaklan include inhibition of luminal CFTR and Ca2+-activated Cl− channels in enterocytes . HPLC fractionation indicated that the three Cl− inhibition actions of Krisanaklan are produced by different components in the herbal extract . Testing of individual herbs comprising Krisanaklan indicated that agarwood and clove extracts as primarily responsible for Cl− channel inhibition . The low cost , broad antidiarrheal efficacy , and defined cellular mechanisms of Krisanaklan suggests its potential application for antisecretory therapy of cholera and other enterotoxin-mediated secretory diarrheas in developing countries .
Secretory diarrhea is a major health challenge in developing countries , representing the second leading cause of mortality globally in children under age 5 [1] . Repeated episodes of hypovolemia from diarrhea can produce malnutrition and impaired development [2] . The mainstay of diarrhea therapy is oral rehydration solution ( ORS ) , which consists of an aqueous mixture of salts and carbohydrates [3] , [4] . Though ORS has reduced mortality from diarrhea four-fold in the last 3 decades , its efficacy is limited , particularly in the young and elderly , and because of practicalities in its availability and compliance [5] . Antisecretory drug therapy for diarrhea may be efficacious when ORS is not available , as during natural disasters , and it may potentiate the efficacy of ORS . The intestinal epithelium absorbs and secretes large volumes of fluid , with net absorption under normal conditions and net secretion in secretory diarrheas . Intestinal fluid secretion involves Cl− transport from the blood into the intestinal lumen through Cl− channels on the enterocyte apical plasma membrane , which include the cAMP-gated channel CFTR ( cystic fibrosis transmembrane conductance regulator ) and one or more CaCCs ( Ca2+-activated Cl− channels ) whose molecular identity is not known [6]–[8] . CFTR is the primary route for Cl− secretion in secretory diarrheas caused by bacterial enterotoxins in cholera and Travelers' diarrhea ( caused by enterotoxigenic E . coli ) . CaCCs are likely involved as well in these diarrheas because of cross-talk between cyclic nucleotide and Ca2+ signaling [9] , [10] , and may provide the primary route for Cl− secretion in some viral and drug-induced diarrheas , including childhood rotaviral diarrhea [11] , [12] and antiretroviral drug-induced diarrhea [13] . The Ca2+-activated Cl− channel TMEM16A is expressed intestinal pacemaker cells , the interstitial cells of Cajal , where it is required intestinal smooth muscle contraction and motility [14] , [15] . TMEM16A is widely expressed in secretory epithelia in the airways and salivary gland , but probably plays at most a minor role as a CaCC in intestinal epithelium [16] . There is currently no approved antisecretory drug for treatment of major secretory diarrheas such as cholera . Our laboratory has identified , by high-throughput screening , several classes of small-molecule CFTR and CaCC inhibitors ( reviewed in ref . [17] ) , and has shown their efficacy in mouse models of secretory diarrheas [18] , [19] . As an alternative approach to the costly and lengthy development of a new chemical entity , here we investigated the possibility that effective , natural-product antisecretory therapeutics may already be available , but unappreciated . Screening of diarrhea remedies from around the world for enterocyte Cl− channel inhibition identified Krisanaklan , a herbal extract used widely in Thailand for treatment of diarrhea , as effective in inhibiting intestinal Cl− secretion and motility . We previously reported that one component of Krisanaklan , eugenol , inhibited the CaCC TMEM16A [20] . Here , we report here on the antidiarrheal efficacy and cellular mechanisms of Krisanaklan , and suggest its potential utility for antisecretory therapy of major , life-threatening diarrheas in developing countries .
This study was approved by the UCSF Institutional Animal Care and Use Committee ( IACUC approved protocol AN089748 ) , and was performed in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . FRT cells stably expressing human CFTR or TMEM16A were generated and cultured as described [16] , [21] . T84 cells ( ATCC CCL-248 ) were cultured as described [22] . The Thai herbal formulation Krisanaklan was purchased from Osotspa Inc . ( Bangkok , Thailand ) . Snapwell inserts containing T84 or FRT cells were mounted in Ussing chambers ( Physiologic Instruments , San Diego , CA ) , as described [16] , [23] . Activators and inhibitors were added to the apical solution and an equal volume of vehicle was added at the same time to the basolateral solution . Symmetrical HCO3−-buffered solutions were used for T84 cells . For FRT cells , the hemichambers were filled with a half-Cl− solution ( apical ) and the HCO3−-buffered solution ( basolateral ) , and the basolateral membrane was permeabilized with 250 µg/mL amphotericin B . Under these conditions short-circuit current is a direct measure of apical membrane Cl− conductance . Cells were bathed for a 10 min stabilization period and aerated with 95% O2/5% CO2 at 37°C . Short-circuit current was measured using an EVC4000 Multi-Channel V/I Clamp ( World Precision Instruments , Sarasota , FL ) . T84 cells were grown on 12-mm diameter collagen-coated transwell inserts ( 0 . 4-µm pore size Costar , Corning , Tewksbury , MA ) . Cells were cultured for 5–7 days to form tight monolayers with transepithelial resistance 900–1 , 000 Ω cm2 . Krisanaklan ( 1 . 5 ml of 6% solution ) in Ringers bicarbonate buffer was added into the basolateral chamber , and 0 . 5 ml of Ringers bicarbonate alone was added into the apical chamber . Apical chamber fluid ( 200 µL ) was collected at 0 , 30 and 60 min ( and replaced with the identical volume of buffer ) . The fluid samples were bioassayed for Cl− transport inhibition by short-circuit current measurement on T84 cells as described above . The percentage transport of inhibitory substance ( s ) was computed from activities of apical samples versus the original basolateral fluid , correcting for dilution . Mice ( CD1 strain , 25–35 g ) were deprived of food for 24 h and anaesthetized with intraperitoneal 2 , 2 , 2-tribromoethanol ( Avertin , Sigma-Aldrich , St . Louise , MO ) ( 125 mg/kg ) . Body temperature was maintained at 36–38°C using a heating pad . Following a small abdominal incision , three closed mid-jejunum loops ( length 20–30 mm ) were isolated by sutures , as described [18] . Loops were injected with 100 µl of PBS or PBS containing cholera toxin ( 1 µg ) without or with Krisanaklan . The abdominal incision was closed with suture and mice were allowed to recover from anesthesia . At 4 h the mice were anaesthetized , intestinal loops were removed , and loop length and weight were measured to quantify net fluid secretion . Fluid absorption was measured separately , from the reduction in loop weight/length ratio at 30 min after injection of 200 µL PBS . PBS containing 10 mM glucose was used as a positive control for fluid absorption . Mice were killed by an overdose of Avertin . Mice ( CD1 strain , weight 25–35 g ) were deprived of food for 24 h before experiments . Krisanaklan ( 3% in 100 µL PBS ) was administered either orally or by intraperitoneal injection . Fifteen min later mice were orally administered a charcoal meal ( 0 . 2 ml of 10% activated charcoal suspended in 5% gum acacia ) with or without 3% Krisanaklan . Thirty minutes later the mice were sacrificed and the small intestine was isolated . The peristaltic index was calculated as the percentage of distance traveled of the charcoal meal relative to the total length of small intestine . Neonatal C57bl/6 mice ( age 5–7 days , weight 1 . 8–2 . 5 g ) were inoculated with 30 µL ( 1 . 2×107 pfu/mL ) of Simian SA-11 rotavirus ( ATCC VR 1739 ) by oral gavage , as modified from prior reported models [10] , [24] . The treated group received 10 µL Krisanaklan one day after rotavirus infection . Stool specimens were collected by gentle palpation of the mouse abdomen 2 day after rotavirus inoculation . For quantification of stool water content we fabricated a polydimethylsiloxane slab of 1 . 5-mm thickness with a 1 . 91-mm diameter hole to contain a cylindrical 4 . 3-mm3 volume of stool , as described [24] . The stool plug was expelled onto absorbent tissue in a humidified atmosphere and allowed to contact the tissue for 1 min . The wetted area was measured and related to absolute water content using stool standards . In some studies the mid-jejunum was perfusion-fixed at 2 days after rotavirus inoculation for preparation of 5-µm thick , hematoxylin and eosin-stained , paraffin-embedded sections . For measurement of cytosolic Ca2+ , FRT-TMEM16A cells were plated in 96-well black-walled microplates . After removal of growth medium 100 µl of 10 µM Fluo-4 NW ( Invitrogen , Carlsbad , CA ) was added and incubated at 37°C for 30 min , then at room temperature for an additional 30 min . Fluo-4 fluorescence was measured with a plate reader at excitation/emission wavelengths of 485/538 nm . cAMP was assayed in T84 cells treated for 30 min with 0 or 10 µM forskolin , without or with Krisanaklan , lysed by repeating freeze/thaw , centrifuged , and the supernatant was assayed ( Parameter cAMP immunoassay kit; R&D Systems , Minneapolis , MN ) . Fractionation was performed on an AKTA Explorer 10 system ( GE Healthcare Life Science , Piscataway , NJ ) equipped with a C18 reversed-phase column ( Varian Pursuit XRs , 250×10 mm , 5 mm particle size , Waldbronn , Germany ) , as described [20] . In separate studies Krisanaklan was dialyzed using 1- , 10- , and 50- kDa cut-off membranes ( Float-A-Lyzer G2 , Spectrum Laboratories , Rancho Dominguez , CA ) . Wild-type CD1 mice ( age 7–10 weeks ) were killed by avertin overdose ( 200 mg/kg ) . The ileum was isolated and washed with ( in mM ) : 120 NaCl , 5 KCl , 1 MgCl2 , 1 CaCl2 , 10 D-glucose , 5 HEPES , and 25 NaHCO3 ( pH 7 . 4 ) . The ends of the ileal segments were tied and connected to a force transducer , as described [25] . Ileal segments were stabilized for 60 min with a resting force of ∼1 mN , with changes of the bathing solution every 20 min . Whole-cell recordings were made at room temperature on T84 cells , and CFTR- and TMEM16A-expressing FRT cells . The bath solution contained ( mM ) : 140 N-methyl-D-glucamine-Cl , 1 CaCl2 , 1 MgCl2 , 10 glucose and 10 HEPES ( pH 7 . 4 ) for the TMEM16A and CFTR . The pipette solution contained ( in mM ) : 130 CsCl , 0 . 5 EGTA , 1 MgCl2 , 1 Tris-ATP and 10 HEPES ( pH 7 . 2 ) . TMEM16A was activated by 400 nM free Ca2+ in the pipette solution . CFTR currents were recorded by test pulse from −80 to +80 mV from a holding potential of 0 mV in the presence of forskolin . Cl− currents in FRT-TMEM16A cells were elicited by applying voltage pulses from a holding potential of 0 mV to potentials between −100 mV and +100 mV with increases of 20 mV . CaCC was activated by 1000 nM free Ca2+ in T84 cells . To record CaCC in T84 cells , external solution contained ( mM ) : 150 NaCl , 6 CsCl , 2 CaCl2 , 1 MgCl2 , 10 glucose and 10 HEPES ( pH 7 . 4 ) were used . The pipette solution contained ( in mM ) : 40 CsCl , 100 Cs-aspartate , 5 EGTA , 1 MgCl2 , 4 . 33 CaCl2 , 4 Na2-ATP and 10 HEPES ( pH 7 . 2 ) . The currents in T84 cells were evoked by test pulse from −100 mV to 100 mV with increases of 20 mV from a holding potential of −50 mV . Pipettes ( 3–4 MΩ ) were fabricated on a model P-97 electrode puller ( Sutter Instrument , Novato , CA ) and polished with a MF-900 Micro Forge ( Narishige Scientific Instruments Laboratories ) . Whole-cell currents were recorded using an Axopatch-200B ( Axon Instruments ) and currents were filtered at 1–2 kHz and digitized at 2–4 kHz . Statistical analysis was done with Prism 5 software ( GraphPad Software Inc . , San Diego , CA ) using 2-tailed Student's t test , Mann-Whitney rank-sum test , or one-way analysis of variance ( ANOVA ) , where appropriate . Data are presented as the mean ± S . E . M . A P value of 0 . 05 or less was considered significant .
The Thai herbal medicine Krisanaklan ( Fig . 1A ) was identified from testing of diarrheal remedies for inhibition of intestinal Cl− channels . Fig . 1B shows inhibition of CFTR Cl− current in a human intestinal epithelial cell line ( T84 cells ) in response to stimulation by the cAMP agonists forskolin , an adenylyl cyclase activator , and IBMX , a phosphodiesterase inhibitor . The IC50 for inhibition of CFTR Cl− current was <0 . 01% Krisanaklan ( 1∶10 , 000 dilution ) , with complete inhibition at higher concentrations . CFTR Cl− current was inhibited by the CFTR inhibitor CFTRinh-172 ( red curve in Fig . 1B ) . Krisanaklan also inhibited CaCC Cl− current in T84 cells following stimulation by ATP , with IC50 ∼0 . 02% Krisanaklan ( Fig . 1C ) . The CaCC measurement was done in the presence of a CFTRinh-172 to eliminate ATP-dependent CFTR Cl− currents that arise from cross-talk between cAMP and Ca2+ signaling . CaCC Cl− current was inhibited by the non-selective CaCC inhibitor tannic acid ( red curve in Fig . 1C ) . Krisanaklan did not inhibit cAMP or Ca2+ signaling in T84 cells . Addition of Krisanaklan up to 0 . 1% did not reduce cytoplasmic cAMP accumulation in response to forskolin ( Fig . 1D ) , nor did it reduce cytoplasmic Ca2+ elevation in response to ATP ( Fig . 1E ) . These results suggest direct action of component ( s ) of Krisanaklan on CFTR and CaCC Cl− channels . Whole-cell patch-clamp was done to further investigate Krisanaklan effects on CFTR and CaCC currents . CFTR Cl− current was measured in CFTR-expressing FRT cells following forskolin addition ( Fig . 2A ) . Approximately linear Cl− currents were seen before and after CFTR inhibition by addition of a 1∶2000 dilution of Krisanaklan . CaCC Cl− current was measured in T84 cells following activation by high pipette Ca2+ in the presence of CFTR inhibitor CFTRinh-172 ( Fig . 2B ) . Outwardly rectifying Cl− currents were seen before and after Krisanaklan addition , which were fully inhibited by the CaCC inhibitor CaCCinh-A01 . Cl− current was also measured in FRT cells expressing TMEM16A ( Fig . 2C ) . The outwardly rectifying currents elicited by high pipette Ca2+ were ∼50% inhibited by a 1∶2000 dilution of Krisanaklan , and fully inhibited by the TMEM16A inhibitor T16Ainh-A01 . To investigate whether the active Cl− inhibitory component ( s ) in Krisanaklan might act from the inside or outside of cells , we used a bioassay to measure transepithelial transport in T84 cells grown on a porous filter . Following addition of Krisanaklan to the basolateral membrane bathing solution , the apical solution was sampled at 30 and 60 min and assayed for CFTR and CaCC activity by short-circuit current in T84 cells . While the component ( s ) of Krisanaklan responsible for CFTR inhibition were cell permeable , those responsible for CaCC inhibition were not ( Fig . 2D ) . Therefore , different components of Krisanaklan are responsible for CFTR and CaCC inhibition activities , as investigated further below . The results also suggest an intracellular site of action for CFTR inhibition and an extracellular site of action for CaCC inhibition . Krisanaklan was tested for antisecretory activity in a mouse model of CFTR-dependent secretory diarrhea caused by cholera toxin and of CaCC-dependent secretory diarrhea caused by rotavirus infection . An established model of cholera toxin-induced intestinal fluid secretion was used in which fluid accumulation is measured in closed loops of mouse mid-jejenum in vivo at 4 hours after injection of cholera toxin into each loop . Fig . 3A shows marked fluid accumulation in a cholera toxin-injected loop compared to a control ( PBS-injected ) loop . Inclusion of small quantities of Krisanaklan reduced loop fluid accumulation . Fig . 3B shows a dose-dependent reduction in intestinal fluid accumulation , with IC50 of 1–2 µl Krisanaklan per loop , with near complete inhibition of loop fluid accumulation at higher concentrations . The determinants of intestinal fluid accumulation include fluid secretion and absorption . To verify that Krisanaklan did not affect intestinal fluid absorption , measurements of fluid absorption were made in closed , mid-jejunal loops at 30 min after injection of 200 µl PBS , in which ∼65% of the injected fluid was absorbed . Fig . 3C shows no significant effects of Krisanaklan on loop fluid absorption . Rotaviral diarrhea in neonates is thought to result from activation of CaCC by the rotaviral enterotoxin NSP4 , which causes elevation of cytoplasmic Ca2+ in enterocytes by mechanisms involving enteric nerves , and perhaps galanin or integrin receptors [26]–[28] . To study Krisanaklan action , neonatal mice were inoculated with live rotavirus by oral gavage , which consistently produced watery diarrhea 1–3 days later . A single dose of Krisanaklan ( or saline control ) was administered at day 1 , and stool water content was determined at day 2 . Fig . 4A ( left ) shows watery stool in rotavirus-inoculated mice , and near-normal , non-watery stool in the Krisanaklan-treated mice . Stool water content was judged both from stool appearance , and semi-quantitatively from the wetted area on absorbent paper after deposition of a defined stool volume ( Fig . 4A , right ) . The prevention of watery stool by Krisanaklan could be a result of its antisecretory action and/or inhibition of rotaviral infection of the intestine . Fig . 4B shows the most characteristic finding of rotaviral infection of the small intestine , prominent enterocyte vacuolization [29] . Similar pathological changes were seen in intestine from Krisanaklan-treated mice , suggesting that Krisanaklan did not prevent the rotavirus infection . Based on our prior study of TMEM16A inhibition by Krisanaklan [20] , we postulated that the antidiarrheal action Krisanaklan may also involve a third mechanism – inhibition of intestinal smooth muscle contraction , as TMEM16A is expressed in interstitial cells of Cajal , where it is required for intestinal smooth muscle contraction [14] . Fig . 5A shows Krisanaklan inhibition of TMEM16A Cl− current in TMEM16A-expressing FRT cells , with IC50 ∼0 . 06% Krisanaklan , and complete inhibition at higher concentrations . Krisanaklan inhibition of intestinal smooth muscle contraction was measured in ex vivo mouse ileal strips using a force transducer and a 37°C physiological bath . Fig . 5B ( top ) shows spontaneous ileal contractions with amplitude ∼1 . 5 mN . In agreement with our prior data [20] , addition of Krisanaklan to the bath produced a concentration-dependent reduction , to near zero , of contraction amplitude , without effect on contraction frequency . Krisanaklan also reduced the amplitude of intestinal contractions following application of the agonist carbachol ( Fig . 5B , bottom ) . To investigate whether Krisanaklan inhibition of intestinal smooth muscle contraction found ex vivo may be relevant to gastrointestinal motility in vivo , we used a standard assay of intestinal motility involving transit of an orally administered activated charcoal meal . While intraperitoneal Krisanaklan at a dose similar to that used in humans significantly reduced peristaltic index , oral Krisanaklan did not ( Fig . 5C ) . The difference is likely due to minimal accumulation of TMEM16A-inhibiting components in Krisanaklan in interstitial cells of Cajal in the intestinal wall following oral administration . We investigated the nature of the component ( s ) responsible for Cl− channel inhibition by Krisanaklan . Initial studies showed that the Cl− channel inhibition activities of Krisanaklan were heat-insensitive ( 100°C for 2 min , data not shown ) . Several rough size fractions of Krisanaklan were prepared by dialysis using 1- , 10- and 50-kDa cut-off membranes and tested for Cl− channel inhibition . Fig . 6A shows inhibition of CFTR by the <1 kDa fraction , but little effect of the >1 , >10 and >50 kDa size fractions , suggesting that the CFTR inhibitor molecule ( s ) have molecular size <1 kDa . Similar CaCC inhibition was seen for <1 and >1 kDa size fractions , whereas the >10 and >50 kDa showed little inhibition ( Fig . 6B ) . Strong TMEM16A inhibition was seen for the <1 kDa fraction , with less inhibition for the higher molecular size fractions ( Fig . 6C ) , suggesting that the TMEM16A inhibitor molecule ( s ) have a molecular size <1 kDa . Fig . 6D shows that the >1 kDa fraction produce little inhibition of intestinal smooth muscle contraction , whereas the original Krisanaklan showed strong inhibition . Fig . 6E shows reverse-phase HPLC fractionation of Krisanaklan , done as reported previously [20] . Testing of individual fractions reveals distinct fractions as responsible for the CFTR , CaCC and TMEM16A inhibition actions of Krisanaklan . CaCC inhibition activity was found in several fractions , suggest a heterogeneous mixture of relatively large molecules as responsible . To determine which of the four herbal constituents of Krisanaklan are responsible for its Cl− channel inhibition activities , extracts were prepared from each individual herb and tested in T84 and FRT-TMEM16A cell cultures . Concentrations were adjusted to correspond to the original Krisanaklan formulation consisting of an ethanol/water ( 54∶46 ) extract in which each 100 mL is extracted from 10 g Aquilaria crassna bark ( agarwood ) , 33 . 3 g clove flower bud , 2 g Terminalia triptera Stapf bark and 4 . 8 g camphor . CFTR inhibition activity was found in the agarwood and clove tracts , but not in the camphor and Terminalia triptera extracts ( Fig . 7A ) . CaCC inhibition activity was found in the agarwood and clove extracts , but not in the camphor and Terminalia triptera extracts ( Fig . 7B ) . TMEM16A inhibition activity was found mainly in the agarwood and clove extracts ( Fig . 7C ) .
There is an unmet need for effective drug therapy for secretory diarrheas , especially in developing countries where cholera and other enterotoxin-mediated secretory diarrheas remain a major cause of morbidity and mortality . Potential targets for antisecretory therapy include the causative bacterial or viral agent ( vaccines and antibiotics ) , elaborated endotoxins and endotoxin-enterocyte interactions , as well as enterocyte signaling effectors ( cAMP , cGMP , Ca2+ ) and membrane transporters involved in fluid secretion ( Cl− and K+ channels , NKCC1 ) and absorption ( NHE3 , SGLT1 ) [6] . Cl− channels are attractive targets for antisecretory therapy because they are the final , rate-limiting step in Cl− ( and hence Na+ and water ) secretion . Unlike vaccines and antimicrobials that target the causative microbial agent , therapies targeting host secretory mechanisms , such as enterocyte Cl− channels , are not subject to the emergence of resistance . Here , we identified a widely used Thai herbal remedy , Krisanaklan , as having broad antidiarrheal efficacy in bacterial and viral models of secretory diarrhea , which , at the cellular level , inhibits the two major enterocyte Cl− channels , CFTR and CaCC . CFTR and CaCCs are responsible for Cl− secretion across the luminal membrane of enterocytes in the intestinal epithelium . Several lines of evidence support the conclusion that CFTR is the major apical membrane Cl− pathway in secretory diarrheas caused by the bacterial enterotoxins in cholera and Traveler's diarrhea; ( i ) The small intestine and colon show robust cAMP-activated CFTR Cl− currents [30]; ( ii ) intestinal Cl− and fluid secretion are reduced in CFTR-deficient mice and humans [31]–[33]; and ( iii ) CFTR inhibitors are effective in various rodent models of cholera [18] , [19] . CaCC ( s ) are likely involved as well in diarrheas caused by bacterial endotoxins , as experimental evidence supports cross-talk in cAMP and signalling mechanisms in which cAMP elevation increases cytoplasmic Ca+2 [9] and Ca+2 elevation increases cytoplasmic cAMP [34] . CaCC ( s ) are proposed to be the primary route for Cl− secretion in diarrheas caused by rotaviral and other viral enterotoxins [24] , [35] and various anti-retroviral and chemotherapeutic agents [13] , [36]; however , definitive quantification of the involvement of CaCC ( s ) in diarrheas awaits their molecular identification . From these considerations therapeutics targeting both enterocyte CFTR and CaCC ( s ) are predicted to have the greatest and broadest efficacy in secretory diarrheas . Krisanaklan is an inexpensive , natural-product extract containing ingredients that fully inhibit the major enterocyte Cl− channels , CFTR and CaCC . There are two antisecretory agents currently under clinical evaluation , one natural product and one synthetic small molecule . Crofelemer , a mixture of proanthocyanidin oligomers extracted from the bark latex of Croton lechleri , was recently approved for HIV-associated diarrhea [37] . Crofelemer is a weak and partial inhibitor of CFTR ( IC50>100 µM ) , though it fully inhibits enterocyte CaCC , albeit with low potency ( IC50∼10 µM ) [23] . Crofelemer is thus unlikely to be beneficial in secretory diarrheas such as cholera and Traveler's diarrhea in which CFTR is the major Cl− secretory pathway and in which fluid secretion is very high . A small molecule , iOWH032 , is in clinical trials for cholera [38] . iOWH032 is a close chemical analog of the glycine hydrazide GlyH-101 [39] that targets the extracellular ( lumen-facing ) surface of CFTR . However , iOWH032 has low CFTR inhibition potency ( IC50>5 µM ) and hence rapid ( seconds or less ) dissociation from CFTR . Mathematical modeling of an orally administered drug targeting the extracellular surface of intestinal crypts predicts little antisecretory efficacy of a micromolar-affinity CFTR inhibitor under conditions of high fluid secretion because of convective washout [40] . Alternative candidates for CFTR-targeted antidiarrheal therapy include glycine hydrazide conjugates with IC50∼50 nM that resist convective washout [19] , [41] , and thiazolidinones and quinoxalinediones that act on the cytoplasmic surface of CFTR with IC50 as low as 4 nM [18] , [21] , [42] , [43] . The three distinct actions of Krisanaklan , including inhibition of CFTR and non-TMEM16A CaCC ( s ) , and TMEM16A , are mediated by different components of the herbal extract . HLPC fractionation showed each of the inhibition activities in different fractions , and testing of size fractions prepared by dialysis indicated that small molecules of <1 kDa molecular size account for the CFTR and TMEM16A inhibition activities , and more heterogeneous , larger molecules for CaCC inhibition . We previously reported that the small molecule eugenol , a major component of clove , as a small-molecule TMEM16A inhibitor that likely accounts , at least in part , for the TMEM16A inhibition activity of Krisanaklan [20] . The molecular identities of the CFTR and CaCC inhibitors in Krisanaklan were not determined in this study , though testing of individual herbs suggest that they arise from two of the four herbal constituents , agarwood and clove . Based on prior studies of Crofelemer [23] and red wines [44] , the compounds responsible for CaCC inhibition are probably relatively large , heterogeneous and polyphenolic , whose molecular identities would be very difficult to determine . Agarwood extracts have been shown to contain several classes of phytochemical components including alkaloids , saponin , tannins , anthroquinones , glycosides and triterpenoids [45] , [46] , some of which may be responsible its Cl− channel inhibition activity . Clove is the dried flower bud of Caryephyllus aromaticus L , which contains the volatile compound eugenol , as well as non-volatile tannins , flavonoids , sterols and glycosides [47] , [48] . Though eugenol and tannins lack CFTR inhibition activity [20] , [44] , flavonoids are known to bind to CFTR and may be responsible for CFTR inhibition . Our results suggest that Krisanaklan , or extracts/components from its individual herbal constituents , is a potential candidate for antisecretory therapy of life-threatening diarrheas in developing countries . The potential advantages of Krisanaklan over alternative antisecretory agents under development include broad Cl− channel specificity with proven efficacy in mouse models , a long history of use in adults and children , low cost , and immediate availability for clinical testing . However , data from in vitro and animal models should be extrapolated cautiously to human diarrheas because of differences in intestinal anatomy , fluid secretion rates and , potentially , enterocyte signaling mechanisms . We also note that , as found for vaccines , the efficacy of antisecretory therapeutics may differ in different target populations because of genetic and environment factors . Notwithstanding these caveats , the preclinical data reported here support clinical trials of Krisanaklan for antisecretory therapy of diarrheas .
|
Secretory diarrhea is a major health challenge in developing countries . Causative agents include bacteria , as in cholera , and viruses , as in childhood rotaviral diarrhea . Though oral rehydration solution ( ORS ) has reduced mortality from diarrhea four-fold in the last three decades , its efficacy is limited , particularly in the young and elderly , and because of practicalities in its availability and compliance . Antisecretory drug therapy for diarrhea may be efficacious when ORS is not available , as during natural disasters , and it may potentiate the efficacy of ORS . As an alternative approach to the costly and lengthy development of a new chemical entity , in this study we investigated the possibility that effective , natural-product antisecretory therapeutics may already be available , but unappreciated . Screening of diarrhea remedies from around the world for enterocyte chloride channel inhibition identified Krisanaklan , a herbal extract used widely in Thailand for treatment of diarrhea , as effective in inhibiting intestinal chloride secretion . We report the antidiarrheal efficacy and cellular mechanisms of Krisanaklan , providing proof-of-concept for its potential utility for antisecretory therapy of major , life-threatening diarrheas in developing countries .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"biology"
] |
2014
|
Antidiarrheal Efficacy and Cellular Mechanisms of a Thai Herbal Remedy
|
Silene latifolia is a dioecious plant with heteromorphic sex chromosomes that have originated only ∼10 million years ago and is a promising model organism to study sex chromosome evolution in plants . Previous work suggests that S . latifolia XY chromosomes have gradually stopped recombining and the Y chromosome is undergoing degeneration as in animal sex chromosomes . However , this work has been limited by the paucity of sex-linked genes available . Here , we used 35 Gb of RNA-seq data from multiple males ( XY ) and females ( XX ) of an S . latifolia inbred line to detect sex-linked SNPs and identified more than 1 , 700 sex-linked contigs ( with X-linked and Y-linked alleles ) . Analyses using known sex-linked and autosomal genes , together with simulations indicate that these newly identified sex-linked contigs are reliable . Using read numbers , we then estimated expression levels of X-linked and Y-linked alleles in males and found an overall trend of reduced expression of Y-linked alleles , consistent with a widespread ongoing degeneration of the S . latifolia Y chromosome . By comparing expression intensities of X-linked alleles in males and females , we found that X-linked allele expression increases as Y-linked allele expression decreases in males , which makes expression of sex-linked contigs similar in both sexes . This phenomenon is known as dosage compensation and has so far only been observed in evolutionary old animal sex chromosome systems . Our results suggest that dosage compensation has evolved in plants and that it can quickly evolve de novo after the origin of sex chromosomes .
In humans , where the evolution of sex chromosomes is probably best known , the XY chromosome pair was originally a recombining pair of autosomes that progressively stopped recombining , most likely because of a series of inversions on the Y chromosome [1]–[4] . This started ∼150 million years ago [5] , [6] and the non-recombining human Y chromosome subsequently suffered from degenerating processes known as Hill-Robertson effects ( inefficient selection and reduced polymorphism , see [7]–[9] ) , which explain the massive loss of Y genes ( ∼97% ) and the concomitant accumulation of DNA repeats on the non-recombining Y compared to the X chromosome and the still recombining pseudoautosomal regions ( PARs ) [2] , [3] . Even the few genes that persisted on the Y show signs of degeneration [10] , [11] . The classical view is that the massive loss of Y-linked genes has been balanced by the evolution of dosage compensation ( equal dosage of X and autosomal transcripts in both males and females [12]–[14] ) , which is achieved by the inactivation of one X chromosome in females [15] . The question whether this three-step scenario ( X–Y recombination suppression , Y degeneration , X dosage compensation ) is similar for all species with sex chromosomes , in particular those with much younger sex chromosomes , has received much attention from evolutionary biologists , and several alternative model organisms to study the evolution of sex chromosomes have emerged , some of them very recently [9] , [16]–[18] . S . latifolia ( white campion ) is one such model organism . It is a dioecious plant from the Caryophyllaceae family with heteromorphic sex chromosomes that have originated only ∼10 million years ago [19]–[22] and is a promising model organism to study sex chromosome evolution in plants [23] , [24] . Previous work suggests that S . latifolia XY chromosomes have stopped recombining gradually [21] , [22] , [25] and that the Y is undergoing degeneration ( gene loss , reduced polymorphism , accumulation of repeats , maladapted proteins , reduced gene expression ) as in animal sex chromosomes [26]–[34] . Despite these highly interesting results , work on sex chromosome evolution in S . latifolia has been limited by the slow pace of sex-linked gene identification ( one to two new genes/year ) [21] , [25] , [35]–[40] . This situation is now changing rapidly , thanks to next-generation sequencing ( NGS ) approaches , which are helping reveal the strong potential of the S . latifolia model [23] , [24] , [41]–[43] . Here we report a study using such an NGS approach , RNA-seq , applied to several males and females of an S . latifolia inbred line . Using a de novo assembly strategy followed by SNP analysis , we identified >1 , 700 sex-linked contigs , increasing by almost 100-fold the number of sex-linked sequences available until recently in S . latifolia . Studying these 1 , 700 sex-linked contigs , we found that expression of alleles on the Y is significantly reduced compared to those on the X chromosome , providing evidence for large-scale ongoing degeneration of the S . latifolia Y chromosome . By comparing the expression of X-linked alleles in males and females , which differ in the number of X chromosomes , we further found evidence of equal dosage of X transcripts among sexes for sex-linked genes showing Y degeneration , a phenomenon known as dosage compensation . To our knowledge , this is the first evidence for dosage compensation in plants and reveals that dosage compensation is not an animal-specific phenomenon . Moreover , the finding of dosage compensation in evolutionary young sex chromosomes has novel implications for the evolution of sex chromosomes because it shows that 10 million years are sufficient to evolve dosage compensation de novo . By contrast , dosage compensation in animals has to date been documented only in >100-million-year-old sex chromosome systems .
We used RNA-seq—a next-generation transcriptome-sequencing approach—to identify new sex-linked genes and to study gene expression ( find more details in Text S1 ) . We obtained ∼35 Gb of sequence data from three males and three females from a ten-generation inbred population of S . latifolia using Illumina technology ( Table S1 ) . Male and female reads were pooled and assembled de novo ( see Material and Methods ) ( Figure S1 ) , and we obtained 141 , 855 contigs ( Table S2 ) . From these , we identified sex-linked contigs using a segregation analysis similarly to [42] , [43] and found 1 , 736 contigs with at least one sex-linked SNP ( Table S2 ) . We tested the reliability of our inference of sex-linkage by first using known autosomal genes [44] to see whether sex-linked SNPs have been wrongly inferred for these , but could not find any for the ten autosomal genes tested ( Table S3 ) . This very low rate of false positives was confirmed when running our scripts to detect sex-linked SNPs on a set of simulated autosomal SNPs ( Text S2 ) . We thus concluded that our inferences of sex-linkage are highly reliable . To estimate how many sex-linked contigs we missed with our method , we checked how many of the previously identified sex-linked genes were among our sex-linked contigs ( Table S3 ) . 42% of these were not found , which means that our rate of false negatives is quite high , and we identified a subset ( probably about half; see Figure S2; Text S1 ) of the sex-linked genes in S . latifolia . Many of our sex-linked contigs should be full-length transcripts as suggested by the size distribution plot ( Figure S3 ) . We used read numbers to estimate expression levels of the sex-linked contigs ( see Material and Methods ) . We first compared expression levels of X-linked and Y-linked alleles in males . The read numbers were normalized to be able to combine data from different male individuals . As shown in Figure 1 , we found that the Y/X expression ratio is significantly less than 1 ( median 0 . 77 , mean 0 . 89 , significant Wilcoxon paired test p<10−16 ) . This is in agreement with previous work on six experimentally identified sex-linked genes [33] and also with recent work using RNA-seq data [42] , [43] . Why Y expression is reduced over evolutionary time is not fully understood . It could be because of the accumulation of slightly deleterious mutations in promoters and cis-regulatory elements , and/or the insertion of transposable elements when the methylation of these elements spreads to nearby genes . However , this trend is considered a hallmark of Y chromosome degeneration and has been observed in several animal systems [45] , [46] . Y degeneration is thus clearly visible in S . latifolia but may not be as pronounced as expected because of haploid selection on pollen preventing the degeneration of many pollen-expressed Y genes [42] ( but see [43] , [47] ) . The observation that many X/Y pairs show reduced Y expression ( Figure 1 ) raises the question whether dosage compensation has evolved in S . latifolia . To test this , we compared expression levels of sex-linked genes between males and females following a normalization procedure that allows comparing different individuals ( see Material and Methods ) . First , we computed the ratio of the expression intensities of X-linked contigs in males and females and called this the Xmale/2Xfemale expression ratio ( to stress the difference in gene copy number between male and female ) . In the absence of dosage compensation , the Xmale/2Xfemale expression ratio is expected to be 0 . 5 , simply because males ( XY ) have one X-linked copy and females ( XX ) have two . This is what we observe for contigs that do not show reduced expression of the Y-linked allele relative to the X-linked allele , i . e . , that have a Y/X expression ratio close to 1 ( median of Xmale/2Xfemale ratio is 0 . 51 for contigs with 1≤Y/X<1 . 5; see Figure 2 ) . However , for contigs with reduced Y expression and therefore low Y/X ratios , we observe an Xmale/2Xfemale expression ratio very close to 1 ( median of contigs with Y/X<0 . 5 is 0 . 93; see Figure 2 ) . This suggests that for contigs with reduced Y expression , for which expression of sex-linked genes would thus be unbalanced between males and females , a mechanism has evolved that compensates for the reduced Y expression by increasing X expression in males . To study this phenomenon further , we compared expression of X-linked and Y-linked alleles in males and females for different Y/X expression ratio categories ( Figure 3 ) . We excluded sex-linked contigs that showed either an elevated Y expression ( high Y/X ratios ) or male-biased X expression ( high Xmale/2Xfemale ratios ) . Such male-biased expression patterns suggest that these genes may be sexually antagonistic genes . The evolutionary dynamics of such genes is known to be distinct from other sex-linked genes and no dosage compensation is expected [48] , [49] . Figure 3 shows the results for the remaining 75% of sex-linked genes . We found that X expression in males increases with decreasing Y expression , which results in similar expression levels of sex-linked contigs in both sexes and provides further evidence of dosage compensation in S . latifolia . Importantly , this result is consistent even when we include only sex-linked contigs with at least two sex-linked SNPs , for which we estimated the rate and number of erroneous sex-linked contigs to be extremely low ( 0 . 001 and 1 . 38 , respectively; see Figure S4 ) . We also looked at expression patterns of the contigs corresponding to known sex-linked genes . Although this analysis can only be qualitative due to the small number of such genes , we found that Y/X ratios for most genes are consistent with previous work [33] and some known sex-linked genes show evidence for dosage compensation ( Table S4 ) .
There was a recent claim of absence of dosage compensation in S . latifolia [42] , which seems to contradict our findings . However , the test for dosage compensation performed in this recent work is very different from ours . As Chibalina and Filatov ( 2011 ) analyzed crosses ( parents and progeny ) , they were able to identify X-linked genes without detectable homologous Y-linked copies ( called hemizygous genes ) . They compared the expression levels of these hemizygous genes between sexes , found a significantly reduced expression in males compared to females , and concluded that this was evidence for the absence of dosage compensation in S . latifolia [42] . Their test however may be overly conservative , as it requires a strict Xmale/2Xfemale ratio of 1 to infer for dosage compensation . Their figure 4 suggests the Xmale/2Xfemale ratio is not 0 . 5 , as expected under a complete absence of dosage compensation , but instead is close to 0 . 7 , which is consistent with many hemizygous genes being dosage compensated . Importantly , the hemizygous genes were interpreted as sex-linked genes with fully degenerated Y copies , which may not always be the case as genes that have recently moved from the autosomes to the X chromosome will also be detected as hemizygous genes but dosage compensation is clearly not expected for those genes [43] . Such gene movement has been documented in S . latifolia [39] and may account for the intermediate Xmale/2Xfemale value ( between 0 . 5 and 1 ) found in [42] . By contrast , we looked for departure from a Xmale/2Xfemale of 0 . 5 and did not restrict the test to sex-linked genes with no Y expression but included the many sex-linked genes with reduced but still detectable Y expression . We thus performed a more permissive test for dosage compensation , which may be more suitable in the case of young sex chromosomes with incipient X chromosome dosage compensation . Dosage compensation is not the only sex-specific gene expression regulation that is expected on the X chromosome . Indeed , X-linked genes involved in sexual conflicts—for instance those underlying sexual dimorphism and having sexually antagonistic effects—can show sex-biased expression and this can substantially affect the global X expression pattern in both sexes if these genes are numerous [50] . A way to distinguish dosage compensation from such sex-specific expression regulation is to look at the X over autosome ( X/A ) expression ratio as only dosage compensation predicts a X/A expression of 1 [50] . However , this test is difficult to perform here for several reasons . First , our set of sex-linked genes is expected to exclude those with very low expression levels because the detection of sex-linked SNPs requires reasonably high read coverage . This should bias upward the average expression level of sex-linked genes compared to the “autosonal” set , which is what we actually found ( the mean number of reads per base is 466 . 7 for sex-linked contigs and 101 . 4 for non–sex-linked contigs ) . Second , we do not have a reliable “autosomal” set as this includes a mixture of autosomal contigs and sex-linked contigs not detected by our method ( ∼40% of all sex-linked genes , see above ) . Although we excluded possible candidates for sexually antagonistic genes ( some of the contigs with high Xmale/2Xfemale may be “male-beneficial and female-detrimental” genes ) , we cannot completely rule out the possibility that others remained in the set of contigs used to assess dosage compensation ( especially some contigs with low Xmale/2Xfemale may be “female-beneficial and male-detrimental” genes ) . However , Figure 3 shows that the increase of X expression in males follows the level of degeneration of Y expression , which is not expected in case of sexually antagonistic selection . Moreover , increased expression of the X-linked allele in males always compensates for the reduced Y expression , such that the total expression of these sex-linked genes is similar in both sexes ( i . e . , X+Y expression in males = X+X expression in females ) , which is not in agreement with sexually antagonistic selection . On the contrary , sexually antagonistic selection predicts between-sex differences in expression of sex-linked genes . The results presented in Figure 3 are thus better explained by dosage compensation than by sexually antagonistic selection . Global dosage compensation has previously been documented in male heterogametic systems ( XY ) such as Drosophila , Caenorhabditis elegans , and mammals [14] , [51] , whereas only partial ( or no ) dosage compensation has been found in female heterogametic systems ( ZW ) [52] . Indeed , in zebra finch , chicken , and crow , no global mechanism to balance avian Z chromosome gene dosage ( such as X chromosome inactivation ) has been found [53]–[56] and in chicken , dosage compensation seems to be local , with only few Z-linked genes being dosage compensated [57] . Similar observations have been made in silkworm [58] , [59] , indicating that the lepidopteran Z is not fully dosage compensated , and also in the parasite Schistosoma mansoni [60] . Moreover , studies on the platypus [61] , [62] and on sticklebacks [63] suggest that partial dosage compensation can also exist in male heterogametic systems ( XY ) . Overall , these new data suggest that full dosage compensation is not a necessary outcome of sex chromosome evolution [50] . An important point of whether dosage compensation will evolve or not is the presence of dosage-sensitive genes on the proto-sex chromosomes , as these genes are the only ones for which dosage compensation is vital [50] , [64] . Although we do not have any data about the fraction of dosage-sensitive genes in the different sex chromosome systems , it has been suggested that resistance to aneuploidy and polyploidization may indicate whether the genome as a whole includes many such genes or not [50] . Polyploidization is known to be common in plants [65] . However , plant polyploids do have dosage problems that cause endosperm development failure and reduced fertility [64] , [66] . Following polyploidization events , the retention of plant duplicate genes seems to be driven by dosage constraints as in animals [64] . All this suggests that the success of polyploids in plants may not be related to lack of dosage constraints but to other reasons ( e . g . , vegetative propagation ) . It is also known that aneuploidy has more severe phenotypic consequences than polyploidy in plants , which further supports the idea of strong dosage constraints in plant genomes [64] . As far as we know , there is no documented case of fertile polyploids in dioecious Silene species and it is possible that the S . latifolia genome includes enough dosage-sensitive genes for dosage compensation to evolve . Our results reveal that dosage compensation is not restricted to animals but also occurs in plants and raise questions about the mechanisms underlying dosage compensation . In animals , three different dosage compensation mechanisms have been uncovered ( reviewed in [67] ) : hyper-expression of X-linked alleles in male Drosophila , down-regulation of the two X-linked alleles in hermaphrodites of C . elegans , and inactivation of one of the two female X chromosomes in mammals . We tested whether such a chromosome-wide inactivation exists in S . latifolia by checking whether both X-linked alleles are expressed in females . Although heterozygosity is low in our X-linked alleles because our individuals are inbred , we found that the level of heterozygosity of the X-linked alleles is similar for sex-linked contigs with dosage compensation and those without dosage compensation ( Table S5 ) . This suggests that both X-linked alleles are expressed , whatever the level of dosage compensation is , and does not support an X-inactivation-like mechanism in S . latifolia . Further work will be needed to identify the molecular mechanism underlying dosage compensation in S . latifolia . Previous work in animals has reported dosage compensation in old X chromosomes ( see above ) and also in young neoX chromosomes such as the D . miranda neoX . The fusion between X and the autosome that formed the D . miranda neoX is very recent ( 1 . 5 million years old ) , but dosage compensation is achieved by a protein complex ( the MSL complex ) that pre-dates neoX formation and has been shown to be very old [68] . Evidence for de novo evolution of dosage compensation in evolutionary young animal sex chromosomes is therefore lacking [50] . In the Silene genus , most species are hermaphroditic or gynodioecious and do not have sex chromosomes . Sex chromosomes have evolved recently in two independent lineages , one including S . latifolia and one containing S . colpophylla [20] , [44] , [69] . Our results therefore reveal that dosage compensation has evolved de novo in evolutionarily young sex chromosomes in probably less than 10 million years . Furthermore , Figure 2 shows that many dosage-compensated contigs have an Xmale/2Xfemale ratio that is not exactly 1 ( although the median is close to 1 , there is no peak at 1 for Y/X<0 . 5 contigs ) . This is consistent with the mechanism being evolutionarily young and not optimized yet . Our results also reveal that dosage compensation can evolve as soon as Y expression starts declining . This way , dosage compensation already exists when the Y copy is ultimately lost ( and can even facilitate such loss , see [70] ) . Instead of being a later step of sex chromosome evolution following Y degeneration , our results suggest that the evolution of dosage compensation and Y degeneration probably occur at the same time .
Plants used in this study belong to a population of S . latifolia that has been inbred for ten generations with brother-sister mating: three males ( U10_11 , U10_49 , and U10_09 ) and three females ( U10_34 , U10_37 , and U10_39 ) that were grown in a temperature-controlled greenhouse . The QiagenRNeasy Mini Plant extraction kit was used to extract total RNA two times separately from four flower buds at developmental stages B1–B2 after removing the calyx . Samples were treated additionally with QiagenDNase . RNA quality was assessed with an Aligent Bioanalyzer ( RIN>9 ) and quantity with an Invitrogen Qubit . An intron-spanning PCR product was checked on an agarose gel to exclude the possibility of genomic DNA contamination . Then , the two extractions of the same individual were pooled . Samples were sequenced by FASTERIS SA on an Illumina HiSeq2000 following an Illumina paired-end protocol ( fragment lengths 150–250 bp , 100 bp sequenced from each end ) . Individuals were tagged and pooled for sequencing in two different runs ( U10_49 male and U10_37 female in the first run and the others in the second ) . See Table S1 for sizes of the different libraries . Our Illumina reads are available in the GEO database ( through the GEO Series GSE35563 ) . De novo assembly was conducted on a computer cluster ( Figure S1 ) . Illumina reads from all individuals were pooled together for assembly with AbySS 1 . 2 . 5 ( E = 10 , n = 5 ) [71] with the paired-end option and with all k-mers ranging from 51 to 96 in order to address variable transcript expression [72] . A k-mer length equal to 51 was the minimum possible to avoid contigs shorter than the reads , and 96 is the maximum allowed by AbySS . Only contigs were kept at this stage , singlets were discarded . Contigs that exactly matched another longer contig were then removed by pairwise comparison of AbySS outputs using Trans-ABySS 1 . 2 . 0 [72] . A non-redundant set of contigs was thus obtained and further assembled through two runs of CAP3 version 12/21/07 [73] . Singlets and contigs were conserved after each CAP3 run . CAP3 runs increased the chance for X and Y copies to be assembled into the same contig , which is crucial for further sex-linked SNP detection . Contigs shorter than 200 bp were not included in the final set of contigs . Illumina reads were mapped onto reference sequences ( final set of contigs and also CDS from known sex-linked genes retrieved from GenBank for adjusting SNP detection , see below ) for each individual separately using BWA 0 . 5 . 9 [74] ( using default parameters for paired-end reads , and gap and mismatch maximum number of 5 as suggested for 100 bp reads in [74] ) , which was shown to be efficient and to use much less RAM than other programs for Illumina read mapping [75] . Alignments of all individuals were then merged together using Samtoolsmerge version 0 . 1 . 12 [76] . The percentage of mapped reads was assessed using Samtoolsflagstat version 0 . 1 . 12 [76] and the average coverage was determined using the Genome Analysis Toolkit ( GATK 1 . 0 . 5315 ) Depth of Coverage [77] . SNPs were detected with the GATK Unified Genotyper ( using the following parameters: -stand_call_conf 4 -stand_emit_conf 0 -mbq 17 -mmq 0 -mm40 40 -bad_mates -dcov 2000 ) [77] , which is considered the best currently available tool for SNP detection [78] . Thresholds for the different SNP detection parameters were set to be very low ( except for the base quality parameter ) in order not to disfavour Y SNPs that are expected to be found in low numbers and low mapping quality if a contig contains mainly X reads , which can happen when X-linked alleles are more strongly expressed than Y-linked alleles [33] . The detected SNPs were then filtered using Perl scripts to retrieve SNPs for which all males are heterozygous ( XY ) and all females homozygous ( XX ) . All contigs with at least one SNP showing this pattern were considered sex-linked . For females , the genotypes inferred by GATK were directly used for analysis . For males , this information is not reliable since the Y-linked allele is expected to be less expressed than the X-linked allele [33] while GATK genotyper makes the assumption that both alleles are expressed at a similar level . The read numbers of each SNP were thus used to infer male genotypes ( see Text S3 for details ) . Polymorphism on the X chromosome ( at least one male or female heterozygous or all individuals homozygous but not for the same polymorphism ) was detected on sex-linked contigs with a similar filter as the one described above . Expression levels of the X-linked and Y-linked alleles in males and both X copies in females were computed by counting reads at sex-linked SNP locations only , and not for the entire contigs , in order to clearly distinguish between X and Y reads . Total read numbers of all X or Y SNPs provided by the GATK Unified Genotyper [77] were summed for each X-linked or Y-linked alleles and each individual separately and then normalized using the total number of mapped reads per individuals ( library size ) and the number of sex-linked SNPs in the contigs:With E = normalized expression level , r = sum of total read counts , n = n sex-linked SNPs , l = normalized library size . The library size of the six individuals was normalized to take into account the difference in mitochondrial , chloroplast , and transposable element ( TE ) transcript quantity between sexes and the difference in rRNA quantity between the first and the second Illumina run . The Arabidopsis thaliana rRNA genes , complete S . latifolia mtDNA genome [79] , S . latifolia chloroplast genes rpoB , rpoC1 , rpoC2 , rps2 , atpI , atpH , atpF , atpA , psbI , psbK , rps16 , matK , psbA , rpl2 , ycf2 , ndhB , rps7 , and the TEs known in Silene [80] were retrieved from GenBank . The read numbers of rRNA , TEs and mtRNA , and cpRNA were determined by mapping the Illumina reads onto the known CDS sequences of these elements using the default parameters in BWA ( results presented in Table S1 ) . The expression levels were normalized for each contig and for each individual in number of reads per kilobase per million mapped reads ( RPKM ) [81] , and then the mean for each sex was computed .
|
The mammalian sex chromosomes originated from an ancestral pair of autosomes about 150 million years ago and the Y chromosome subsequently degenerated , losing most of its genes . During this process , a phenomenon called dosage compensation evolved to compensate for the gene loss on the Y chromosome and to equalize expression of X-linked genes in the two sexes . In humans , this is achieved by inactivating one of the two X chromosomes in females . Dosage compensation has also been reported in other animal XY systems such as fruit flies and worms , each 100 million years old or more . Here we studied dosage compensation in plants . We used high-throughput RNA sequencing in male and female Silene latifolia ( white campion ) —a dioecious plant whose XY chromosomes originated only about 10 million years ago—to identify hundreds of sex-linked genes . Analysis of their expression patterns in males and females revealed equal doses of sex-linked transcripts in both sexes , regardless of the degree of reduction of Y expression due to degeneration . Our results thus show that dosage compensation occurs in plants and is thus not an animal-specific phenomenon . They also reveal that proportionate dosage compensation can evolve rapidly de novo after the origin of sex chromosomes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] |
[
"sequence",
"analysis",
"genome",
"expression",
"analysis",
"plant",
"science",
"genomics",
"genome",
"evolution",
"evolutionary",
"biology",
"plant",
"evolution",
"plant",
"biology",
"plant",
"genomics",
"biology",
"computational",
"biology",
"comparative",
"genomics",
"genetics",
"and",
"genomics"
] |
2012
|
Rapid De Novo Evolution of X Chromosome Dosage Compensation in Silene latifolia, a Plant with Young Sex Chromosomes
|
Microsatellites have been found to be useful in determining genetic diversities of various medically-important parasites which can be used as basis for an effective disease management and control program . In Asia and Africa , the identification of different geographical strains of Schistosoma japonicum , S . haematobium and S . mansoni as determined through microsatellites could pave the way for a better understanding of the transmission epidemiology of the parasite . Thus , the present study aims to apply microsatellite markers in analyzing the populations of S . japonicum from different endemic areas in the Philippines for possible strain differentiation . Experimental mice were infected using the cercariae of S . japonicum collected from infected Oncomelania hupensis quadrasi snails in seven endemic municipalities . Adult worms were harvested from infected mice after 45 days of infection and their DNA analyzed against ten previously characterized microsatellite loci . High genetic diversity was observed in areas with high endemicity . The degree of genetic differentiation of the parasite population between endemic areas varies . Geographical separation was considered as one of the factors accounting for the observed difference between populations . Two subgroups have been observed in one of the study sites , suggesting that co-infection with several genotypes of the parasite might be present in the population . Clustering analysis showed no particular spatial structuring between parasite populations from different endemic areas . This result could possibly suggest varying degrees of effects of the ongoing control programs and the existing gene flow in the populations , which might be attributed to migration and active movement of infected hosts from one endemic area to another . Based on the results of the study , it is reasonable to conclude that genetic diversity could be one possible criterion to assess the infection status in highly endemic areas . Genetic surveillance using microsatellites is therefore important to predict the ongoing gene flow and degree of genetic diversity , which indirectly reflects the success of the control program in schistosomiasis-endemic areas .
Schistosomiasis is one of the most important neglected tropical diseases affecting almost 240 million people throughout the world with more than 700 million considered to be at risk of infection [1] . Five species of schistosomes are known to cause human infection: Schistosoma haematobium , S . mansoni , S . mekongi , S . intercalatum , and S . japonicum . Among these species , S . japonicum is considered the most virulent because of the larger number of eggs it can produce as compared to other species , causing severe disease pathology . In addition , the zoonotic nature of S . japonicum contributes to increased disease transmission , making schistosomiasis control more difficult [2 , 3] . Microsatellite markers have recently been used in determining S . japonicum genetic diversity and in estimating the levels of gene flow in the population . Previous studies have recommended the use of microsatellite markers to determine schistosome genetic diversity because of their codominant expression and their ability to serve as neutral markers [4 , 5] . The easy observation of heterozygosity and the reasonable number of alleles per polymorphic locus in the samples make microsatellite analysis a more powerful tool in genetic studies than the use of rapid amplified polymorphic DNA ( RAPD ) and mitochondrial DNA [6] . Studying the genetic variation of S . japonicum populations provides an opportunity to link some genotypes associated with disease prevalence , which can then be used in formulating effective control measures . Previous studies using microsatellite markers suggested that the prevalence of S . haematobium and S . mansoni infections could be closely related with the parasite’s genetic variation . High prevalence of infection was observed in those areas with high genetic diversity and low prevalence in those areas with low genetic diversity [7 , 8] . This information makes DNA microsatellites useful in population genetics studies . In addition , microsatellite markers had been used in the identification of different geographical strains of S . japonicum in China [5] . Previous studies suggested that these markers could provide useful information for assessing the efficacy of mass drug administration ( MDA ) [9] . In the Philippines , schistosomiasis has shown considerable variations in the intensity and prevalence of the disease [10] . We hypothesized that the population genetic structure of schistosome parasites might partly contribute to these differences . The parasite population in each endemic area was therefore characterized for their genetic backgrounds using the microsatellite markers . The information found in this study can therefore provide basic information on the population genetic structure of S . japonicum in the Philippines that can be used as basis to evaluate and modify the widely used current control strategies for human schistosomiasis . For example , cognizant of the genetic types of the parasites , interventions can be modified so as to address differences in disease prevalence .
Animal experiments done in this study were conducted according to ethical guidelines for the use of animal samples permitted by the University of the Philippines Manila Institutional Animal Care and Use Committee ( 2011–009 ) , as well as by Obihiro University of Agriculture and Veterinary Medicine ( Permit No . 28–30 ) . Infected mice were anesthesized using isoflurane before they were sacrificed . Perfusion method with normal saline solution was done to collect the adult S . japonicum . The snail intermediate host , O . h . quadrasi , were collected from seven municipalities ( in seven provinces ) in 2013 to 2015 where the disease is endemic , namely Catarman ( in Northern Samar ) , Gonzaga ( in Cagayan Province ) , New Corella ( in Davao del Norte ) , Irosin ( in Sorsogon ) , Talibon ( in Bohol ) , Alang-Alang ( in Leyte ) and Socorro ( in Oriental Mindoro ) ( Fig 1 ) . The snails were crushed between glass slides to examine the presence of cercariae under the microscope . The cercariae were then pooled separately for each endemic municipality and used for mice infection . Ten BALB/c mice were infected percutaneously with 50 cercariae from each municipality . The infected mice were sacrificed six weeks after the infection , and the adult worms were collected from their mesenteric veins and washed with saline for DNA extraction . Genomic DNA was extracted from individual male and female adult worms using the DNeasy Blood & Tissue Kit ( QIAGEN , Japan ) following the manufacturers’ protocol . Table 1 showed the total number of DNA samples tested in each endemic municipality . PCR amplifications were performed using Veriti 96-well Thermal Cycler ( Applied Biosystems , Carlsbad , CA ) . Amplifications were performed in 10 μl reactions containing 1 μl of 10X PCR buffer , 0 . 4 μl of 1 . 5 mM MgCl2 , 0 . 2 μl of 2 . 5 mM dNTP , 0 . 2 μl each of 10 pmol/μl primer , 0 . 1 μl of 5 U/μl Taq DNA polymerase ( Takara , Otsu , Japan ) , and 1 μl of template DNA . The conditions for thermal cycling were as follows: 5 minutes at 94°C , followed by 30 cycles of 1 minute at 94°C , 1 minute at locus-specific temperature , 1 minute at 72°C , with a final extension at 72°C for 10 minutes [11] . The DNA of each individual S . japonicum worm was genotyped using the previously characterized microsatellite loci RRPS , M5A , TS2 , MPA , 2AAA , J5 , SJP1 , SJP5 , SJP6 , and SJP9 [11 , 12] . In our study , we have screened twenty microsatellite markers but among which only ten worked well . The 5’ end of the forward primer for each locus was fluorescently labeled with 6-FAM , VIC and Ned dyes . Different dyes were used for those loci with overlapping fragment size . Two μl of the PCR product with LIZ 600 labeled size standard ( Applied Biosystems ) was subjected to the 3500 ABI Prism Genetic Analyzer for fragment analysis assay . The allele sizes were determined using the Gene Mapper software version 4 . 0 ( Applied Biosystems ) . In each run , S . japonicum sample from Gonzaga , Cagayan , which has good DNA volume and concentration , served as the reference genotype for which the microsatellite sizes for the 10 loci had been determined by sequencing . S . japonicum Yamanashi strain ( Japanese isolate ) was also genotyped as a control group to confirm that the microsatellite markers could differentiate between samples from different origin . A total of 201 DNA samples were tested; however , only 186 were successfully genotyped due to poor DNA quality . For each population , the genetic diversity was examined by calculating the number of alleles using rarefaction analysis . Expected heterozygosity ( gene diversity ) ( He ) and observed heterozygosity ( Ho ) were determined using the GenAlEx 6 . 5 software [13] . Rarefaction analysis was performed to make the alleles comparable in the population . Genetic differentiation was determined using Wright’s F-statistics ( Fst ) in Arlequin , and the significance of the Fst values was tested at p value <0 . 05 [14] . The following qualitative guidelines were used for the interpretation of Fst genetic differentiation: 0–0 . 05 ( little ) , 0 . 05–0 . 15 ( moderate ) , 0 . 15–0 . 25 ( great ) , and >0 . 25 indicate ( very great genetic differentiation ) [5] . The Analysis of Molecular Variance ( AMOVA ) was used to partition the genetic variation within and among populations using the software Arlequin version 3 . 5 . The inbreeding coefficient ( FIS ) , which measures the extent of nonrandom mating , was computed in the study . Nonrandom mating occurs when there is inbreeding . Principal Coordinate Analysis ( PCoA ) was done to determine the clustering pattern of S . japonicum population based on their genetic distance using GenAlEx 6 . 5 . Cercariae derived from a snail infected with only a single miracidium is assumed to be genetically identical . Hence , duplicate multi locus genotypes in a population are a consequence of clonal replication within snails [12 , 15] . The presence of duplicate multilocus genotypes in adult worms was identified as one possible source of bias [15] . Duplicate multilocus genotype ( MLG ) was therefore removed , leaving a single representative of each in the dataset . Recent studies revealed that removal of clones in the dataset improved the assignment and clustering pattern of S . japonicum population [15] . The GENECLASS software 2 . 0 was used to identify migrant individuals [16] . To visualize relationships among populations , a Neighbor-joining tree was constructed based on FST genetic distance using 100 bootstrap replications in POPTREE2 [17] . The FST is one of the well-known parameters used in measuring genetic differentiation between populations using microsatellite data [18] . The S . japonicum Yamanashi strain was used as an outgroup . Sequences of microsatellite loci reported here have been deposited in GenBank with the following accession numbers , RRPS ( U22167 ) , M5A ( AF244896 ) , TS2 ( AF244896 ) , MPA ( U11895 ) , 2AAA ( M32280 ) , J5 ( M26212 ) , SJP1 ( EU262604 ) , SJP5 ( EU262608 ) , SJP6 ( EU262609 ) and SJP9 ( EU262612 ) .
A total of 186 individual S . japonicum worms collected from seven endemic municipalities were analyzed . Highest gene diversity indices ( He ) was observed in Catarman ( 0 . 727 ) followed by Irosin ( 0 . 694 ) , Socorro ( 0 . 677 ) , and Gonzaga ( 0 . 605 ) while the lowest was in Alang-Alang ( 0 . 495 ) . Those from New Corella ( 0 . 587 ) and Talibon ( 0 . 566 ) were comparable . Similarly , allelic richness after sample size correction was highest in Irosin ( 4 . 630 ) followed by Catarman ( 4 . 500 ) and Socorro ( 4 . 280 ) and lowest in Alang-Alang ( 2 . 570 ) followed by Talibon ( 2 . 920 ) ( Table 1 ) . Population-specific inbreeding coefficient was determined in this study to measure the extent of nonrandom mating . Highest inbreeding coefficient values ( FIS ) was observed in Irosin ( 0 . 239 ) while the lowest was in Catarman ( 0 . 012 ) ( Table 1 ) . The lowest inbreeding coefficient values in Catarman may be related to the increased heterozygosity in this area . Inbreeding increases the homozygosity of the alleles . The pairwise FST values ranged from 0 . 019 to 0 . 0188 , indicating varied levels of pairwise population genetic differentiation ( Table 2 ) . Great genetic differentiation was observed in the New Corella samples . The AMOVA showed that greater genetic variation in the samples occurred within the population ( 91 . 95% ) rather than among populations ( 8 . 05% ) ( Table 3 ) . The PCoA showed no particular geographical structuring among the S . japonicum populations ( Fig 2 ) . The neighbor-joining tree method showed clustering of the samples into two groups . Populations from Catarman , New Corella , Gonzaga and Talibon grouped together ( 86% NJ bootstraps ) , whereas populations from Alang-Alang Socorro and Irosin belong to a separate cluster ( 61% NJ bootstraps ) ( S1 Fig ) . However , there was no correlation between the clustering of populations and their geographic distribution as shown in the neighbor joining tree ( S1 Fig ) . These findings further support the results of the PCoA ( Fig 2 ) , suggesting the existing gene flow in the population . Two subgroups were observed in Catarman ( Northern Samar ) using the PCoA analysis ( Fig 2 ) . The presence of two subgroups in the Catarman samples may account for the high genetic variation within population ( Table 1 ) . Six samples , namely 3 from Gonzaga , 2 from Irosin and 1 from New Corella , were identified by GENECLASS as migrants ( Table 4 ) . These individuals showed a probability below 0 . 05 .
In this study , based on the hypothesis that the population genetic structure of S . japonicum might explain the variations in the intensity and prevalence of schistosomiasis in the Philippines , the genetic polymorphism of the parasite population from different endemic areas was examined . A large number of different alleles were observed in the samples examined , especially in Irosin , Catarman and Socorro where high prevalence of infections was reported [10] ( S1 Table ) . There is a greater potential for these populations to possess the alleles responsible for the parasite infectivity , causing high infection [7 , 19] . These findings were in agreement with that of previous studies where the prevalence of infection was directly proportional to the number of alleles [7 , 8 , 22] . This situation somehow follows a general pattern in our current study where high prevalence of infection either in humans and snail hosts was observed in those areas with high allelic richness while low or zero prevalence in those areas with low allele numbers ( S1 Table ) . However , this is in contrast to our results in Alang-Alang where low number of different alleles has been observed where high prevalence of the infection was also reported . This could be due to the prolonged utilization of praziquantel from the annual MDA since Leyte has been one of the oldest endemic foci in the Philippines . To confirm such findings the effect of drug selective pressure brought by praziquantel on the parasite genetic diversity should be analysed . It has been known in other parasitic infections such as malaria that selective drug pressure brought by extensive drug use can lead to a reduction in genetic diversity of the parasite [20 , 21] . Currently , there are no microsatellite markers that can be linked with parasite infectivity . The alleles that might contribute to the high infection rate might be present in those area with high prevalence , however further studies should be done to confirm this . In this study , it should be noted that the prevalence data presented in S1 Table was collected from 2013 to 2015 while , our samples obtained from Talibon ( Bohol ) were collected prior to this period . Bohol is considered as a near-elimination area based on the absence of human cases for many years now . However , the presence of infection in water buffaloes continues to indicate an ongoing transmission even if there are no more human cases ( S1 Table ) . Hence , the possibility of human infection is always present . Among the seven endemic municipalities analyzed , the Catarman samples showed the highest gene diversity indices ( Table 1 ) . Catarman has been reported with high prevalence of infection both in humans and snail intermediate hosts ( S1 Table ) . In addition , this study also revealed that water buffaloes and dogs in this municipality had high prevalence of infection ( S1 Table ) . High infection rate in humans and animal hosts will then increase the probability of snail infection . A study by Rudge et al . ( 2008 ) in the Philippines using S . japonicum larval stages , found high levels of parasite gene flow between humans and dogs suggesting strongly the frequent transmission of S . japonicum infection across host species and between villages [23] . Furthermore , the role of animals in disease transmission was further supported by a population genetics study in China , where S . japonicum from cattle showed high genetic diversity in the marshland areas , whereas parasites from humans and dogs were more diverse in the hilly region [24 , 25] . These previous studies have therefore demonstrated the contribution of animal host species in the genetic diversity , and gene flow pattern of the parasite . Thus , the zoonotic nature of S . japonicum infecting animals should be seriously considered in the increased disease transmission [26 , 27 , 28] . Currently , we are now performing direct genotyping of stool-derived eggs collected from humans and animals particularly in water buffaloes and dogs using microsatellites . We will measure the infection intensities in humans and animals together with the parasite’s genetic variation in our ongoing study . Moreover , in this study we chose to use adult worms isolated from mice experimentally infected with snail-derived cercariae for genotyping . This is because adult worms can generally provide DNA with higher quality and quantity suitable for genotyping than snail-derived cercariae or eggs in stool samples . However , fitness of parasites to mice may serve as a bias to the genetically diverse population , leading to bottlenecking of genotypes [29 , 30] . The S . japonicum cercariae shed from infected snails collected from endemic areas will also be analyzed in our future studies . Because there are several studies on parasite’s population genetics by using cercariae-derived samples [23 , 24 , 25] , such the direct genotyping may be feasible and provide an advantage to skip in vivo passage . The level of genetic differentiation differs between endemic areas . Great genetic differentiations were observed in the New Corella samples than those from other endemic sites ( Table 2 ) . The large geographical distance separating New Corella from other endemic sites could possibly limit the contact between the hosts , eventually resulting to high genetic differentiation in this municipality . New Corella is located in Southern Mindanao and is expected to be more genetically differentiated because of its geographic location ( Fig 1 ) . As seen in Fig 1 , New Corella is the farthest of the endemic municipalities being separated by a wide distance from other endemic municipalities . Previous studies showed that the high genetic differentiation observed among peripheral populations such as those of New Corella can be explained by their strong spatial isolation [31] . Furthermore , the possibility of the snail hosts influencing the genetic variation of the parasite population should also be taken into consideration . Presently , there is no study using microsatellites on the genetic variation of the snail population in the sampling areas . However , previous studies on mitochondrial DNA had provided some insights into the genetic variation of the snail population of S . japonicum , and suggested that examination of naturally infected snails may exhibit co-evolutionary relationships with their parasites [32] . Thus , a snail population may reflect the population genetic parameters of their parasites [33 , 34] . Nevertheless , it is worth mentioning that these previous studies on the genetic diversity of Oncomelania populations are based on mitochondrial markers . Hence , future studies using microsatellites on the snail populations from each endemic area is essential to obtain results which can be analyzed together with those of S . japonicum population . The genetic variation observed using AMOVA was greater within each S . japonicum population ( 91 . 95% ) than the variation among the populations ( 8 . 05% ) ( Table 3 ) . This might be due to the snails being infected by genetically different cercariae having multiple genotypes within the endemic areas [7 , 22] . Mixing of infected snails and of their parasites brought about by flooding may explain the higher genetic variation within the population [5 , 9] . Also , the continuous rainfall and subsequent floods in these endemic areas might facilitate host-parasite contact , exposing people and animals to contaminated waters that result to higher infection . Thus , people and animals moving from one village to another to escape flooding , take advantage of employment opportunities and there is also animal trade where water buffaloes are exported to other areas could facilitate parasite transmission , contributing to high genetic variation within each endemic area . Another reason could be due to the snail sampling being done in three villages for each endemic municipality where a high village-level variation might exist . Genetic variance within population was accounted for most of the genetic diversity of S . japonicum population in endemic provinces in China [5] . S . japonicum samples obtained from different endemic areas did not form a particular spatial structuring . The lack of geographical structuring suggests that there is still an ongoing gene flow among the S . japonicum populations in all the study areas despite execution of control measures [22] . These findings might imply that there is a continuing transmission of S . japonicum across geographic areas , and therefore reflect the inadequate effect of MDA implementation . The current national control strategy for schistosomiasis in the Philippines is annual MDA using 40 mg/kg of praziquantel in all schistosomiasis-endemic villages including the sampling areas . However , the compliance rate was reported to be <50% [35 , 36] . The ongoing gene flow in the populations might be attributed to migration and movement of infected hosts as also suggested otherwise by previous studies done on S . mansoni [22] . The infected hosts could therefore serve as means of allele dispersal in endemic sites . Therefore , the existence of gene flow among the schistosome populations might increase the opportunity for the spread of alleles conferring parasite traits such as infectivity , virulence and drug resistance [8 , 22] . Two subgroups were observed in the Catarman samples using the PCoA analysis ( Fig 2 ) , indicating that co-infection with several genotypes of the parasite might be infecting the hosts in this endemic site . Catarman is surrounded by other endemic areas such as Leyte , Negros Occidental , Bohol in the Visayas , so the possibility of intermixing of the parasite is very high leading to high genetic variation within the area . A higher transmission and infection success is expected to occur more in mixed parasite genotypes than in single-genotype infection as reported in previous studies [37] . There might be a decrease in the effectiveness of the host immune system to cope with the infection due to the simultaneous attack of the parasite with different genotypes leading to a higher infection success [37] . Furthermore , the high genetic diversity in Catarman may be explained by the lowest inbreeding coefficient values . Inbreeding increases the similarity of the alleles in the parasite population . Previous studies suggested that co-infection by multiple genotypes decreases the possibility of inbreeding [37] . Some parasite populations in Gonzaga , New Corella and Irosin were identified as migrants using Gene Class 2 . 0 software ( Table 4 ) . Gonzaga has just been identified as a new endemic focus at the start of the 21st century [38] , and it is presumed that some parasite populations are introduced in this area from Talibon and Catarman . Infected people or animals might have moved from these areas and started the disease transmission in Gonzaga . The theory proposed for the emergence of schistosomiasis in Gonzaga was based on the history of a big geothermal project by PNOC ( Philippine National Oil Company ) in Gonzaga that recruited workers from the Visayas and Mindanao ( Fig 1 ) . The movement of people from these endemic areas into Gonzaga brought in cases , and with the presence of snail hosts in the area , the emergence of the disease became just a matter of time [38] . Interestingly , migrants detected in Irosin were rooted from Gonzaga , further providing evidence for the continuous transmission flow of the parasite among the endemic municipalities in the Philippines ( Table 4 ) . Migrant detection in the study supports the clustering analysis using the neighbor joining method wherein the population , including those individuals that were considered to be migrants , clustered together with their source population; for instance , Gonzaga received migrants from Talibon , which clustered together in the NJ tree ( S1 Fig ) . Genetic diversity found in this study among the parasite population in each endemic site in the Philippines is vital in the parasites’ ability to survive the effect of selective pressures such as those brought by drug treatment . At the same time , selection pressures increase the frequency of favorable alleles across all populations [4] . In this sense , the present finding of high diversity among the parasite populations imply that the MDA with praziquantel has varying degree of impact in interrupting the parasite’s life cycle . Aside from looking at several factors that can contribute to possible treatment failures including low compliance and the quality of the drugs used , it is also important that the effects of MDA be monitored on schistosome populations for its genetic background . This can be done by using microsatellite markers to measure the genetic diversity parameters which include the allelic richness and the heterozygosity of the alleles in the parasite population before and after MDA implementation . Alleles contributing to the severity of the disease such as the ones responsible for the fecundity and survival of S . japonicum inside the host should be identified and need to be further studied . In conclusion , the use of microsatellites in this study has shown that there is an ongoing gene flow among the S . japonicum population from different endemic areas , indicating the active movement of infected humans and animals from one endemic area to another . Aside from the control programs being implemented in each endemic area , an effective surveillance to monitoring these movements in humans and animals in each endemic site should be in place . Thus , a better cooperation between the medical and veterinary sectors would be highly recommended to ensure a strengthened control program for schistosomiasis . In addition , the diversity will indirectly explain the varying degree of the effects of the ongoing control programs done in these endemic areas . A regular MDA should be implemented and monitored regularly for its efficacy in endemic areas . Considering that only 10 microsatellite markers were analyzed in this study for determining genetic diversity and gene flow of the parasite , we therefore recommend the use of additional highly polymorphic microsatellite markers not only for S . japonicum , but for S . mansoni and S . haematobium to be used in future studies for more precise analysis .
|
Schistosomiasis is one of the important neglected tropical diseases endemic in 78 countries throughout the world . The disease is caused by the parasitic worms known as schistosomes . In the Philippines , S . japonicum is the causative agent of the disease . The prevalence of the disease varies in endemic areas , suggesting that the parasite populations might differ in their genetic composition and infectivity to the human host . In this study , DNA samples of adult worms from seven endemic municipalities were analyzed . Characterization of S . japonicum samples in different endemic sites with varying prevalence provides information on the genetic diversity of the parasite . Results of this study showed that samples in high prevalence endemic areas like Irosin , Catarman and Socorro were genetically diverse as compared to other areas . Information on parasite genetic diversity is therefore important in planning disease control strategies . The results suggest ongoing parasite transmission across geographic endemic areas which should be monitored and used as reference for genetic diversity of the schistosomes attributed to geographic areas , thus a safeguarding precaution should be implemented to ensure localized elimination of the disease .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"schistosoma",
"invertebrates",
"biogeography",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"helminths",
"population",
"genetics",
"parasitic",
"diseases",
"animals",
"microsatellite",
"loci",
"gastropods",
"gene",
"types",
"population",
"biology",
"snails",
"schistosoma",
"japonicum",
"geography",
"molluscs",
"phylogeography",
"pathogenesis",
"host-pathogen",
"interactions",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"genetics",
"of",
"disease",
"organisms"
] |
2017
|
Geographic strain differentiation of Schistosoma japonicum in the Philippines using microsatellite markers
|
Improving accuracy in genetic studies would greatly accelerate understanding the genetic basis of complex diseases . One approach to achieve such an improvement for risk variants identified by the genome wide association study ( GWAS ) approach is to incorporate previously known biology when screening variants across the genome . We developed a simple approach for improving the prioritization of candidate disease genes that incorporates a network diffusion of scores from known disease genes using a protein network and a novel integration with GWAS risk scores , and tested this approach on a large Alzheimer disease ( AD ) GWAS dataset . Using a statistical bootstrap approach , we cross-validated the method and for the first time showed that a network approach improves the expected replication rates in GWAS studies . Several novel AD genes were predicted including CR2 , SHARPIN , and PTPN2 . Our re-prioritized results are enriched for established known AD-associated biological pathways including inflammation , immune response , and metabolism , whereas standard non-prioritized results were not . Our findings support a strategy of considering network information when investigating genetic risk factors .
The discovery of disease-associated genomic variation has numerous clinical and scientific applications , including earlier disease prognosis , improved understanding of disease pathophysiology , and development of personalized treatment therapies [1] . A commonly used technique for identifying these mutations is the genome wide association study ( GWAS ) approach [2] . Typically , a large sample of affected and unaffected individuals are genotyped for many single nucleotide polymorphisms ( SNPs ) using a high-density microarray chip and then test statistically if the allele frequency of each variant is associated with disease status [2] . Significant associations in this first step ( “discovery phase” ) are deemed to be robust if they replicate in an independent cohort ( “replication phase” ) . In this study , we focused on improving the replicability of GWAS results for Alzheimer disease ( AD ) , although our methodology is applicable to genetic data for other diseases and traits . AD is a neurodegenerative disease resulting in irreversible dementia and memory loss with elevated prevalence in older populations [3] . Recent estimates suggest that approximately 5 . 4 million Americans have AD , and the number of cases of AD is expected to increase dramatically in future years if medical advances continue to improve life expectancy , thereby allowing more individuals to reach ages where AD is on the rise [3] . Genetic studies of AD have led to identifying numerous AD associated genes such as APP [4] , PSEN1 [5] , and PSEN2 [6] for early onset AD ( EOAD ) , as well as APOE [7 , 8] and SORL1 [8 , 9] for late onset AD ( LOAD ) . Common variants in more than 20 other genes have been robustly associated with AD risk [8] . However , not all AD associated genes will reach genome wide significance in current datasets of sample sizes below 100 , 000 individuals . It is well recognized that incorporating other forms of biological data improves confidence in genetic findings [10–12] . Our computational framework is based on the following biological hypothesis . If a known AD variant is associated with a gene that is involved in a particular biological process ( BP ) ( e . g . inflammation ) , we assume as a probabilistic prior that other AD variants might be associated with proteins involved in this BP or proteins that physically interact with this BP . This hypothesis can be tested computationally using a protein interaction network [13–15] by extending the “guilt by association” principle via propagation of probabilistic evidence in a network [16 , 17] . This general idea has similarity to the Google ranking algorithm of web pages , in which a web page that has a short link distance to many “important” pages will itself be considered “important” . In the case of protein interactions , guilt by association-based inference is typically performed by inspecting the function of direct neighbors of a predicted disease gene in a protein-interaction network . This approach has been incorporated in multiple interpretation systems as well as commercially such as Ingenuity Pathway Analysis ( IPA ) . However , it has been shown that network propagation , diffusion or other related methods that go beyond simple neighbor-based analysis can carry functional or disease associations further in the network with improved predictive accuracies [10 , 11] . This idea extends to predicting both gene function and disease phenotypes associated with genes [11 , 18–22] . We hypothesize that this general framework , and network diffusion in particular , can be extended to aid prioritization of AD genes . Although the underlying biology of AD may be far more diverse than a single function , there are several biological pathways that are aberrantly activated in AD brains , and not surprisingly , most of the genes identified by AD GWAS contribute to these pathways [23] . For example , a primary indicator of AD is the accumulation of amyloid beta plaques in the brain , resulting from mis-processing of APP protein [23] . We developed a novel re-prioritization approach that can be integrated easily into the current genetic analysis design ( Fig 1 ) . First , we curated the AD literature to produce a set of approximately 60 robust AD ( RAD ) genes that includes those that have been associated with AD at the genome-wide significance level or that contain variants shown to affect AD-related processes directly ( Table 1 ) . We then constructed a network of protein-protein interactions and applied network diffusion to score and rank genes based on their proximity to the RAD genes . Network diffusion allows modeling of indirect interactions , modules and protein complexes that are not modeled if only the direct interactions of proteins are considered . Next , we combined our genetic association results with the network diffusion scores to produce a newly re-prioritized ranking of genes . Finally , we validated our methodology using a novel approach involving bootstrap aggregation on one of the largest assembled genetic datasets of AD . Network-augmented genetic results have measurably improved replication rates in this validation approach . We also show that our main results and key predictions were essentially unchanged after restricting the RAD set to 19 genes which have had been functionally validated as well as replicated in independent datasets .
We assembled a PPI network using interactions pooled from multiple PPI databases ( ConsensusPathDB [13] , iRefIndex [14] , and Human Interactome Y2H [15] ) inspired by recent work [21] . Pooling interactions from these three databases resulted in a connected network that includes a large percentage of the genes in our GWAS dataset . We then determined if the RAD genes are proximal within this network . The first proximity measure tested was the average shortest path ( ASP ) distance [50] . The ASP distance between RAD genes , determined by cross-validation ( See Methods ) , is much smaller than would be expected by random chance ( Table 2 ) . One problem is that ASP distance between RAD genes and genes with many interactions ( the number of interactions a gene has corresponds to its “degree” and high degree genes are considered to be hubs ) tends to be small ( Table 3 ) . In this situation , all hub genes will be falsely predicted to be AD-related . Thus , we incorporated instead the Regularized Laplacian diffusion kernel [51] which penalizes paths going through hubs . The diffusion distance between RAD genes is smaller than would be expected by chance ( p = 0 . 00054 ) ( Table 2 ) . Simultaneously , the problematic hub genes in the network have discounted scores as demonstrated by the notable drop in ranking of the 10 genes with the highest number of overall interactions ( Table 3 ) . We next tested if genes with high diffusion scores replicate more frequently in order to demonstrate that diffusion scores are informative when used in conjunction with genetic data . Bootstrap aggregation [52] was applied to our genetic dataset to produce a large number of pairs of discovery and replication datasets ( See Methods ) . In each discovery + replication pair , we conducted a standard genetic workflow , beginning with a screen in the discovery dataset followed by validating top findings in the replication dataset . For each pair , a replication rate was calculated by determining the percentage of genes that surpass a given significance threshold also replicated . To test if network diffusion scores improved replication , we altered the standard discover + replication approach . We ranked genes by their network diffusion score and then iteratively dropped genes that had ranking diffusion scores below a given stringency threshold . At first we retained only genes in the 50th percentile of network scores , then gradually increased the threshold to only include genes in the 60th , 70th , 80th , and 90th percentiles . For each threshold , we computed the replication rate and compared to the baseline . As shown in Fig 2 , filtering based upon network score percentile noticeably increased replication rate . Genes with a–log ( p-value ) of > 6 replicated at a rate of approximately 16% in simulations ( farthest right purple point ) , while additional strict network filtering improved the replication rate to nearly 34% ( farthest right red point ) . Since filtering on network diffusion score improved replication rate , we next sought to integrate the network diffusion scores and genetic results into a single score . First , we converted the p-value of each gene from genetic analysis into a Z-score ( “GWAS Z-Scores” ) and then converted the network diffusion percentile of each gene into a Z-score ( “Network Z-scores” ) . Linear regression analysis showed that the Network and GWAS Z-scores are independent ( Fig 3A ) . Next , we assigned each gene a replication rate based upon how frequently the gene replicated in our bootstrapped validation datasets ( See Methods ) . We observed that replication rates were higher for genes with higher network Z-scores compared to genes with lower network Z-scores ( Fig 3B ) . To combine the Network and GWAS Z-scores , we developed an approach that uses a linear support vector machine ( SVM ) [53] to determine how heavily each type of score should be weighted in order to maximize replication rate ( See Methods ) . These weights were then used in conjunction with the meta-analysis method for combining summary results implemented in METAL [54] . The weights predicted by the SVM ( Fig 4 ) were 0 . 703 ( GWAS ) and 0 . 297 ( Network ) . As further confirmation , we conducted binomial ( logit family ) logistic regression using network and GWAS Z-scores as predictors and the replication class ( high/low ) as the outcome . Both network and GWAS score were significant , ( GWAS: coefficient = -0 . 659 , p <2 . 0×10−16 ) ( Network: coefficient = -0 . 229 , p = 0 . 0016 ) . The coefficients derived from logistic regression are very similar to the SVM-derived weights ( GWAS weight = 0 . 742 , Network weight = 0 . 258 ) . Next , we applied our combined approach genome-wide , excluding the RAD genes and genes containing significantly associated variants ( p <1 . 0x10-7 ) to focus on novel candidates . Among the genes with largest combined Z-scores ( Table 4 , S1 Table ) , several have important roles in inflammation . CR2 ( p = 5 . 95×10−7 ) is a receptor protein involved in immune response ( genecards . com [55] ) . SHARPIN ( p = 1 . 43×10−5 ) is a component of the LUBAC complex that plays a regulatory role in inflammation [55] . PTPN2 ( p = 3 . 21×10−5 ) is a phosphatase that also serves an important role in regulation of inflammation and glucose homeostasis [55] . The Bonferroni-corrected significance threshold when considering only genes in the 75th percentile of network scores is p = 1 . 46 x 10−5 , although this is likely to be overly strict since proximally located genes are not inherited independently . We performed pathway analysis using Gene Set Enrichment Analysis ( GSEA ) [56] to determine if AD-related pathways are more enriched when genes are ranked by their combined Z-scores versus GWAS-only Z-scores ( See Methods ) . Notably , ranking genes based upon combined Z-scores resulted in several significantly enriched AD-related pathways including immune response , FOX03 targeting ( indicates enrichment for aging ) , and hippocampal development ( Table 5 ) . By comparison , ranking genes based only upon their GWAS Z-scores resulted in virtually no significant pathways entirely ( Table 6 ) .
GWAS of AD and AD-related endophenotypes have discovered and replicated associations with more than 60 genes ( Table 1 ) , many of which have roles in AD-related pathways ( amyloid β aggregation , inflammation , cholesterol transport , immune response , etc . ) . To identify additional AD-related genes , we hypothesized that genes having suggestive evidence for association from a genome-wide screen and protein-level interactions ( both direct and indirect ) are more likely to replicate . This idea has been referred to as functional linkage [57] . To test this hypothesis , we developed a novel approach for improving the prioritization of candidate disease genes that incorporates a network diffusion of scores from known disease genes using a protein network and integration with GWAS risk scores . We tested this approach on a large AD GWAS dataset and validated the performance of the methodology using bootstrap aggregation . Several novel AD genes were predicted including CR2 , SHARPIN , and PTPN2 . Part of the motivation for our approach was to identify genes that are more obviously biologically relevant to AD . This is exemplified by SHARPIN , whose principal known function is to form the LUBAC complex and prevent inflammation , a major process through which amyloid aggregation and AD are thought to develop [23] . Similarly , CR2 , a homolog of CR1 which is a well-established AD gene [8] , is involved in immune response . Many immune response genes are differentially expressed between healthy and AD brains , and investigations into the connection between expression in cell types and the presence of AD has led to growing interest in the role microglial cells ( a first responder in the immune response pathway ) [58] . Finally , PTPN2 is involved in multiple AD-related pathways; it has roles in negatively regulating inflammation and de-phosphorylation of key glucose metabolism kinases including INSR and EGFR [59] . The AD-related roles of each of our novel AD gene predictions , in combination with their strong network and genetic scores , make them highly promising candidates . One biological form of functional linkage that does not require direct physical interaction is membership in the same signaling pathway or protein complex . For example , our study identified interaction between FOXO and INSR that is consistent with evidence of a multi-link signaling pathway comprised of direct physical interactions in the insulin-signaling pathway [60] . By comparison , neighborhood enrichment approaches ( i . e . , testing a gene’s direct interactions ) cannot detect indirect interactions . Furthermore , neighborhood enrichment approaches are unreasonable for AD because some RAD genes are network hubs ( e . g . , APP has more than 2000 interactions ) which would result in an unreasonably high number of genes having AD-enriched neighborhoods . Some distance metrics capture indirect interactions by calculating the proximity between a pair of genes based upon short paths between them in the network . However , after testing a simple distance metric known as average shortest path ( ASP ) , we observed that hub genes were still the top-ranked predicted genes . Since hub genes have many interactions , they tend to have short overall paths to any genes in a network , although their functions are highly generic and unlikely tied to a particular disease . Ubiquitin C ( UBC ) , for example , has nearly 9 , 000 interactions; however , this is simply because protein degradation is essential for regulating the vast majority of proteins . Therefore , a more nuanced network propagation approach can aid in making disease specific inferences . Network diffusion is a widely used class of spectral graph clustering methods that have been applied to many computational disciplines [51] . We used this approach to propagate evidence in the form of AD scores throughout the network . A protein in the network that has a short “diffusion distance” to one or more well-established AD genes will receive a high network risk score . Notably , we observed that network diffusion down-weights hubs while simultaneously outperforming ASP distance when applying leave-one-out cross-validation to the RAD genes . Many diffusion kernels have been proposed in graph theory , however the Regularized Laplacian [51] approach used in this study has the highly desirable properties of requiring very little parameterization ( in fact , only a single parameter is required to be set ) and also more computationally efficient than other diffusion kernels . Network diffusion methods have been applied in other genetics research contexts such as labeling somatic network mutations in cancer [61] , characterizing gene sets [62] , and predicting risk genes for amyotrophic lateral sclerosis [21] . We also observed that genes with high diffusion scores tended to replicate more frequently in our 125 pairs of bootstrapped discovery and replication datasets . However , network Z-scores and GWAS Z-scores in the full dataset were not strongly correlated . Taken together , these observations indicate the importance of considering jointly protein interaction data and genetic results even though they are independent because the integration of both types of information will likely yield noticeable improvement in replicability of findings . Since our bootstrapping procedure required splitting the original dataset , the simulations were conducted using datasets that contained only one-half of the total sample . This suggests that our network scores aided in determining which genetic associations were real in datasets with reduced power . We note that our bootstrapping approach was performed on the same data from which we derived the GWAS Z-scores used to train the SVM . Therefore , the selection of combination weights may have been biased in favor of GWAS Z-scores . Furthermore , it is unclear whether the weight combination used in this study ( 0 . 297/0 . 703 ) would be appropriate for combining genetic and network data for other disorders or traits . The GWAS approach has a very limited capability to identify the entire set of genes which contribute to the risk of a complex disease like AD , even in datasets containing up to 100 , 000 individuals , because some genes do not contain variants that are sufficiently frequent and/or exert a large enough effect to yield a statistically significant association . To overcome this limitation , we developed a novel SVM approach to integrate the genetic and network scores by propagating GWAS Z-scores in a PPI network . In the AD example presented here , we initialized the RAD genes to have an identical high score in the network , thereby allowing re-prioritization of genes in any AD dataset regardless of the internal Z-scores of the RAD genes . We acknowledge that our initial choice to treat each RAD gene equally may be controversial . Arguably , we could have seeded our analyses with GWAS Z-scores for each RAD gene from the original studies . However , our approach permits unbiased exploration of interactions of all plausible AD genes and does not require adjustment to these Z-scores for sample size or allele frequencies . Moreover , results derived from weighted RAD genes would be dominated by interactions with APOE for which the significance level exceeded a–log ( p-value ) of more than 100 in several datasets ( compared to < 10 for most other RAD genes in the total group of datasets ) . Also , several key AD-related genes ( e . g . , APP , PSEN1 and PSEN2 ) which show little evidence for association with individual SNP or gene-based tests for AD would be undervalued in analyses using weighted Z-scores . In order to make our software maximally flexible and support weights derived from confidence in the seed genes , we implemented an option for users to specify unequal weights on the seed genes at their own discretion . A potential concern about our results is the strategy for selecting RAD genes because many significant GWAS findings include variants located in intergenic regions . The most parsimonious explanation is that the variant responsible for the association peak influences the nearest gene , but there is abundant evidence suggesting this assumption is often incorrect . To address this issue , we repeated our analyses using a more restricted set of RAD genes that included only those supported by genome-wide significant evidence of association with AD risk and replication in independent datasets or by other genetic evidence plus experiments linking them to AD-related pathophysiology . Our leave-one-out cross validation approach demonstrated that the genes in the restricted RAD set had closer network proximity to each other than would be expected by chance ( p = 5 . 93x10-5 , S2 Table ) . The statistical support for the novel genes CR2 ( p = 4 . 09x10-7 ) , SHARPIN ( p = 1 . 10x10-5 ) , and PTPN2 ( p = 2 . 41x10-5 ) remained the same ( S3 Table ) . Finally , combined Z-scores that were derived using diffusion from the more conservative RAD gene set yielded similar AD-related pathways such as Fx03 targets ( FWER p = 0 . 064 ) , antigen processing ( FWER p = 0 . 02 ) , and hippocampal development ( FWER p = 0 . 065 ) ( S4 Table ) . These results confirm that the genes with a clear functional role in AD produce network diffusion-based predictions that are consistent with the results presented here . Curiously , the inclusion or exclusion of the portion of RAD genes that have an ambiguous or non- validated functional role in AD did not affect our results . We also acknowledge that several of the novel putative AD genes may have been erroneously prioritized because they are in the same locus with RAD genes . This concern is unlikely noting that there are several instances where a genetic association peak includes multiple genes that may have a possible functional role in AD ( e . g . , the MS4A gene cluster [8] ) . Although one of our novel AD genes , CR2 , is located close to CR1 , which is an unambiguous RAD gene given its robust replication in GWAS and effect on deposition of neuritic amyloid plaque [63] , CR2 is also an intriguing AD candidate gene because it has been shown to regulate hippocampal neurogenesis [63] . Thus , our findings suggest that our approach will aid in predicting truly multiple AD-related genes at a locus , however additional biological evidence may be required in some instances to make this distinction . Previous AD studies have implicated inflammation and immune response genes , but we did not observe enrichment for these pathways when incorporating only GWAS scores in the analysis . However , these and other recognized AD-related pathways emerged after applying our network re-prioritization method ( Table 6 ) suggesting that incorporation of network data can help minimize discrepancies in predictions across different genetic datasets . On the other hand , other well-established AD-related pathways , including cholesterol metabolism and endocytosis , were not detected by our approach . Further inspection of the results revealed , for example , that enrichment for the cholesterol homeostasis pathway is not significant when applying GSEA to the genetic data only ( FWER p = 1 ) . This pathway as defined in the Molecular Signatures Database ( MSigDB ) is very broad and contains many genes that are weakly associated with AD which consequently diminish the enrichment of the set . The evidence for this pathway is greater in the analysis using only network scores ( FWER p = 0 . 18 ) , which indicates our method still improves the detection of cholesterol homeostasis . Even pathways such as HDL-mediated lipid transport that were enriched in analyses considering only genetic data ( largely due to the strong signal from APOE ) were not ranked highly by our network diffusion algorithm because RAD genes such as APOE are ignored to minimize bias . Although merging of multiple databases to obtain a very highly connected network is a requirement for the diffusion algorithm to work properly , our approach offers several advantages in comparison to other network-based approaches including biological transparency , ease of integration with a variety of GWAS methods , and the ability to balance data-driven statistics and biological prior probabilities . The extensive simulations we conducted provide a general basis for further establishing the practicality of genetic and network-based integration . Our network methodology was developed with the goal of accommodating known complications of genetic analysis . The software developed for this study is open source , accessible to most users ( incorporated in an R package ) , and applicable to any set of variant- or gene-level disease association results . Importantly , it requires only a set of GWAS results and a list of previously known disease genes and , therefore , does not necessitate changes to previously established genetic analysis pipelines . Although we used an SVM procedure to determine the weights for the score combination , a user can specify any weights or simply use our defaults that are based on the 0 . 297/0 . 703 ratio determined by SVM . Our package is accessible through GitHub ( https://github . com/lancour/ignition ) .
A set of genes ascribed to AD with a high degree of certainty was assembled through curation of published findings ascertained through PubMed searches that emerged from studies using a variety of approaches including GWAS of AD risk and AD-related endophenotypes , family-based linkage analysis , positional cloning , whole exome sequencing ( WES ) , and candidate gene testing ( CGS ) ( Table 1 ) . Criteria for inclusion in this set included ( 1 ) genome-wide significance for GWAS and WES studies ( p < 5x10-8 ) and LOD score > 3 for linkage studies and ( 2 ) replication of association signals in independent datasets; or ( 3 ) biological evidence that demonstrate functional relevance to AD of associated variants or the encoded protein . A set of interacting gene-gene pairs ( in HGNC symbol format ) is required as input for this software . To compile this set , three databases ( RefIndex v14 [14] , ConsensusPathDB v31 [13] , and Human Interactome Y2H DB vHI-II-14 [15] ) were selected based on their demonstrated utility in recent work [21] . iREFINDEX and ConsensusPathDB interactions were filtered to remove self and complex ( more than two proteins ) interactions . The ConsensusPathDB interactions are given in uniProt ID format , which were converted to HGNC symbols using the official website ( http://www . genenames . org ) . iREFINDEX provides a HGNC symbol for each interactor of an interaction when possible , and so only interactions which had a HGNC for both interactors were kept . The Human Interactome DB already provides a set of binary gene-gene interactions in HGNC format , so no processing was required . The union of the processed sets from each database was used as the final interaction set . The unified set contains 19 , 972 unique gene symbols and 236 , 642 interactions . These databases are curated collections of experimentally determined interactions ( typically binding or affinity ) reported in the literature , such as from co-immunoprecipitation , as well as predicted interactions in a small number of databases . Network diffusion is a very well-studied spectral approach to graph clustering and annotation [17 , 51 , 64 , 65] . It attempts to mimic node-to-node distance in the graph that in turn aims to capture functional relevance . The first step of the diffusion method is to model the protein interactions as a network . A network is comprised of a set of nodes , V , and a set of edges between nodes , E . For this work , nodes represent genes , and edges represent an interaction present in the unified set . Although we use unweighted edges in this work , our network methods and software are able to receive weighted input as well , such as protein interactions with confidence measures taken from STRING [66] . The construction of diffusion kernels using weighted edges has been well studied and is equally valid [51] . n is the number of nodes in the network , which is 19 , 972 ( yielding 236 , 642 edges ) . All network methods were implemented in R . The regularized Laplacian kernel [51] is constructed by: K= ( I+αL ) −1 ( 1 ) where K is the resulting kernel , I is the identity matrix , L is the graph Laplacian , and alpha is a constant ( see S1 Text and [51] for additional details ) . For this study , an alpha value of 0 . 1 was used , consistent with other work in this field [17] . Next , a network diffusion score was computed for each gene . To do this , the diffusion score vector , y , was initialized to be a length n vector that contains 1’s in the indices of the RAD genes , and 0’s otherwise . Risk scores for all genes in the graph were then derived by multiplication of K by the diffusion score vector y: ỹ = Ky . To test if RAD genes had closer than random diffusion proximity to other RAD genes in a network , leave-one-out cross validation [67] was applied to the RAD gene set . First , a single RAD gene from the RAD set was set to 0 in the initial diffusion score vector , y . Then , diffusion scores were computed based upon this new initialization of y . The diffusion scores were sorted and the sorted rank of the removed RAD gene’s diffusion score was determined in comparison to all other non-RAD genes . This process was repeated for each gene in the RAD set , resulting in a list of ranks . If diffusion proximity is informative and potentially predictive , the average rank of the RAD genes should be significantly lower than the average rank of all genes , ( n+1 ) / 2 , which was verified using a one-tailed t-test . The Alzheimer’s Disease Genetics Consortium ( ADGC ) is an NIA-funded project whose goal is to identify genes associated with an increased risk of developing late-onset Alzheimer disease ( LOAD ) by assembling and analyzing genetic and phenotypic data from large cohorts containing rigorously evaluated AD cases and cognitively normal controls of various ethnic ancestries . Details of ascertainment , collection , quality control ( QC ) , and analysis of genotype and phenotype data in the individual datasets of the ADGC are provided elsewhere [8 , 68] . Here we examined genotype data that were generated using high-density SNP microarrays from 32 prospective , case-control , and family-based studies of LOAD comprising 16 , 175 case and 17 , 176 controls of European ancestry . After QC steps to filter low-quality SNPs and individuals with low genotype call rates , principal components ( PCs ) of ancestry were computed within each dataset using EIGENSTRAT [69] and a set of 21 , 109 SNPs common to all genotyping platforms and datasets in order to account for population substructure in genetic association analysis . Samples with outlier PC values >six standard deviations from the mean were excluded from subsequent analyses . Genotypes for a much larger set of SNPs were imputed using the Haplotype Reference Consortium panel release 1 . 1 [70 , 71] , which includes 64 , 976 haplotypes derived from 39 , 235 , 157 SNPs , and the Michigan Imputation Server ( https://imputationserver . sph . umich . edu/ ) running MiniMac3 [72 , 73] . Association of AD with the imputed dosage of the minor allele for each SNP ( a quantitative estimate between 0 and 2 ) genome-wide was conducted using logistic regression models implemented in PLINK [74] that included covariates for age-at-onset/age-at-exam , sex , the first three PCs , and an indicator variable for each dataset . Joint analysis was chosen in favor of meta-analysis to avoid problems that could be introduced if bootstrap aggregation under-sampled small cohorts , resulting in unreliable association estimates for those cohorts . To account for relatedness in family datasets , subsets of maximally-unrelated affected and unaffected individuals were sampled from each pedigree . Each variant was annotated to a gene region according to RefSeq release 69 [75] using the program ANNOVAR [76] . Then , each gene was assigned the minimum p-value of all variants annotated to it , after applying the following formula: PgGene′=1− ( 1−PgBestSNP ) N+12 ( 2 ) where N is the number of variants analyzed that were annotated to the gene . Previously , this correction [77] has been shown to perform comparably to more complex adjustments based upon gene length , recombination hotspots , and similar gene features [78] . Since the availability of large AD genetic datasets is limited , bootstrap aggregation [52] was used to generate a high number of datasets for method validation . First , the full ADGC dataset was equally separated into discovery and replication halves . Then , 25 iterations of bootstrap aggregation were applied to the discovery half and then the replication half . The resultant 25 discovery and 25 replication datasets were then matched ( D1 and R1 , D2 and R2… . D25 and R25 ) . To further ensure robustness , the splitting procedure was repeated a total of 5 times , with 25 iterations of bootstrap aggregation applied each time , resulting in 125 total pairings ( D1 and R1 , D2 and R2 . …D125 and R125 ) . Each pairing represents a discovery dataset as well as an independent replication dataset . For each pairing , the previously described genetic analysis was conducted on the discovery half . Then all genes that passed a designated significance threshold ( the number of passing genes is denoted as r ) were selected to be tested again in the replication half using a significance threshold of ( 0 . 05 / r ) . The replication rate was computed by determining the percentage of passing genes in the discovery half that also passed in the replication half . A replication rate was estimated for each pairing , and the mean replication rate was then determined . Next , the replication rate was re-determined for each pairing , with the added criterion that selected genes must also have a top percentile network diffusion score ( top 10th , 20th , 30th , 40th , and 50th were tested ) . The average replication rate for each filtering threshold was compared to the average replication rate without filtering . The p-values from genetic analysis of the ADGC dataset were converted to Z-scores using the qnorm function in R . Then , the network diffusion scores were converted into percentiles . The percentiles are transformed into Z-scores using the qnorm function , with the additional specification of lower . tail = F . The weighting scheme from METAL was applied to combine the GWAS and network Z-scores: Zcombined=w1*Zgwas+w2*Znetworkw12+w22 ( 3 ) Although any weight selection can be used , the weights were “learned” using an SVM [53] due to the observation that the GWAS and network scores did not contribute equally to predicting replication rate . First , a replication rate was determined for each gene . If a gene had a p-value of <0 . 05 in d discovery datasets and a replication p-value of <0 . 05 in r of the paired replication datasets , it was assigned a replication rate of r/d . To reduce model overfitting , create sufficient separation between the classes , and achieve a balance of high and low replicating genes , only high replication genes ( ≥0 . 7 , n = 676 ) and low replication genes ( <0 . 1 , n = 475 ) representing approximately 8 . 4% of the total genes with both a network and GWAS scores were extracted . By comparison , using a threshold of 0 . 8 or 0 . 9 would result in an imbalanced training set with very few high replication genes because highly replicating genes are uncommon . A linear SVM [53] was trained using the network Z-scores and the genetic association Z-scores as features , and “high” and “low” as the classes . The resulting slope of decision boundary was then used to determine appropriate weights ( w1 = 0 . 703 , w2 = 0 . 297 ) . Pathway enrichment was performed using the Gene Set Enrichment Analysis ( GSEA ) software [56] . GSEA’s pre-ranked analysis tool requires that the user provide a numeric measure for ordering genes . To establish a baseline , enrichment was done using our internal GWAS Z-scores to order genes . Then , enrichment was done using the alternative ordering genes based upon their combined Z-scores ( see above for combination method ) . The gene sets tested for enrichment were the GSEA C2 pathways in MSigDb , which are the “curated gene sets” compiled from multiple sources including KEGG [60] , Reactome [79] , and domain experts . The significance threshold was set at FDR < 0 . 25 , as suggested previously for this hypothesis generating approach [56] . The use of de-identified human subject information for this study was approved by the Boston University Institutional Review Board .
|
Integrating multiple types of -omics data is a rapidly growing research area due in part to the increasing amount of diverse and publicly accessible data . In this study , we demonstrated that integration of genetic association and protein interaction data using a network diffusion approach measurably improves reproducibility of top candidate genes . Application of this approach to Alzheimer disease ( AD ) using a large dataset assembled by the Alzheimer’s Disease Genetics Consortium identified several novel candidate AD genes that are supported by pre-existing knowledge of AD pathobiology . Our findings support a strategy of considering network information when investigating genetic risk factors . Finally , we developed a transparent and easy-to-use R package that can facilitate the extension of our methodology to other phenotypes for which genetic data are available .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"genetic",
"networks",
"protein",
"interactions",
"pathology",
"and",
"laboratory",
"medicine",
"neurodegenerative",
"diseases",
"protein",
"interaction",
"networks",
"immunology",
"signs",
"and",
"symptoms",
"network",
"analysis",
"genome",
"analysis",
"alzheimer's",
"disease",
"computer",
"and",
"information",
"sciences",
"inflammation",
"proteins",
"proteomics",
"dementia",
"mental",
"health",
"and",
"psychiatry",
"immune",
"response",
"biochemistry",
"diagnostic",
"medicine",
"gene",
"identification",
"and",
"analysis",
"neurology",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"gene",
"prediction",
"genetics",
"of",
"disease",
"computational",
"biology",
"human",
"genetics"
] |
2018
|
One for all and all for One: Improving replication of genetic studies through network diffusion
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.